Pergamon

Transfus. Sci. Vol. 18, No. 3, pp. 447458, 1997

# 1997 Published by Elsevier Science Ltd. All rights reserved
Printed in Great Britain
PII: S0955-3886(97)00043-X
0955-3886/97 $17.00 + 0.00

The Red Cell Storage Lesion and

its Implication for Transfusion
Ian Chin-Yee, MD, FRCPC*{
Naveen Arya, BScy
Mark S. d'Almeida, PhD*

cal ofcer, O.H. Robertson, is generally

credited with extending the storage period for red cells up to 21 days by using
citrate and glucose solution (RousTurner medium) which combined an
anticoagulant with a nutritive substance. Subsequent generations of preservative media contained reduced
amounts of anticoagulant to minimize
citrate toxicity and added phosphate to
improve tonicity and pH. These modications have only modestly extended
the storage period from 21 up to 42 days.
More recent developments include the
addition of purine nucleosides to storage
media to maintain red cell adenosine
triphosphophate (ATP) and experimentation with additives to maintain 2,3diphosphoglycerate (2,3-DPG). These
trends represent a shift away from the
traditional blood bank focus on RBC
viability to include the quality of the
transfused red cells.
The ``storage lesion'' can be dened
as a constellation of changes occurring
to the red cell which eventually results
in irreversible damage and reduced
post-transfusion survival. The progressive corpuscular damage ultimately
limits the storage period. Most
reviews2,3 have highlighted the importance of physical alterations in the red
cell membrane resulting in part from
abnormalities in red cell metabolism
during the storage period. Although
traditionally the term ``storage lesion''
has been restricted to corpuscular
damage, recent evidence has shown that

& Storage of red blood cells in preservative medium is associated with metabolic, biochemical and molecular
changes to erythrocytes collectively
referred to as the ``storage lesion.'' In
addition to corpuscular injury, bioreactive substances including cytokines and
lipids accumulate in the medium during
storage. We review evidence for those
storage related changes and potential
clinical implications for red blood
cell transfusion. # 1997 Published by
Elsevier Science Ltd &

HISTORY OF BLOOD STORAGE

The development of transfusion medicine as a specialty can be attributed to
our ability to store blood ex vivo in its
liquid state. Early attempts at preservation focused on preventing coagulation
by the use of soluble salts such as oxalic
acid or sodium citrate.1 In these anticoagulant solutions blood could be stored
for only a few days making transfusion
therapy impractical. A Canadian medi-

*Department of Haematology, London Health Sciences

Centre, University of Western Ontario, Canada.
yFaculty of Medicine, University of Western Ontario,
Canada.
{Author for correspondence at Room 2760, Phase 1, London Health Sciences Centre, 800 Commissioners Road
East, London, Ontario N6A 4G5, Canada. Fax:1-(519)685-8477; e-mail: [email protected]

447

448 Transfus. Sci. Vol. 18, No. 3

a number of bioreactive substances

accumulate in the medium during storage.4,5 These substances also inuence
the quality of the transfused blood product and may account for some of the
adverse reactions to transfusion. This
review will examine: the current
knowledge of the red cell storage lesion,
evidence for the accumulation of other
bioreactive substances during storage of
RBC, and potential clinical consequences of transfusing a ``stored'' red blood
cell product.

NORMAL RED CELL AND OXYGEN

DELIVERY
Over a century of research has helped
unravel the intricacies of these highly
specialized cells, evolved for the efcient delivery of oxygen (O2) to tissue.
To accomplish this goal, atmospheric
O2 is transferred from the lungs to haemoglobin contained within the RBCs at
the alveolar/capillary interface in the
lung. The red cell must maintain heme
iron in a functionally reduced state
(Fe2+) to fully bind O2. After passage
through the lungs, haemoglobin-O2
binding sites are near complete saturation and the process of transferring
bound O2 to tissue begins. Adequate
levels of 2,3- DPG facilitate O2 unloading, particularly in the microcirculation
where partial pressure of O2 is low and
the majority of O2 is transferred to tissue.
Recently, the traditional view of
the red cell and the respiratory cycle has
been expanded to include the role that
haemoglobin plays in binding nitric
oxide (NO), an important regulator of
vascular tone.6 In a series of experiments Jia et al.6 demonstrated that haemoglobin binds NO in at least two
distinct sites, the direct attachment of
NO to the heme iron or at the cysteine
residue (Cys 93) as s-nitrosylated haemoglobin (SNOHb). SNOHb levels were
found to be high in the arterial circulation and low in venous blood in keeping
with uptake in lung and release of NO

