Transfusion and Apheresis Science 27 (2002) 83–92

www.elsevier.com/locate/transci

Massive blood transfusion in the elective surgical setting

Wendy N. Erber *

Department of Haematology, Western Australian Centre for Pathology and Medical Research (Path Centre), Locked bag 2009,
Nedlands, WA 6909, Australia

Abstract
Massive haemorrhage in elective surgery can be either anticipated (e.g. organ transplantation) or unexpected.
Management requires early recognition, securing haemostasis and maintenance of normovolaemia. Transfusion
management involves the transfusion of packed red cells, platelet concentrates and plasma (fresh frozen plasma and
cryoprecipitate). Blood product support should be based on clinical judgment and be guided by repeated laboratory
tests of coagulation. Although coagulation tests may not provide a true representation of in vivo haemostasis, they do
assist in management of haemostatic factors. Below critical levels (prothrombin time or activated partial thrombo-
plastin time >1.8; fibrinogen <1.0 g/l; platelet count <80  109 l 1 ) it is difficult to achieve haemostasis. Despite
seemingly adequate blood component therapy there remain situations where haemorrhage is uncontrollable. In this
setting, alternative approaches must be considered. These include the use of other blood products (e.g. prothrombin
complex concentrates; fresh whole blood; fibrin glue) and pharmacological agents (e.g. aprotinin). Complications of
massive transfusion result in significant morbidity and mortality. These may be secondary to the storage lesion of the
transfused blood products, disseminated intravascular coagulation, hypothermia or hypovolaemic shock. The use of
fresh blood products and leucocyte-reduced packed red cells and platelets, may minimise some of the adverse clinical
sequelae.
2002 Published by Elsevier Science Ltd.

Keywords: Haemorrhage; Massive transfusion; Blood products; Surgery

1. Introduction used are the transfusion of >3000 ml, or more

Massive transfusions can occur in the settings
of trauma, surgery and obstetrics and pose a major
therapeutic challenge to clinicians and transfusion
personnel. There is no universally accepted defi-
nition of ‘‘massive transfusion’’, but those used
are based on both the volume of transfusion and
the rate at which this occurs. The most commonly
*
Tel.: +61-8-9346-2893; fax: +61-8-9346-3848.
E-mail address: wendy.erber@health.wa.gov.au
Erber).
1473-0502/02/$ - see front matter
2002 Published by Elsevier Science Ltd.
PII: S 1 4 7 3 - 0 5 0 2 ( 0 2 ) 0 0 0 2 9 - 0
than 10 units of packed red cells, over 24 h. Other
definitions are:
(a) the replacement of 50% of blood volume in 3 h,
(b) blood loss of 150 ml/min [1],
(c) loss of 1.5 ml blood per kg body weight per
minute over 20 min, and,
(d) when blood loss is so rapid and severe that
blood product support with red cells and vol-
ume replacement with fluids exceeds the com-
pensatory mechanisms of the body [2].
(W.N. Massive transfusion can occur in a number of
elective surgical settings including cardiac surgery
84 W.N. Erber / Transfusion and Apheresis Science 27 (2002) 83–92
Table 1
Elective surgical settings associated with massive haemorrhage
and transfusion
Type of elective surgery Mechanism of massive
haemorrhage
Complex and re-do surgery Surgical damage
Neurosurgery Thromboplastin release
Cancer surgery Procoagulopathic cytokines
DIC
Cardiopulmonary bypass Anticoagulant therapy (e.g.
surgery heparin)
Platelet dysfunction:
• Antiplatelet therapy prior
to surgery
• Platelet activation from
bypass
Platelet consumption
Hypothermia
Haemodilution
Liver transplantation Hepatic ischaemia; anhepatic
Pre-existing coagulopathy
Haemodilution
Activation of coagulation
Activation of fibrinogenoly-
sis and fibrinolysis
Reperfusion syndrome (hypo-
tension)
Hypothermia
Aortic aneurysm repair Aortic cross-clamp
Hepatic ischaemia/anoxia
Reperfusion syndrome
Sepsis DIC
with cardiopulmonary bypass, transplantation
surgery, orthopaedics and neurosurgery [3] (Table
1). The massive transfusion situation can be:
(1) Anticipated: Elective surgery with a high
probability of large volume blood loss, e.g. re-do
cardiac surgery and liver transplantation. Planning
by the surgeon, anaesthetist, haematologist and
blood service should ensure that appropriate
blood products are available; or,
(2) Unexpected: Massive haemorrhage occur-
ring unexpectedly in an elective surgical setting
requiring massive transfusion, and, where required
blood products may not be immediately available.
