Primers in Medicine

Transfusion of Packed Red Blood Cells The

Indications Have Changed
B y A l a n C o o k , M D ; a n d N a t e M i l l e r, M D

Abstract
Whole blood/packed red blood cells (pRBC) units transfused in the U.S. totaled 13,785,000 in 2011.1 A single
institution in South Dakota transfused 6,485 units of pRBC in 2013. Current thresholds for transfusion have
changed and each transfusion has the risk of causing an adverse reaction; thus, it is important to ensure pRBCs
are administered appropriately. Due to these changes and the potential risks associated with transfusion, we
reviewed the literature regarding appropriate indications for transfusion of pRBC. Our review specifically
focused on four disease entities: iron-deficiency anemia, acute upper gastrointestinal (GI) bleeding, acute coronary syndromes, and chronic ischemic heart disease. Based on our findings, we recommend utilizing an overall
conservative approach to the transfusion of pRBC. In patients with iron-deficiency anemia, first try alternative
methods to improve hemoglobin levels; in those with acute GI bleeding, transfuse for hemoglobin less than 7
g/dL; in patients with acute coronary syndromes, let symptoms/signs be your guide; and in patients with ischemic
heart disease, transfuse for hemoglobin levels less than 8 g/dL or if they are symptomatic. Most importantly, be
cautious to not fixate on numbers alone; always incorporate patients symptoms and co-morbidities when considering whether to transfuse pRBCs.

Introduction
Current thresholds for the transfusion of packed red blood
cells (pRBC) have recently been challenged. These
changes/recommendations will impact physicians clinical
practice both nationwide, and in South Dakota. Whole
blood/packed red blood cells (pRBC) units transfused in
the U.S. totaled 13,785,000 in 2011.1 One institution in
South Dakota transfused 6,485 units of pRBC in 2013
alone.
Clinical guidelines on appropriate transfusion practices
were published in 2012.2 In addition, new data provides
additional guidance for the transfusion of pRBC in
specific clinical populations.3-5 Given the recent changes
proposed and their potential impact on clinical practice,
we performed a literature review to specifically address
appropriate indications for the transfusion of pRBC in
four distinct (but not necessarily mutually exclusive)
clinical populations: iron deficiency anemia, acute upper
gastrointestinal (GI) bleeding, acute coronary syndrome,
and chronic ischemic heart disease. In addition, we will
542

briefly discuss possible adverse reactions one may

encounter with transfusion of pRBC.
Iron Deficiency Anemia
Iron deficiency anemia (IDA) is an entity we often
encounter in clinical practice; depending on the cohort
examined, it can affect anywhere from 2 to 12 percent of
the U.S. population.6 The classic findings in IDA are a
microcytic, hypochromic anemia (reduced mean corpuscular volume [MCV]) and mean corpuscular hemoglobin
concentration [MCHC], respectively) associated with
reduced serum ferritin, increased transferrin, reduced
serum iron, reduced transferrin saturation, increased red
blood cell distribution width (RDW), and a reduced
reticulocyte count.6-8
IDA is the most common cause of microcytic anemia; the
differential includes anemia of chronic disease (ACD),
thalassemia, and sideroblastic anemia. IDA may also
present as a normocytic anemia. In this case, it must be
distinguished from nutritional deficiencies (e.g., folate,

Primers in Medicine
vitamin B12), anemia of renal insufficiency, hemolytic
anemia, primary bone marrow disorders, and ACD.6-8

common at an estimated incidence of 20 per 100,000

persons in the U.S.15-16

Serum ferritin is the most useful test in distinguishing IDA

from other causes of anemia, and should be ordered early
when IDA is suspected. One study9 found a serum ferritin
less than or equal to 12 mcg/L had a sensitivity of 25
percent and a specificity of 98 percent in the diagnosis of
IDA. When the cutoff value was raised to less than or
equal to 30 mcg/L, the sensitivity and specificity were 92
and 98 percent, respectively, giving a positive predictive
value of 92 percent. Therefore, a low ferritin is near
diagnostic of depleted iron stores.