during systemic circulation. Donation

of NO by SNOHb causes vasodilation
suggesting that red cell haemoglobin
could potentially autoregulate blood
ow. The red cells, therefore, can be
viewed as regulating the two critical
components of O2 delivery, the binding
and release of O2, and the distribution
of blood ow within the microcirculation through regulation of NO homeostasis.
In addition to maintaining functional haemoglobin, to deliver O2
within the microcirculation the red cell
(diameter 810 ), must maintain exibility to negotiate capillary bed with
luminal diameters of 35 . The
remarkable deformability of the RBC is
principally dependent on membrane
surface area to volume, cytoplasmic
viscosity and to lesser degree on membrane exibility. The lipid bilayer
membrane is composed of choline-containing phospholipids (phosphatidyl
choline and sphingomyelin) predominantly in the outer layer and amino
phospholipids (phosphatidylserine and
phosphatidylethanolamine) found preferentially in the inner surface. The
lipid bilayer is penetrated by integral
membrane proteins and is supported by
a highly organized cytoskeleton, a
lamentous network of proteins
including spectrin, ankyrin, actin and
band 4.1.
Maintenance of the membrane
properties and functional haemoglobin
are in part dependent upon the energy
metabolism of the RBC. During storage,
metabolic changes in the RBC are
believed to underlie many of the functional alterations in the membrane,
ultimately reducing deformability and
limiting cell viability following transfusion. The red cell derives its energy primarily from the glycolysis of glucose to
lactic acid supplying the necessary fuel
to build high energy phosphate bonds of
ATP. Thus one of the goals of most preservative media has been the maintenance of adequate red cell energy
metabolism and in particular the levels
of ATP.

Red Cell Storage Lesion 449

RBC Shape and Metabolism During

Storage
A predictable sequence of shape changes
occur to the red cell during the storage
period. From its normal biconcave disc
shape a stored RBC undergoes crenation
forming an echinocyte and subsequently swells to form a crenated sphere
with multiple long spicules called a
spheroechinocyte.3 These spicules are
eventually shed as lipid vesicles leading
to spherocytosis. Transformation from a
biconcave disc to a sphere is associated
with a loss of surface to volume ratio,7
an increase in mean cell haemoglobin
increased
concentration
(MCHC),8
osmotic fragility and a loss of deformability.9 Several lines of evidence
associate morphologic changes during
storage with erythrocyte ATP depletion.
Nakao et al.10 reproduced the shape
changes by depleting ATP in erythrocytes and demonstrated that, by restoring ATP levels with adenosine,
echinospherocytic cells could revert to
their previous discoid shape. Moreover,
Haradin et al.7 showed a high correlation between the percent of cells that
failed to revert to a discoid shape and
post transfusion RBC survival. There
are, however, important differences
between profound ATP depletion produced articially and the changes
occurring during prolonged storage.
Typically, red cell ATP slowly declines
by 8090% during storage and is associated with more gradual morphologic
changes. After 8 weeks of storage a large
number of irreversible spherocytes persist despite rejuvenation of normal ATP
levels.3 More importantly, biomechanical measurements of RBC surface to
volume ratio and cytoplasmic viscosity
appear to precede the reduction in
ATP suggesting that ATP depletion is
not the cause of these early changes
which characterize the storage lesion.11
Overall, there is a poor correlation
between RBC ATP levels and posttransfusion survival except when ATP
levels are below 50%.12 This inconsistency is further emphasized by studies

demonstrating that maintaining supranormal ATP levels have little impact

upon
post-transfusion
survival.13
Whereas red cell ATP depletion in vitro
can reproduce many shape changes
observed during storage, ATP induced
defects alone do not explain RBC storage lesion or determine post transfusion viability.
Although ATP depletion may not
directly account for the membrane
lesion, it has been postulated that a
reduction in ATP may lead to an
increase in secondary mediators of
injury such as intracellular calcium or a
reduction in phosphorylation of other
protein or lipid kinases, important in
maintaining red cell integrity.3 An analysis of inositol phospholipids shows a
shift in balance of the phosphorylationdephosphorylation cycle towards
dephosphorylation
in
erythrocytes
during their second and third week of
storage.14 An increase in the dephosphorylated form, phosphatidylinositol4-phosphate, is accompanied by the
appearance of echinocytes suggesting a
relationship between dephosphorylation
and shape change.
Calcium has drawn much attention
as an important molecule in red cell
injury during storage. In particular,
micromolar concentrations of calcium
(less than 100 micromoles) cause intracellular red cell potassium loss with
cellular dehydration, echinocytosis and
microvesiculation.15 Calcium loading
induced by ionophores, in vitro, differs
signicantly from storage-related changes, reproducing only the very late
biomechanical changes of ATP depletion and not the early alterations in
surface area to volume relationships.
The use of EDTA to reduce calcium
accumulation has had no effect on red
cell shape change produced by ATP
depletion.11 More importantly, the red
cells possess a calcium-magnesium
ATPase that is active at storage temperatures (4 C) and can pump out calcium at a rate of almost 5000 times the
steady state leak into cells.16 It is only
under conditions of extreme ATP

450 Transfus. Sci. Vol. 18, No. 3

depletion, well beyond those observed

during standard storage condition, that
calcium ATPase fails and signicant
changes in intracellular calcium are
observed.

Biomechanical Changes in RBC

During Storage
Since depletion of erythrocyte ATP may
simply represent a metabolic marker of
cellular injury during storage, other
measures that correlate with posttransfusion survival have been sought.
The early observation of Haradin et al.7
between post-transfusion survival and
percentage of cells rejuvenated to their
normal disc shape suggested that sensitive biomechanical measurements may
provide a marker of the storage lesion.
Initial attempts to correlate increased
osmotic fragility with surface to volume
ratio showed a relatively poor correlation but subsequent studies suggested
that the lack of correlation was masked
by changes in volume due to alterations
in the cellular ionic content and lactate
accumulation during storage.12 If red
cells were returned to an isotonic state,
there was a small population of stored
RBCs with increased osmotic fragility
and reduced post transfusion viability.
Osmotic fragility of RBC stored in
Adsol preservative solution initially
demonstrate a single normally distributed population but after extended storage of up to 60 days, two distinct
populations emerge indicating a subpopulation of susceptible osmotically fragile RBCs.17 It is speculated that the
fragile population represents older erythrocytes that are particularly prone to
storage injury.
An increase in MCHC observed
during storage is attributed to cell
shrinkage due to loss of water and other
intracellular ions.8 During normal RBC
life span, an increase in MCHC is
observed in the denser older erythrocytes that can increase internal viscosity of a cell up to 10-fold.18 Preservative
solutions may swell cells, masking their