In both situations the principles underlying the
management are:
(a) the early recognition of massive blood loss,
(b) resuscitation to prevent shock and tissue hyp-
oxia, and,
(c) blood transfusion.
These issues will be discussed in more detail
with emphasis on the approach to transfusion
management.
2. Physiology and management of massive haemor-
rhage
A discussion of massive transfusion is not
complete without some mention of the physiology
and management of haemorrhage. Haemorrhagic
shock is characterised by a number of physiolo-
gical changes including:
(a) reduced cardiac output,
(b) hypoxic cell damage with metabolic effects,
(c) activation of the coagulation and fibrinolytic
cascades,
(d) activation of inflammatory mediators which
cause cellular damage,
(e) endothelial cell damage which activates intra-
vascular coagulation leading to a consumptive
coagulopathy.
The extent, severity and consequences of these
changes are proportional to the volume of blood
loss. The immediate resuscitation priorities in
massive haemorrhage are:
(1) Blood volume replacement: Fluid resusci-
tation with crystalloid or colloid is essential to
maintain intravascular volume. Seventy percent
loss of red cells together with a low haemoglobin
(Hb) can be tolerated only if intravascular volume
is maintained. The body’s ability to compensate is
limited to 30% loss of blood volume, after which
hypovolaemic shock occurs. Large volumes of
crystalloid or colloid fluids can impact on hae-
mostasis (e.g. haemodilution).
(2) Maintain tissue oxygenation: Tissue oxy-
genation requires adequate circulating haemoglo-
bin, the ability to increase cardiac output and
adequate oxygen delivery. In general, anaemia can
be tolerated as long as there is normovolaemia (see
 W.N. Erber / Transfusion and Apheresis Science 27 (2002) 83–92
Fig. 1. Transfusion therapy in massive haemorrhage.
above) and there is a compensatory increase in
cardiac output. Thirty percent loss of blood vol-
ume is the critical level at which red cells must
be replaced (Fig. 1).
(3) Achieve haemostasis: The highest priority
should be given to controlling haemorrhage. Sur-
gical bleeding must be curtailed and any coagul-
opathy corrected. Normovolaemia is required to
achieve haemostasis. These steps will minimise the
need for massive transfusion.
3. Transfusion management: general principles
Massive transfusion is a difficult clinical man-
agement problem requiring communication and
collaboration between the surgeon, anaesthetist,
transfusion laboratory, haematologist and blood
service [4]. In addition to the acute event, it re-
quires the long-term management tools provided
by review, guidance and audit functions of the
institutional transfusion committee. The aim of
transfusion management is to replace the deficit of
85
cells (red cells and platelets) and plasma proteins.
The timing and monitoring of blood product ad-
ministration is complex. There are no simple for-
mulae. Each case is a complex mixture of factors
and successful management requires input from
all participants because:
(a) The cause of the blood loss needs to be deter-
mined (e.g. surgical; haemostatic failure).
(b) There are no precise laboratory indices to as-
sist in blood product management.
(c) There is a need to anticipate changes in clinical
status and blood product requirements.
(d) The logistic issues in obtaining appropriate
blood products from the blood service.
These are usually best managed when the sur-
gical team is in open communication with a hae-
matologist.
Blood product replacement therapy should be
guided by the general principles of securing hae-
mostasis, maintaining intravascular volume and
oxygen delivery together with sound clinical
86 W.N. Erber / Transfusion and Apheresis Science 27 (2002) 83–92
judgement. Transfusion guidelines, including for-
mula-based approaches (e.g. fresh frozen plasma to
be given after a fixed number of packed red cells),
are not advocated as they do not improve the
outcome of massive transfusion [4–6]. Early and
aggressive blood component replacement is rec-
ommended as this can circumvent clinical problems
associated with massive blood loss and transfusion
(e.g. hypothermia; acidosis from hypoperfusion).
4. Red blood cells
Packed red blood cell concentrates (together
with crystalloid or colloid volume replacement) are
used virtually universally for red cell transfusion;
in most countries whole blood is no longer rou-
tinely available. The red cells must be ABO and
Rh compatible. In the elective surgical setting the
blood group will commonly be known from pre-
operative pre-transfusion testing. For some pro-
cedures blood may already be crossmatched or a
group and antibody screen performed; in these
situations ABO compatible or crossmatched blood
should be immediately available in the event
of either anticipated or unexpected haemorrhage.