Acute gastrointestinal bleeding can result in massive

blood loss and early assessment of bleeding is crucial.
Although a rapid decrease in hemoglobin should raise
concern, it is not sensitive in the acute setting as a
variable amount of time is required for the extra vascular
fluid to equilibrate with the intravascular space.
Therefore, it is important to monitor for signs of hemodynamic instability. It has been suggested that blood loss less
than 200 mL does not affect blood pressure or heart rate;
whereas losses in excess of 800 mL will result in an expected
decrease in blood pressure of 10 mmHg and an increase in
heart rate of 10 beats per minute.17 Additional signs and
symptoms which may be precipitated by blood loss include
orthostatic hypotension, shortness of breath, fatigue,
pallor, palpitations, and chest pain.17 A study of 123
patients with lower GI bleeding identified the following as
independent correlates of severe bleeding: heart rate
greater than 100 beats per minute, systolic blood pressure
less than 115 mmHg, syncope, non-tender abdominal
exam, bleeding per rectum during the first four hours of
evaluation, aspirin use, and more than two active comorbid conditions.18 Digital rectal examination should
also be included in the initial assessment to exclude
lesions of the anus and distal rectum.17

As an acute phase reactant, ferritin is elevated in

inflammatory states; thus, when serum ferritin is normal,
distinguishing IDA from ACD can be difficult. This
problem is further exacerbated by the fact that serum iron,
serum transferrin, and transferrin saturation are non-diagnostic in distinguishing these two conditions (which in
fact may exist simultaneously). Measurement of serum
transferrin receptor (sTfR) in conjunction with other iron
indices or C-reactive protein (CRP) has been proposed to
help in confirming the diagnosis. However, sTfR comes at
increased cost and CRP has low specificity. ACD tends to
produce a less profound anemia, and when anemia is
severe enough to consider transfusion, IDA may be more
likely. Even the accuracy of bone marrow biopsy,
considered the gold standard for assessment of iron stores,
has become controversial.10 This highlights the necessity
of utilizing not only laboratory tests, but also patients
symptoms, past medical history and signs when making
the diagnosis of IDA.
Once diagnosed, iron deficiency anemia is best treated
with oral or intravenous iron; blood transfusions are
generally only recommended in specific clinical situations
(acute coronary syndromes, shock, etc).11 After initiation
of iron therapy, a reticulocytosis should be noted in three
to four days and hemoglobin should rise 1 g/dL per week.12
A discussion of when to use oral iron as opposed to
intravenous iron is beyond the scope of this article.
Acute Upper Gastrointestinal Bleeding
Based on data from the National Hospital Discharge
Summary, the average annual incidence of hospitalizations
due to gastrointestinal bleeding has been estimated at
approximately 350 per 100,000 persons in the U.S.13 The
incidence of acute upper gastrointestinal bleeding in the
U.S. is 95 per 100,000 persons and carries a mortality of 5
percent.14-15 Acute lower gastrointestinal bleeding is less

Classically, lower GI bleeding is associated with hematochezia, and upper GI bleeding is associated with
hematemesis and melena. Melena can rarely occur due to
lesions of the right colon, and large volumes from an upper
GI source can produce hematochezia, especially in cases of
increased GI motility. A small study of patients suspected
of lower GI bleeding based on the presence of
hematochezia, hemodynamic instability and absence of
hematemesis, melena and bloody nasogastric aspirate were
found to have an upper source of bleeding 15 percent of
the time. A similar association with hematochezia and
upper GI bleeding has also been found.20 Furthermore,
patients and physicians may have variable accuracy in
describing blood color and its categorization as either
melena or hematochezia.21 Laboratory testing to consider
in GI bleeding includes CBC and serial hemoglobin,
coagulation indices, blood type and crossmatch.17 An
elevated blood urea nitrogen to creatinine ratio suggests
an upper source of bleeding.22-24
Studies of critically ill patients have revealed a restrictive
transfusion strategy (transfusion indicated when
December 2015

543

Primers in Medicine
hemoglobin dropped below 7 g/dL) may be of benefit.24 In
addition, Hebert et al. found that cardiac events
(myocardial infarction and pulmonary edema) were more
frequent when a liberal as opposed to a restrictive transfusion strategy was employed.24 However, in these studies,
individuals with diagnoses of active blood loss were
excluded. Thus, Villanueva et al. conducted a study that
evaluated a restrictive versus liberal transfusion strategy in
individuals with acute upper gastrointestinal bleeding.5

safety measures of in-hospital death, recurrent MI, or new

or worsening congestive heart failure were met more often
in the liberal transfusion group as opposed to the restrictive
transfusion group (38 percent versus 13 percent,
p <0.046). However, this data was driven by the metric
new or worsening congestive heart failure; in addition this
study was conducted as a pilot study to inform if further
study is warranted, thus these findings should be interpreted
with caution.