true volume changes.8 Turner et al.8

simulated transfusion by incubating
cells stored in CPDA-1 for greater than
21 days in autologous plasma for 24 h
and found a shrinkage to their true
volume. As cells returned to their true
volume, MCHC increased and their lterability decreased.
More sophisticated measurements
of biomechanical properties of the red
cell include micropipette measurements of the aspirated red cell membrane or ektacytometry, a technique
which measures shape changes during
laminar shear stress monitored by laser
refractometry. The advantages and disadvantages of each of these techniques
has been previously reviewed.19 Each of
these methods are sensitive to different
cellular properties. Filtration, ektacytometry and micropipette aspiration into
3  pipettes reects the surface area to
volume relationships and cytoplasmic
viscosity. Aspiration into 0.5  pipettes
measures the isolated membrane property referred to as elastic shear modulus.2 During storage of red cells, subtle
changes in surface to volume relationships are detectable by ektacytometry in
hypotonic conditions at 12 weeks20
and show a high correlation with posttransfusion survival. In contrast, little
change is noted in the membrane elastic
shear modulus. A high correlation
between sensitive measurements of
surface to volume ratio, cytoplasmic
viscosity and post-transfusion viability,
highlights the importance of maintaining red cell deformability if the cells are
to survive in the microcirculation of
organs such as the spleen. Changes in
deformability do not preclude that other
factors are important in clearance of
injured stored red cells.
Changes in biomechanical properties of the red cell membrane could
result from alteration in the lipid
bilayer, membrane proteins, the cytoskeleton, or the interaction between
these three components. The importance of the progressive loss of lipid
membrane during the storage period
rst described by Haradin,7 has been

Red Cell Storage Lesion 451

re-emphasized in contemporary studies

aimed at counteracting microvesiculation, which appear to have the greatest
effect on post-transfusion viability.21
Lipid microvesicles, measuring 80 to
200 nm, can be detected at approximately two weeks of storage in the
sediments from high speed centrifugation of plasma of stored whole blood.
Surprisingly the microvesicles are to a
large degree derived from the younger
population of RBCs rather than the
more dense, older cells as might be
expected.22 The lipid vesicles from
young and old cells do not differ qualitatively maintaining a normal cholesterol to phospholipid ratio and contain
haemoglobin as well as other integral
membrane proteins including Band 3,
4.1 and 7.23,24 Conspicuously absent
from the microvesicles is spectrin, the
most abundant protein in the cytoskeleton. This has lead to much speculation about a potential abnormal
interaction between the lipid bilayer
and the underlying cytoskeleton. It is
postulated that even subtle alterations
in the lipid bilayer interaction with
spectrin or other membrane proteins
could signicantly alter the surface area
to volume relationship. Quantitative
reduction in spectrin or other proteins
has not been observed in random samples of outdated blood.3 Qualitative
changes in spectrin's ability to form a
normal complex with protein 4.1 and
actin have been described.25 This qualitative abnormality in spectrin shares
many of the features rst described in
patients with the clinical phenotype of
hereditary spherocytosis. In this defect,
a decreased association of spectrin to
protein 4.1 results from spectrin's
increased susceptibility to oxidative
damage.26 The function of these abnormal spectrin molecules could be
improved by antioxidants. Oxidation of
spectrin has also been shown to occur
during storage and correlates with
membrane vesiculation in stored blood
suggesting a relationship between
microvesiculation
and
oxidative
injury.27

LIPID PEROXIDATION DURING

STORAGE
Oxidative damage to the membranes of
RBCs can effect membrane lipids as
well as involve proteins such as spectrin. Lipid peroxidation, the oxidative
deterioration of polyunsaturated fatty
acids as a result of oxygen free radicals,
is an important mechanism for cellular
injury, RBC aging and eventually death
of RBCs.28 The generation of oxygen
free radicals can result from interactions with environmental O2 and/or
haemoglobin bound O2 (endogenous
source) and may be responsible for oxidant membrane damage during storage.
The relative importance of lipid peroxidation during RBC storage is, however,
still unclear. Various antioxidant
mechanisms, including superoxide dismutase (SOD), glutathione peroxidase
(GPX), catalase and methehaemoglobinreductase, are in place in the RBC to
safeguard against this type of injury. As
stored blood approaches its outdated
period, the formation of Heinz bodies
(oxidized haemoglobin) has been noted
accompanied by a decrease in gluthionine levels, suggesting that gluthionine
may play a central role in RBC protection.29 Bhargava et al.30 found that oxidative injury to RBCs as measured by
malondialdehyde levels (MDA) was
minimal during blood storage in CPDA1 under standard blood bank conditions,
while the level of anti-oxidant enzymes
SOD and GPX were stable. In contrast,
Knight et al.28 found increasing levels of
MDA during storage with little relationship to gluthionine levels and that
metal chelators such as desferoxamine
effectively reduced MDA accumulation.
By adding the antioxidant drug, tirilazad
mesylate, to RBCs, stored in Adsol, Epps
et al.17 were able to decrease the number
of osmotically fragile RBC and the level
of F2- isoprostanes, a marker of lipid
peroxidation. Further study is required
to determine the role of oxidative injury
to erythrocytes and if the addition of
various antioxidants will improve
storage and post transfusion viability.