Difficulty may arise for patients known to have
clinically significant red cell antibodies when un-
expected massive haemorrhage occurs.
If no pre-operative transfusion testing has been
performed and the patient’s blood group is not
known, uncrossmatched group O packed cells may
be required in the event of unexpected haemor-
rhage. By providing group O red cells there is no
delay and compatible blood is immediately avail-
able. If there is sufficient time to perform an ABO
and Rh blood group, uncrossmatched group-
specific red cells can be provided in an emergency.
Fully crossmatched compatible blood may take up
to 45 min, and even longer if circulating red cell
antibodies are detected. In the massive transfusion
situation, after 10 units packed cells have been
issued within a 24 h period, group specific blood
can be issued without the need for crossmatching.
What are the qualities of the red cells that
should be considered in the massive transfusion
setting? Red cells are transfused to optimise oxy-
gen delivery to tissues. In a massive transfusion it
is important that red cells with maximal function
are transfused to optimise oxygen delivery whilst
minimising any potential adverse effects. Red cell
storage results in depletion of 2,3-diphosphogly-
cerate (DPG) which increases red cell oxygen
affinity and reduces oxygen delivery, as well as
increasing red cell membrane rigidity with resul-
tant reduced red cell survival. These changes,
which can have adverse clinical sequelae in recip-
ients of massive transfusions, can be minimised if
red cells stored for less than two weeks are used.
Leucocyte-reduced packed red cells are also rec-
ommended; these units may cause less cytokine-
mediated organ damage, one of the major causes
of morbidity following massive transfusions.
Maintaining a high haematocrit (>0.34) has been
reported to reduce the bleeding time and non-
surgical blood loss. This approach may reduce the
need for transfusions of platelets and plasma [7].
When large volumes of cold blood are to be
infused, consideration should be given to the using
a blood warmer. This assists in preventing the
complications of hypothermia, namely vasocon-
striction, cardiac arrhythmias and coagulopathy.
The trade-off is that blood warmers slow the in-
fusion rate. Intraoperative cell salvage and rein-
fusion of washed autologous blood can be used as
an alternative to or to supplement homologous
banked blood in surgical settings where the shed
blood is not contaminated by bacteria or malig-
nant cells. The salvaged blood is immediately
available, compatible and ‘‘fresh’’, thus having
functional advantages over stored blood.
5. Platelets
Dilutional thrombocytopenia is common with
the massive transfusion of packed red cells and
manifests as microvascular bleeding, i.e. oozing,
mucosal bleeding, bleeding from cannula sites
or unexpected wound bleeding [8]. The rate of
development of thrombocytopenia is variable. In
general, the platelet count is rarely below
100  109 l 1 after 1.5 blood volume replacement
but falls to 50  109 l 1 after two blood volumes
[9]. The dilutional thrombocytopenia is less severe
than would be predicted by dilution alone since
 W.N. Erber / Transfusion and Apheresis Science 27 (2002) 83–92
Table 2
Critical haemostatic values in massive transfusion
Laboratory Therapeutic Haemostatic
measure goals failure
Fibrinogen >1.4 g/l <1.0 g/l
PT/APTT <1:5  normal >1:8  normal
Platelet count >100  109 l 1 <80  109 l 1
there is compensatory release of stored platelets
(e.g. from the spleen, where one-third of platelets
are sequestered, and premature release from the
bone marrow and lungs) [8,10]. Disseminated
intravascular coagulation (DIC), hypersplenism,
platelet consumption and pre-existing platelet dys-
function may worsen the thrombocytopenia.
Platelet replacement therapy should be guided
by repeated platelet counts. Platelet counts above
100  109 l 1 usually provide adequate haemosta-
sis. Haemostasis becomes difficult with platelet
counts below 80  109 l 1 , levels usually occurring
after 1.5–2 blood volumes have been replaced with
red cells (Table 2 and Fig. 1). Prophylactic platelet
transfusions are generally not indicated, as they do
not prevent diffuse microvascular bleeding or re-
duce overall blood product requirements [10,11].
Empirical platelet therapy may be indicated if
there is known platelet dysfunction (such as occurs
with cardiopulmonary bypass, or for patients with
known platelet dysfunction secondary to aspirin
therapy or renal dysfunction). Platelet transfusion
therapy will assist in correcting the platelet dys-
function due to prior drug therapy if the drug has
been eliminated from the circulation.