The Transfusion Strategies for Acute Upper

Gastrointestinal Bleeding was a single center, randomized
clinical trial conducted from 2003 to 2009.5 The study
assigned individuals with acute upper GI bleeds to receive
one of two transfusion strategies. The restrictive strategy
group received blood transfusions only if the hemoglobin
level was less than 7 grams per deciliter. The liberal
strategy group received blood transfusions only if the
hemoglobin level was less than 9 grams per deciliter.
When the decision to transfuse was made, one unit was
administered with hemoglobin checked following
transfusion of each unit. Physicians were allowed to
deviate from the protocol if symptoms of anemia
developed. Significant exclusion criteria included patients
with massive exsanguination, acute coronary syndromes,
and lower gastrointestinal bleeds.

The MINT trial found conflicting results in comparison to

the CRIT trial.26 This study showed individuals assigned to
a liberal transfusion strategy (hgb less than 10) versus
restrictive transfusion strategy (hgb less than 8) demonstrated lower mortality (1.8 versus 13 percent, p<0.032).
However, the liberal transfusion groups mean age was
67.3 years as opposed to the conservative transfusion
groups mean age of 74.3 years. In addition, this too was a
pilot study conducted to determine if further investigation
is warranted, thus it was not appropriately powered to
make definitive conclusions on mortality.

Once exclusionary criteria were applied, 444 individuals

were randomized to the restrictive-strategy group and 445
the liberal strategy group. Mortality at 45 days was
significantly lower in the restrictive group vs. the liberal
group (5 vs. 9 percent, p=0.02). In addition, the restrictive
strategy group had a statistically significant decrease in the
rate of further bleeding (10 vs. 16 percent, p =0.01) when
compared to the liberal-strategy group. Nine hundred
sixty-seven more units were transfused in the liberal group
as opposed the restrictive-strategy group. This too was
statistically significant.5
Acute Coronary Syndromes
Data are conflicting if an aggressive versus restrictive
transfusion strategy is warranted in patients with acute
myocardial infarctions.25,26 Observational data currently
bring into question if the benefits of a more aggressive
transfusion strategy in patients with acute myocardial
infarction outweigh the risks. The CRIT randomized study
evaluated a restrictive (hematocrit less than 24 percent)
versus liberal (hematocrit less than 30 percent) transfusion strategy in patients with acute myocardial infarction
(40 percent STEMI and 60 percent NSTEMI).25 The
544

Chronic Ischemic Heart Disease

There currently are few high-quality studies that advocate
for or against the aggressive use of blood transfusions in
patients with chronic ischemic heart disease.
Recommendations are instead made based on observational
studies and sub-group analysis of randomized studies. A
multicenter, cohort study published in 2009 showed
critically ill patients with a cardiovascular diagnosis who
experienced moderate anemia were at increased risk of
death in comparison to non-anemic patients.27 However, a
retrospective cohort study of patients in the CathPCI
registry showed patients who received transfusion were
more likely to experience in hospital myocardial
infarction, stroke, congestive heart failure, cardiogenic
shock or death.28 Furthermore, a sub-group analysis of one
multi-center, randomized clinical trial showed there was
no difference in mortality in patients with chronic
ischemic heart disease who received a liberal (transfused if
hgb was less than 10 g/dL) versus a restrictive (transfused
if hgb was less than 9 g/dL) transfusion policy.24
Adverse Reactions
A discussion on appropriate indications for transfusion of
packed red blood cells would be incomplete if the risks of
transfusion were not discussed. Physicians are instructed
to do no harm. Every intervention has risk and benefits,
both known and unknown. The decision to transfuse
pRBC is no different. In 2011, 50,570 transfusion

Primers in Medicine
reactions occurred (includes pRBC and cumulative blood
components transfused); the reported incidence is 2.4
reactions per 1,000 units transfused. The two most
common adverse reactions include febrile, non-hemolytic
transfusion reactions, and mild-moderate allergic
transfusion reactions.1 Potential life-threatening transfusion reactions include transfusion related acute lung injury
(TRALI) and acute hemolytic transfusion reactions.
Febrile, non-hemolytic transfusion reactions (FNHTR)
and mild-moderate allergic transfusion reactions are
non-life threatening. A retrospective analysis showed
FNHTR occurred in 0.14 percent of all pRBC
transfusions.29 Symptoms generally occur one to six hours
post transfusion: these symptoms include fever, shaking
chills, and at times, mild dyspnea. They generally resolve
on their own, however acetaminophen can be utilized to
treat patients symptoms. Mild to moderate allergic
reactions manifest as urticarial eruptions; these can be
treated with diphenhydramine.
While the incidence of acute hemolytic transfusion
reactions are rare (less than 0.1 percent), they are the
most common life-threatening reactions.30 These
reactions occur in cases of blood incompatibility;
pre-formed antibodies attack foreign antigens, resulting in
brisk hemolysis. Initial symptoms present similarly to that
of FNHTR; fevers, chills, and mild dyspnea. However,
acute hemolytic transfusion reactions can result in
hypotension, shock, and eventual death. Initial treatment
includes stopping the blood transfusion and administering
normal saline. The blood bank must be notified
immediately. Necessary labs include a plasma free
hemoglobin, direct antibody test, lactate dehydrogenase
(LDH), haptoglobin, and repeat type and cross-match.
TRALI more commonly occurs with the transfusion of
plasma; however, it can also occur when pRBC are
transfused. In order for TRALI to be diagnosed, patients
must develop evidence of acute lung injury within six
hours of transfusion of blood products. In addition, other
etiologies of acute lung injury must be ruled out.31
Discussion
Given the risks and benefits in these unique, but not
separate, clinical populations, what is a clinician to do?
We advocate physicians consider utilizing more restrictive
transfusion thresholds. For iron deficiency anemia, try
alternative methods for increasing patients hemoglobin
(oral iron, IV iron, etc). In cases of acute upper
gastrointestinal bleeding, we agree with Villanueva et al.5