452 Transfus. Sci. Vol. 18, No. 3

RBC MEMBRANE PHOSPHOLIPID

ASYMMETRY
Loss of membrane lipid through vesiculation not only directly reduces surface
area but could potentially alter membrane phospholipid asymmetry. As previously mentioned, the normal lipid
bilayer contains an asymmetric distribution of choline and aminophospholipids actively maintained through energy
dependent aminophospholipid translocases. Alterations in membrane assymmetry leading to increased expression
of negatively charged aminophospholipids such as PS on the outer membrane
surface has been postulated to be a
major mechanism of normal red cell
senescence
and
clearance.31 The
increased PS expression may be related
to the susceptibility of PS rich unsaturated fatty acids to peroxidative breakdown during aging resulting in adhesion
of senescent RBCs to endothelium in
the spleen and their subsequent clearance.31 Although it is tempting to postulate that storage simply accelerates
membrane changes associated with
normal aging process, studies examining this hypothesis have shown conicting results. Geldwerth et al.32
demonstrated a reduction in activity of
aminophospholipid translocases during
storage that altered transbilayer mobility of PS, phosphotidylethanolamine
and phophotidylcholine but was unable
to detect any alteration in surface
PS expression using a phopholipase
technique. Diaz et al.33 generated
phenotypically aged cells by dialauroylphosphatidyl-induced vesiculation or in
vitro storage (not under blood bank
conditions), resulting in membrane
phospholipid scrambling, increased
outer membrane PS exposure and
enhanced clearance of the RBC. Using a
sensitive ow cytometric assay with a
uorescently labelled Annexin V, a protein ligand for PS, we have not observed
a signicant increase in RBCs expressing PS in 35 day old blood stored in
Nutricel under blood bank conditions
(Chin-Yee, unpublished observation).

Other similarities between normal

red cell aging and the storage lesion
have been observed. Dense senescent
RBCs have signicant reductions in
surface area/volume due to osmotic
water loss and membrane microvesiculation (reviewed in Ref. 34), similar to in
vitro storage. Formation of senescent
cell antigens from alterations in band 3
or galactosyl carbohydrates have been
postulated to lead to immunoclearance
of aged cells by binding autologous
antibodies.34 Antibodies to ``old'' stored
blood have also been described with
specicity for galactosyl carbohydrates
in two patients35 as well as a general
increase in binding of puried autologous IgG in stored RBCs.36 Other age
associated RBC changes have not been
extensively studied during storage.

NEW DEVELOPMENTS IN RBC

STORAGE
Reducing RBC microvesiculation has
become a major goal of the current generations of storage media. An unexpected benet of storing blood in plastic
containers made of polyvinyl chloride
(PVC) was the presence of leachable
plasticizers such as di-2-ethylhexylphthalate (DEHP). During storage, DEHP
dissolves into the blood with a signicant amount of phthalate associated
with the RBC membrane. The phthalate
appears to have a membrane stabilizing
effect since its elimination increased
microvesiculation by 50%, elevated
supernatant haemoglobin by 70%37 and
compromised post-transfusion recoveries. Although, no clinical consequences
of phthalate toxicity have been documented PVC-DEHP storage containers
have largely been replaced with PVCbutyryl-trihexyl.
Other methods to reduce microvesiculation have used hypotonic storage
media and other additives such as glycerol. The original rationale for the use
of hypotonic storage containing NH4Cl
medium was to increase cell volume
and prevent membrane loss through

Red Cell Storage Lesion 453

lipid vesiculation.38 Ironically, the

hypotonic
preservative
medium
increased the ATP levels which is currently viewed as the most important
factor in the improved post-transfusion
RBC's survival.38 Furthermore, the
amount of microvesicles shed is similar
in both hypotonic preservatives and
isotonic medium Adsol.39 Glycerol21
has a more direct effect on membrane
integrity reducing total amount of
microvesiculation and protein bands 3
and 4.1 in shed vesicles after 12 weeks
of storage. These investigators suggest
that reduced loss of critical membrane
proteins in microvesicles was an indicator of improved RBC preservation.

THE EFFECT OF WHITE BLOOD

CELL ON RBC STORAGE
Changes in RBC are not only affected by
intracorpusclar factors but exogenous
effects of contaminating white blood
cells (WBC) and their byproducts present in the storage medium. Leukocytes
may exert an indirect effect on RBC by
consuming glucose needed by the RBC
or by releasing bioreactive substances in
the storage medium injurious to the
RBC.40 An increase in haemolysis and
potassium leakage resulting from
altered membrane permeability during
storage has been attributed to leukocyte
enzymes such as elastase, collagenase
and cathepsin G and/or activated neutrophils liberating toxic O2 species.41
Greenwalt et al.41 found that leukodepleted units of RBC were signicantly
better after 56 days of storage with
reduced supernatant potassium, haemolysis, total vesicle membrane protein
shed and higher morphology scores
compared to non-leukodepleted RBCs.
These results support the hypothesis
that enzymes and eicosanoids released
by degenerating leukocytes and platelets may be damaging to stored erythrocytes. Leukodepleted blood also had
better preservation of metabolic parameters (including ATP), signicantly
reduced lactate production and glucose

consumption when compared to standard red cell concentrates and extended

the shelf life of stored cells (reviewed in
Ref. 40).