The dose of platelets should be determined by
the required increase in platelet count in the cir-
culation. One dose of random donor platelets from
a whole blood donation containing 5  1010 plate-
lets should increase the platelet count by 10  109
l 1 ; with one dose of apheresis platelets, contain-
ing 3  1011 platelets, the increase should be 60 
109 l 1 . In a massive haemorrhage there is rapid
platelet consumption and repeated platelet trans-
fusions may be required. Platelets should be as
fresh as possible (e.g. less than three days storage)
to maximise platelet quality and hence haemostatic
potential in vivo. The platelet storage lesion, which
affects the platelet membrane, granularity and cy-
tokine production, results in reduced function and
87
increased risk of adverse reactions. Leucocyte-re-
duced platelets are recommended to ensure that
only low levels of pro-inflammatory cytokines are
infused. Although a significant volume of plasma
(50 ml per random donor platelet unit and 300
ml per apheresis unit) accompanies platelet trans-
fusions, the majority of the coagulation factors in
this plasma have been inactivated by room tem-
perature storage.
6. Coagulation factors
The haemostatic defects in massive haemor-
rhage and transfusion are primarily due to dilu-
tional coagulopathy, together with the dilutional
thrombocytopenia, and DIC. Prolonged shock
and large volumes of crystalloid and colloid also
play a significant role in the coagulopathy [12].
DIC, a major complication of massive transfusion,
may be secondary to shock, hypothermia, tissue
thromboplastins or cellular damage.
The loss of one blood volume and replacement
with packed red cells, removes 70% of coagula-
tion factors and is not generally associated with
a bleeding diathesis. At this level, with only 30%
residual coagulation factors, the PT and APTT are
1.5 times the midpoint of the normal reference
range (Fig. 1) [13]; these levels provide sufficient
functional clotting factor activity. Haemostasis is
only compromised when clotting factors fall to
<30% (Table 3) [14–16]. With <20% factors, the
PT and APTT increase to 1.8 times normal, at
which level it is difficult to achieve haemostasis
[17,18]. Plasma replacement with fresh frozen
plasma and/or cryoprecipitate is therefore recom-
mended to maintain the PT and APTT at or below
1.5 times the control (Fig. 1).
Table 3
Blood volume lost, percent patient blood volume exchanged
and residual coagulation factors in massive transfusion
Blood volumes % Recipient % Coagulation
replaced by blood volume factors
transfusion exchanged
1 65–75 30
2 85–95 15
3 95–99 5
88 W.N. Erber / Transfusion and Apheresis Science 27 (2002) 83–92
The PT, APTT, fibrinogen and platelet count,
the most useful tests of haemostasis, can be per-
formed at all times in most laboratories and within
45 min of receipt of the sample in the laboratory.
Point-of-care tests are also available for PT, APTT
and fibrinogen. D-dimer tests can be used to con-
firm or exclude DIC but lacks specificity in surgi-
cal patients. Results of haemostasis tests should be
used as a guide to assist with plasma transfusion.
The correlation of coagulation test results with the
clinical situation is imperfect, in part because
the tests are performed at 37 C whilst the mas-
sively transfused patient is commonly hypother-
mic: normal laboratory coagulation results may be
seen in a hypothermic patient with a clinically
significant in vivo clotting defect [19].
Fibrinogen is the first of the coagulation factors
to fall in massive transfusion, the decrease in fi-
brinogen being directly proportional to the degree
of haemodilution [20]. After 1.5 blood volumes
have been replaced, the fibrinogen falls to the
critical level of 1.0 g/l; lower levels are usually
associated with bleeding. Prolongation of the
thrombin time (TT) only occurs with extremely
low fibrinogen levels, at non-haemostatic levels,
and hence is less valuable than a fibrinogen esti-
mation in massive haemorrhage and transfusion.
The TT can also be prolonged in cardiac surgical
patients receiving heparin; a reptilase time may
be required to determine the effect of heparin on
coagulation tests.
Thromboelastography (TEG) provides a global
assessment of haemostatic function and is used by
some centres to guide blood product replacement
[21]. The tracings, which can be generated within
30 min, assess the interaction of coagulation fac-
tors and platelet function, fibrinolytic activity and
clot stability and supplement routine coagulation
testing. TEG is particularly helpful in the presence
of heparin, with fibrinolysis and platelet dysfunc-
tion. TEG has been used in hepatic and cardiac
surgery and, to a lesser extent, in massive haem-
orrhage. As TEG does not quantify the relative
contributors to the bleeding process, it should not
be used alone in determining the aetiology of
bleeding. The other limitations of TEG are that it
is usually performed in the operating theatre by a
skilled operator, usually the anaesthetist, and this
may not be practical in the setting of massive
haemorrhage: the anaesthetist may have other
pressing problems!