that blood transfusions should be reserved for patients

with hemoglobin levels less than 7 g/dL or those who are
symptomatic (hypotension, dyspnea, chest pain, etc.).
However, it is more difficult to make recommendations for
patients with acute coronary syndromes and chronic
ischemic heart disease. In these instances, we recommend
following the 2012 AAAB guidelines for pRBC
transfusions. The AABB recommends in patients with
preexisting cardiovascular disease, considering transfusion
for hemoglobin levels less than 8 or in the presence of
symptoms.2 However, they could neither recommend for
nor against liberal versus restrictive transfusion thresholds
in hemodynamically stable patients with acute coronary
syndromes. Instead, let symptoms be your guide on
whether patients with acute coronary syndromes should
have their anemia treated aggressively (transfusion for hgb
greater than 8) or conservatively (transfusions for hgb less
than 8).
In summary, most recent studies advocate for restrictive
transfusion thresholds; we agree with this approach.
However, clinicians need to ensure the decision to
transfuse is not based solely on a number; remember to
take into account patients symptoms as well. Weigh the
risks and benefits of each unit of blood given; in doing so
you will be ensuring your patients receive the quality of
care they deserve.
REFERENCES
1. http://www.hhs.gov/ash/bloodsafety/2011-nbcus.
2. Carson JL, Grossman BJ, Kleinman S, et al. Red blood cell transfusion: A clinical
practice guideline from the AABB. Ann Intern Med. 2012;157(1):49-58.
3. Sherwood MW, Wang Y, Curtis JP, Peterson ED, Rao SV. Patterns and outcomes
of red blood cell transfusion in patients undergoing percutaneous coronary
intervention. JAMA. 2014;311(8):836-43.
4. Silvain J, Abtan J, Kerneis M, et al. Impact of red blood cell transfusion on platelet
aggregation and inflammatory response in anemic coronary and noncoronary
patients: The TRANSFUSION-2 study (impact of transfusion of red blood cell on
platelet activation and aggregation studied with flow cytometry use and light
transmission aggregometry). J Am Coll Cardiol. 2014;63(13):1289-96.
5. Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper
gastrointestinal bleeding. N Engl J Med. 2013;368(1):11-21.
6. Clark SF. Iron deficiency anemia. Nutr Clin Pract. 2008;23:128-41.
7. Guyatt GH, Oxman AD, Ali M, Willan A, McIlroy W, Patterson C. Laboratory
diagnosis of iron-deficiency anemia: An overview. JGIM 1992;7:145-53.
8. Tefferi A. Anemia in adults: A contemporary approach to diagnosis. Mayo Clin
Proc. 2003;78:1274-80.
9. Mast AE, Blinder MA, Gronowski AM, Chumley C, Scott MG. Clinical utility of
the soluble transferrin receptor and comparison with serum ferritin in several
populations. Clin Chem 1998;44:45-51.
Please note: Due to limited space, we are unable to list all references. You may
contact South Dakota Medicine at 605.336.1965 for a complete listing.
About the Authors:
Alan Cook, MD, PGY1, Internal Medicine Residency Program, University of South
Dakota Sanford School of Medicine.
Nate Miller, MD, Core Faculty Member, Internal Medicine Residency
Program/Academic Assistant Professor, University of South Dakota Sanford School of
Medicine; Avera Medical Group Hospitalists.

December 2015

545

Copyright of South Dakota Medicine is the property of South Dakota State Medical
Association and its content may not be copied or emailed to multiple sites or posted to a
listserv without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.