Bioreactive Substances in Storage

Medium
In addition to the impact of contaminating WBCs and their byproducts on
red cell viability, attention has recently
focused on the potential consequences
of transfusing bioreactive substances
including histamine,42 lipid43 and cytokines44 which have been identied in
the storage medium. Cytokines known
to increase during storage of RBC or
platelets include interleukin 1 beta
(IL1), tumour necrosis factor alpha
(TNF), and interleukin 8 (IL8).42,44,45
Both TNF and IL8 are derived from
WBC and can potentially activate
neutrophils. Several of the cytokines
generated during storage including
TNF, IL1 and IL8 have potent pyrogenic activity, can recruit neutrophils
from the bone marrow and cause further
release of cytokines.4
The consequences of transfusing
bioreactive substance such as histamine, lipid and cytokines is an important area of ongoing study. In platelet
transfusions,5 febrile reactions are
strongly correlated with IL1 and IL6 in
the plasma supernatant. The effect of
the cytokine generated during storage of
RBC concentrates is less obvious but
may account for some febrile transfusion reactions.46
Lipids generated during storage of
RBCs43 are also capable of directly
priming neutrophil NADPH oxidase.
Sillimann et al.43 demonstrated that the
supernatant of stored blood contain
lipids with platelet activating factor
(PAF)-like activity capable of priming
neutrophils, the effector cell implicated
in tissue injury.47 Unlike cytokines
generated during storage, the PAF-like
lipids do not appear to be the product of
contaminating WBC since they have
been found in products with relatively

454 Transfus. Sci. Vol. 18, No. 3

few leukocytes. Recently, these investigators48 perfused an isolated endotoxin

treated rat lung preparation with plasma
from 42 day old blood or PAF-like lipids,
both, of which produced vasoconstrictor
responses and lung damage. These
results suggest that the priming of neutrophils by plasma supernatant from
stored blood can mediate tissue injury
in a setting of previous endotoxic insult.
Excessive neutrophil activity has been
implicated not only in the rare complication of transfusion related acute lung
injury (TRALI)49 but in adult respiration
distress syndrome (ARDS), ischemia
reperfusion and other inammatory
conditions common in the critically
ill.47 The potential interaction between
a patient's medical condition and transfusion may be particularly important
since the critically ill population typically has the highest transfusion
requirements and may be at greatest
risk of the adverse consequences from a
stored product.

CLINICAL CONSEQUENCES OF
TRANSFUSING STORED BLOOD
Despite the evidence for a red cell storage lesion and bioreactive substances
accumulating in the stored product, the
clinical consequences of transfusing
``old'' stored blood are uncertain. Prolonged storage has a known impact on
post-transfusion RBC survival3 but
other adverse effects are uncommonly
identied. Although this could be used
as an argument against the clinical
relevance of any of the observed changes, it more likely represents the ability
of the body to deal with perturbations to
the system. The ability of the reticuloendothelial
system
to
remove
damaged non-deformable red cells will
reduce the potential impact these cells
may have on the microcirculation. The
patient's ability to cope with a suboptimal product, however, should provide
little comfort to clinicians or patients
since the majority of transfusions are
given to patients who are ill with

potentially impaired reticuloendothelial

function.
In the absence of good clinical evidence, some clinicians have advocated
the use of ``fresh'' blood in specic
patient population. In neonates, for
example, the accumulation of potassium in the storage media has led some
blood banks to either provide ``fresh''
blood or washed units for transfusion. In
the massively transfused patients,
transfusion of large amounts of stored
blood, decient in 2,3-DPG may have
adverse clinical consequences on O2
delivery in patients whose tissue balance is precarious.50 Fresh blood has
also been advocated for patients who
require frequent transfusions because
improved viability of RBC reduce the
requirements for transfusion, minimizing the complication of iron overload.
Even in these specic situations, blood
bank practice varies widely with no
consensus when ``fresh'' blood products
are indicated.
The lack of evidence for clinical
sequelae of transfusing a ``sub-optimal''
product may also be in part related to
the manner by which we evaluate
transfusion efcacy. The goal of RBC
transfusion is to increase haemoglobin
concentration, thereby improving convective O2 delivery to the tissue. It is
implicitly assumed that an increase in
haemoglobin following transfusion will
increase the O2 content of blood and
deliver O2 in a form that can be readily
used by tissue. Actual measurements of
O2 delivery and utilization are rarely
monitored in clinical practice where
most clinicians rely on an increase in
haemoglobin as a marker of efcacy.
The efcacy of a red cell transfusion,
however, should ideally be measured by
clinical benet or a surrogate marker
such as an improvement in tissue oxygenation.51 Standards developed to
quality assure the RBC components
have arbitrarily dened the adequacy of
RBC transfusion in terms of viability
(greater than 70% survival stored RBCs
in 24 h following transfusion) rather
than efcacy per se. Post-transfusion