The laboratory measures of haemostasis pro-
vide a guide only to clotting factor replacement
therapy. Clinical judgement is essential and plasma
replacement therapy should not be delayed or with-
held whilst awaiting results of laboratory tests. It is
important to maintain haemostasis rather than
treat a coagulopathy when it has fully developed.
7. Plasma transfusion
Plasma transfusions should be used to correct
specific coagulation abnormalities. Fresh frozen
plasma (FFP), which contains all coagulation
factors including fibrinogen (2–5 mg/ml), is most
commonly used as first-line haemostatic therapy
with massive transfusion of packed red cells. The
timing of FFP should be guided by results of PT,
APTT (threshold of 1:5 normal) and the fibri-
nogen level. The fibrinogen should be maintained
at >1.4 g/l to prevent haemostatic failure due to
hypofibrinogenaemia (Table 2). Because of the
need to thaw FFP there is a minimum delay of 30
min from request to availability, during which time
the clinical situation can change dramatically. On
the other hand, cryoprecipitate, which is of smaller
volume (10–20 ml), can be made available rapidly
and is an alternative source of fibrinogen as first-
line plasma replacement therapy.
The dose of FFP is usually based on patient
weight (i.e. one unit FFP per 10–20 kg body
weight) and should be sufficient to increase the
coagulation factor levels to above the critical value
(i.e. 30%) and to control blood loss. An aggressive
approach to FFP replacement is recommended
as this minimises the coagulopathy, assists in
controlling bleeding and should reduce the de-
mand for other blood products. Formula-based
dosing of FFP (e.g. one unit FFP per four units
packed cells transfused) and prophylactic plasma
therapy are not advocated; these approaches do
not reduce the overall bleeding or blood product
requirements [6,11,13,17].
Cryoprecipitate is a source of fibrinogen (150–
300 mg/unit), Factor VIII and von Willebrand
 W.N. Erber / Transfusion and Apheresis Science 27 (2002) 83–92
factor. A unit is of smaller volume (10–20 ml) than
FFP and therefore can be thawed and administered
rapidly if there is an urgent need for fibrinogen
replacement. Cryoprecipitate can be transfused in
the following settings:
(a) early in massive haemorrhage as a source of fi-
brinogen (i.e. management of dilutional hypo-
fibrinogenaemia);
(b) following FFP (if there is persistent hypofibri-
nogenaemia); or,
(c) when the fibrinogen level is disproportionately
low compared with other factors (e.g. as occurs
with fibrinogenolysis).
The dose of cryoprecipitate is generally 2 ml/kg
body weight and one unit should increase the
fibrinogen level by 0.1 g/l.
8. Persistent haemorrhage and microvascular bleed-
ing
Despite appropriate coagulation factor replace-
ment and platelet support, haemostatic failure
with persistent blood loss can still occur. This may
manifest as microvascular bleeding or persistent
haemorrhage from the surgical site. At this stage
the causes of haemostatic failure are multifactorial
and include:
(a) persistent surgical blood loss,
(b) delayed replacement of clotting factors and
platelets,
(c) persistent haemodilution causing a dilutional
coagulopathy and thrombocytopenia,
(d) consumption of coagulation factors and plate-
lets (i.e. DIC),
(e) reduced synthesis of coagulation factors,
(f) reduced bone marrow thrombopoiesis,
(g) the effects of pharmacological agents (e.g. an-
tiplatelet agents; anticoagulants),
(h) persistent hypothermia due to shock, loss
of thermal regulation and the administration
of cold fluids including packed red cells,
(i) underlying medical problems (e.g. renal or he-
patic dysfunction).
89
This is an extremely difficult management
problem and there are no simple formulae or rec-
ommendations for ongoing transfusion support.
At this stage management must be determined by
the underlying clinical status of the patient, the
type of surgery, the presence of DIC and known
drug exposure. Hypothermia may be a major
contributor with reduced hepatic synthesis of co-
agulation factors and impaired platelet function;
it is therefore important for normothermia to be
achieved. A number of approaches using other
blood products and pharmaceutical agents can
also be considered.