Red Cell Storage Lesion 455

viability would be best dened as a

marker of transfusion ``activity'' rather
than efcacy. The activity of a red cell
transfusion demonstrated by the results
obtained from a biological assay (RBC
survival) would be considered insufcient to meet suggested Food and Drug
Administration (FDA) requirements of
clinically signicant benet which has
been suggested for the evaluation of
RBC substitutes.51 It is ironic that the
criteria now being suggested for the
evaluation of red cell substitutes have
rarely been applied to red cell transfusion or the effects of storage on transfusion efcacy.
Nevertheless, a few studies in critically ill patients have directly measured
tissue oxygenation following red cell
transfusion in patients.5254 Dietrich et
al.52 assessed the benet of RBC transfusion in terms of tissue O2 utilization
(O2 consumption, decreased lactate) or
decrease myocardial work and found no
improvement in these parameters.
Other studies have employed measurements of gastric mucosal pH (pHi),53,54
an indicator of mucosal oxygenation,
and have demonstrated splanchnic
ischemia in critically ill patients following red cell transfusion. Furthermore, Marik et al.54 found an
association between the fall in gastric
pHi and transfusion of red blood cells
stored for greater than 15 days. These
studies in the critically ill suggest that
RBC transfusion may not acutely
improve tissue O2 utilization and a possible inuence of storage duration on
RBC efcacy.
The efcacy of red cell transfusion
can be more readily studied in animal
models. Unfortunately, there are relatively few published animal studies that
have investigated the impact of red cell
transfusion on the cardiovascular system
and O2 delivery/utilization. Simchon
and colleagues55 have studied the effect
of reduced deformability of glutaraldehyde treated erythrocytes on regional
blood ows. These investigators demonstrated that in rats exchange-transfused
with less deformable erythrocytes, there

was an increased entrapment of these

cells in the microcirculation and subsequent blockage of ow to selective vascular beds. The majority of altered
erythrocytes were found to be entrapped
in organs of the splanchnic vascular bed
(spleen, and liver) as well as in the lungs
and sternum. Blood ow to these
regions was also depressed in accordance with the degree of erythrocytes
entrapment in each specic organ.
While glutaraldehyde treatment does
not completely mimic the effect of the
storage-related deformability changes, it
suggests that reduced deformability not
only affects post-transfusion viability
but may have direct effects of microvascular blood ow and O2 delivery.
Whether poorly deformable stored erythrocytes become entrapped in the
microcirculation and impair, rather
than improve, blood ow and O2 delivery to certain vascular beds remains to
be demonstrated.
In a separate study that investigated
the inuence of RBC deformability on
canine renal blood ow (measured by
electromagnetic ow cytometry), glutaraldehyde stiffened erythrocytes were
not found to alter ow to the kidneys56
which was consistent with the report by
Simchon.55 Levi et al.56 did nd, however, that plasma renin activity
increased by about 50% (2.9 to 4.4 ng
mL1 h1) and suggested that this
increase in plasma renin activity may be
signicant in some types of hypertension. Whether transfusion of poorly
deformable stored RBCs also increases
renin activity is not known but deserves
further study.
Studies done in our laboratory have
directly examined the impact of transfusion on O2 delivery and uptake in different disease states including sepsis. In
septic rats isovolemically haemodiluted
to the point of O2 supply-dependency,
the direct effect of transfusing ``old'' rat
red cells and ``fresh'' cells on arterial
lactate and O2 consumption was measured by a Krogh respirometer.57 Using
this model, a consistent lack of efcacy
of rat blood stored for greater than 21

456 Transfus. Sci. Vol. 18, No. 3

days to acutely improve tissue O2 consumption and decrease arterial lactate

was observed when compared to fresh
cells. A similar lack of efcacy has been
observed in non-septic animals. These
observations raise important questions
about the impact of storage on the RBC
ability to acutely deliver O2 in a utilizable form. To properly investigate the
effect of RBC transfusion on the cardiovascular system and regional haemodynamics, it is necessary to study the
microcirculation, the major site of all
gas and nutrient exchange between the
blood and tissues. Preliminary studies
from our research group using intravital
microscopy suggest that stored RBC
become entrapped in the microcirculation of the gut mucosa and that the
normal microcirculatory arrangement
may be disrupted.58 Further research
examining the consequences of RBC
transfusion, is required to clarify the
impact of ``the storage lesion'' on oxygenation.

SUMMARY
Research over the past three decades has
claried the metabolic, biomechanical
and molecular changes occurring during
the storage of RBC. Several factors
appear to contribute to erythrocyte
injury including metabolite depletion,
lipid and protein peroxidation and WBC
derived factors including enzymes and
cytokines. The ``storage lesion'' results
from the cumulative effect of these
inuences on RBCs and directly limits
post-transfusion viability. In the storage
medium, numerous bioreactive substances accumulate with potential
proinammatory effects. The clinical
and physiologic impact of storage-related changes to RBC and storage medium
on tissue oxygenation requires further
study. The current evidence suggests,
however, that blood bankers need to be
aware of both the quality of the red cell
and the byproducts generated in the
storage medium, either of which is theoretically harmful to the recipient. Further

investigation to distinguish between

adverse effects related to corpuscular
injury as opposed to byproducts in the
storage medium is critical if we are to
eventually improve the quality of the
transfused product.

Acknowledgements
M. S. d'Almeida is the recipient of a Canadian Red Cross
Society Career Development Fellowship. We are grateful
for the assistance of Dr R. M. Barr and A. Heard in
reviewing the manuscript and B. Stevenson for her
administrative assistance.

REFERENCES
1. Diamond L: A history of blood transfusion, in Wintrobe, MM, (ed.): Blood,
Pure and Eloquent. New York,
McGraw-Hill Book Company, 1980,
659.
2. Card RT: Red cell membrane changes
during storage. Trans Med Rev 1988;
2:4047.
3. Wolfe LC: The membrane and the
lesions of storage in preserved red cells.
Transfusion 1985; 25 (3):185202.
4. Davenport RD, Kunkel SL: Cytokine
roles in haemolytic and nonhaemolytic
transfusion reactions. Transfus Med Rev
1994; 8:157.
5. Heddle NM, Klama L, Singer J, et al.:
The role of the plasma from platelet
concentrates in transfusion reactions.
New Engl J Med 1994; 331 (10):625628.
6. Jia L, Bonaventrua C, Bonaventura J, et
al.: S-nitrosohaemoglobin: a dynamic
activity of blood involved in vascular
control. Nature 1996; 380:221226.
7. Haradin AR, Weed RI, Reed CF: Changes in physical properties of stored erythrocytes. Transfusion 1969; 9:229237.
8. Turner S, Williams AR, Rees JMH: The
role of mean corpuscular haemoglobin
concentration in limiting the storage
life of human blood. Vox Sang 1987;
52:177181.
9. Beutler E, Kuhl W, West C: The osmotic
fragility of erythrocytes after prolonged
liquid storage and after reinfusion.
Blood 1982; 59:11411147.
10. Nakao M, Nakao T, Yamazoes S: Adenosine triphosphate and maintenance of
shape of human red cells. Nature 1960;
187:945946.
11. Feo C, Mohandas N: Clarication of role

Red Cell Storage Lesion 457

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.
23.

24.

of ATP in red-cell morphology and

function. Nature 1977; 265:166168.
Dern R, Brewer G, Wiorkowski J: Studies on the preservation of human
blood. The relationship of erythrocyte
adenosine triphosphate levels and other
in vitro measures to red cell storageability. J Lab Clin Med 1967; 69:968978.
Wood L, Beutler E: The viability of
human blood stored in phosphate adenine media. Transfusion 1967; 7:401
408.
Swietochowska K, Piascik R, Jaroszewicz K, et al.: Human stored blood inositol phopholipids. Acta Physiologica
Hungarica 1991; 78:283291.
Palek J, Liu S: Effect of procaine HCl on
ATP: calcium-dependent alteration in
red cell shape and deformability. Blood
1977; 50:155164.
Szasz I, Sarkadi B, Gardos G: Changes in
the Ca2+-transport processes of red cells
in cold storage in ACD. Br J Haematol
1978; 39:559568.
Epps DE, Knechtel TJ, Baczynskyj O, et
al.: Tirilazad mesylate protects stored
erythrocytes against osmotic fragility.
Chemistry and Physics of Lipids 1994;
74:163174.
Cokelet GR, Meiselman HJ: Rheological
comparison of haemoglobin solutions
and erythrocyte suspensions. Science
1968; 162:275277.
Mohandas N, Chasis JA: Red blood cell
deformability, membrane material properties and shape: regulation by transmembrane, skeletal and cytosolic
proteins and lipids. Semin Hemat 1993;
30:171192.
Card RT, Mohandas N, Perkins HA, et
al.: Deformability of stored red blood
cells. Relationship to degree of packing.
Transfusion 1982; 22:96101.
Dumaswala UJ, Dumaswala RU, Levin
DS, et al.: Improved red blood cell preservation correlates with decreased loss
of bands 3, 4.1, acetylcholinestrase, and
lipids in microvesicles. Blood 1996;
87:16121616.
Greenwalt TJ, Dumaswala UJ: Effect of
red cell age on vesiculation in vitro. Br J
Haematol 1987; 68:465467.
Greenwalt TJ, Bryan DJ, Dumaswala UJ:
Erythrocyte membrane vesiculation and
changes in membrane components during storage in citrate-phosphate-dextrose-adenine-1. Vox Sang 1984; 47:261
270.
Rumsby MG, Trotter J, Allan D, et al.:
Recovery of membrane micro-vesicles

25.

26.

27.

28.
29.

30.

31.

32.

33.

34.
35.

36.

37.

from human erythrocytes stored for

transfusion: a mechanism for the erythrocyte discocyte-to-spherocyte shape
transformation. Biochem Soc Trans
1977; 5:126128.
Wolfe LC, Byrne A, Lux SE: Molecular
defect in the membrane skeleton of
blood bank stored red cells. Abnormal
spectrin-protein 4.1-actin complex formation. J Clin Invest 1986; 78:1681
1686.
Becker P, Morrow J, Lux S: Abnormal
oxidant sensitivity and beta-chain
structure of spectrin in hereditary
spherocytosis associated with defective
spectrin-protein binding. J Clin Invest
1987; 80:557565.
Wagner GM, Chiu DTY, Qju JH, et al.:
Spectrin oxidation correlates with
membrane vesiculation in stored RBCs.
Blood 1987; 69:17771781.
Knight JA, Voorhees RP, Martin L, et al.:
Lipid peroxidation in stored red cells.
Transfusion 1992; 32:354357.
Lochant N, Noble N, Myrhe B, et al.:
Antioxidant metabolism during blood
storage and its relationship to posttransfusion red cell survival. Amer J
Hematol 1984; 17:237249.
Bhargava AB, Pavri RS, Bhatia HM: Susceptibility of red cells to oxidativeinjury during blood perservation. Indian
J Medical Research 1988; 87:202205.
Schroit AJ, Madsen JW, Tanaka Y: In
vivo recognition and clearance of red
blood cells containing phosphatidylserine in their plasma membranes. J Biol
Chem 1985; 260:51315138.
Geldwerth D, Kuypers FA, Butikofer P,
et al.: Transbilayer mobility and distribution of red cell phospholipids during
storage. J Clin Invest 1993; 92:308314.
Diaz C, Morkowski J, Schroit AJ: Generation of phenotypically aged phosphatidylserine-expressing erythrocytes by
dilauroylophophatidylcholine-induced
vesiculation. Blood 1996; 87:2956
2961.
Bartosz G: Erythrocyte aging: Physical
and chemical membrane changes. Gerontology 1991; 37:33.
Krugluger W, Koller M, Hopmeier P:
Development of a carbohydrate antigen
during storage of red cells. Transfusion
1994; 34:496500.
Gareau R, Brisson GR, Goulet H, et al.:
Blood banking-induced senescent modications on red blood cells. Cell Molec
Bio 1992; 38:395398.
Estep TN, Pedersen RA, Miller TJ, et al.:

458 Transfus. Sci. Vol. 18, No. 3

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.
48.

Characterization of erythrocyte quality

during the refrigerated storage of whole
blood
containing
di-(2-ethylhexyl)
phthalate. Blood 1984; 64:12701276.
Meryman HT, Hornblower LS, Syring
RL: Prolonged storage of red cells at 4
degrees celcius. Transfusion 1986;
26:500505.
Greenwalt
TJ,
McGuinness
CG,
Dumaswala UJ, et al.: Studies in red
blood cell preservation. Vox Sang 1990;
58:9499.
Andreu G: Early leukocyte depletion of
cellular blood components reduces red
blood cell and platelet storage lesions.
Semin Hemat 1991; 28:2225.
Greenwalt TJ, Sostok CZ, Dumaswala
UJ: Studies in red blood cell preservation: 2. Comparison of vesicle formation, morphology, and membrane lipids
during storage in AS-1 and CPDA-1. Vox
Sang 1990; 58:9093.
Smith KJ, Sierra ER, Nelson EJ: Histamine, IL-1, and IL-8 increase in packed
RBCs stored for 42 days but not in RBCs
leukodepleted pre-storage. Transfusion
1993; 33:53S.
Silliman CC, Clay KL, Thurman GW, et
al.: Partial characterization of lipids that
develop during the routine storage of
blood and prime the neutrophil NADPH
oxidase. J Lab Clin Med 1994; 124:684
694.
Muylle L, Peetermans ME: Effect of prestorage leukocyte removal on the cytokine
levels
in
stored
platelet
concentrates. Vox Sang 1994; 66:1417.
Stack G, Baril L, Napychank P, et al.:
Cytokine generation in stored, white
cell-reduced, and bacterially contaminated units of red cells. Transfusion 1995;
35:199203.
Federowicz I, Barrett BB, Andersen JW,
et al.: Characterization of reactions after
transfusion of cellular blood components that are white cell reduced
before storage. Transfusion 1996; 36
(1):2128.
Malech HL, Gallin JI: Neutrophils in
human diseases. New Engl J Med 1987;
317:687694.
Allard J, Silliman CD, Ambruso D, et
al.: A model of transfusion related acute

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

lung injury (TRALI). Respiratory and

Critical
Care
Medicine
1996;
153(Abstract)(4):May 12.
Popovsky MA, Moore SB: Diagnostic
and pathogenetic considerations in
transfusion-related acute lung injury.
Transfusion 1985; 25:573577.
Sugerman HJ, Davidson DT, Vibul S,
et al.: The basis of defective oxygen
delivery from stored blood. Surgery,
Gynecology, and Obstetrics 1970:733
741.
Center
for
Biologics
Evaluation
Research: Points to consider on efcacy
evaluation of haemoglobin-and peruorocarbon-based oxygen carriers. Transfusion 1994; 34:712713.
Dietrich KA, Contrad SA, Herbert CA,
et al.: Cardiovascular and metabolic
response to red blood cell transfusion in
critically ill volume resusciated nonsurgical patients. Crit Care Med 1990;
18:940944.
Silverman HJ, Tuma P: Gastric tonometry in patients with sepsis: Effects of
dobutamine infusions and packed red
blood cell transfusions. Chest 1992;
102:184188.
Marik PE, Sibbald WJ: Effect of storedblood transfusion on oxygen delivery in
patients with sepsis. JAMA 1993;
269:30243029.
Simchon S, Jan KM, Chien S: Inuence
of reduced red cell deformability on
regional blood ow. Am J Physiol 1987;
253:H898.
Levi E, Baskurt OK, Dikmenoglu N, et
al.: Effect of erythrocyte deformability
on renal haemodynamics and plasma
renin activity. Am J Nephrol 1992;
12:3740.
Fitzgerald RD, Martin CM, Dietz GE, et
al.: Transfusing RBCs stored in CPDA-1
for 28 days fails to improve tissue oxygenation in rats. Crit Care Med, 1997;
25:726732.
Martin CM, Iwao Y, Potter R, et al.:
Decreased mucosal capillary perfusion
following transfusion of stored blood in
control and septic rats. American Thoracic Society 1996 International Conference 1996;(Abstract) New Orleans,
Louisiana: 1115 May.