9. Other blood products
9.1. Autologous cell salvage blood
Salvaged autologous blood shed during surgery
may be washed and transfused. This provides a
source of fresh autologous red cells, thereby
maximising oxygen delivery, avoiding the conse-
quences of the red cell storage lesion and reducing
the demands on the blood service in providing
compatible homologous blood. This is particularly
useful in ‘‘clean’’ surgery with massive haemor-
rhage, such as cardiac and vascular surgery, where
the blood is not contaminated with bacteria or
malignant cells.
9.2. Prothrombin complex concentrate
Prothrombin complex concentrate (PCC) pro-
vides a source of Factors II, IX and X in a small
volume and can be administered rapidly. PCC may
be beneficial in patients with persistent bleeding
and known hepatic impairment or vitamin K de-
ficiency [22]. PCC should be administered to
achieve 50–100% levels of prothrombin complex
factors.
9.3. Recombinant factor VIIa
Recombinant factor VIIa (rVIIa) has been
shown to be effective in reducing blood loss in
patients with haemophilia with factor VIII inhib-
itors. There have been preliminary reports of the
90 W.N. Erber / Transfusion and Apheresis Science 27 (2002) 83–92
successful use of rVIIa in traumatic bleeding
[23,24], intraabdominal bleeding with DIC [25]
and reducing blood loss in liver transplantation
[26,27]. These reports suggest that rVIIa may also
have a role in massive transfusion which is often
accompanied by DIC. Recombinant VIIa is ex-
pensive and further clinical studies are required to
confirm its value and define its role in massive
transfusion.
9.4. Fresh whole blood
Fresh (<24 h old) unrefrigerated whole blood
contains fresh red cells, functional coagulation
factors and platelets and has appropriate colloid
oncotic pressure. It should therefore be considered
in the setting of uncontrollable haemorrhage fol-
lowing 1.5–2 blood volumes have been replaced
with blood products. It has the advantage of
having functional oxygen-carrying red cells, can
provide rapid blood volume expansion simulta-
neously with intact haemostatic agents and does
not compound the effects of hypothermia. A major
disadvantage of fresh whole blood is that it is
difficult to obtain from the blood service. This is
because virtually all whole blood donations are
centrifuged, separated into components and re-
frigerated soon after collection. In addition the
need to perform screening tests for potential in-
fectious agents can delay the release of blood
products by 24–48 h. In an emergency, alternative
sources of whole blood may be required (e.g.
emergency blood donor panels) and there may be
insufficient time for full infectious agent screening
prior to transfusion. This is a controversial issue
and has precluded the use of fresh whole blood
in many centres.
Emergency donor panels are available in remote
centres in some countries (e.g. regional Australia).
These panels are made up of volunteer donors in
the local community who are willing to donate
whole blood in a life-threatening emergency. The
donors are ABO-grouped and are screened for
infectious agents on a regular basis (at least six-
monthly) by the blood service. The panel can be
activated in an emergency (e.g. massive trans-
fusion) and blood collected from donors of the
appropriate blood group. This blood remains un-
refrigerated and is transfused immediately as fresh
whole blood. Infectious agent testing is performed
retrospectively. This system has limitations and
hence fresh whole blood is transfused rarely.
Despite the limitations of access, there are an-
ecdotal reports of the success of fresh unrefriger-
ated whole blood in life-threatening situations of
uncontrollable haemorrhage with seemingly ade-
quate replacement with plasma and platelets [28].
Since there have been no randomised trials to
confirm the efficacy, define the dose or demon-
strate safety of this approach, and, since it appears
unlikely that such trials could be performed, this
approach will remain controversial. Assessment of
risks and benefits in an individual case is required.
The use of unrefrigerated fresh whole blood may
seem justified to the responsible clinicians in an
emergency where there is uncontrolled bleeding
following appropriate large volume blood com-
ponent therapy, or if there is insufficient compo-
nent therapy available, and death seems inevitable
otherwise.
9.5. Fibrin glue
Fibrin glue is a human-derived tissue adhesive
that can be applied topically to bleeding raw sur-
faces when there is uncontrolled local bleeding.
This is prepared from fibrinogen in cryoprecipitate
or autologous plasma.
10. Pharmacological agents
Procoagulant pharmacological agents should be
considered in massive transfusion to:
(a) reverse anticoagulant drugs that may be inter-
fering with coagulation (e.g. heparin);
(b) inhibit fibrinolysis; and,
(c) increase the production of coagulation factors.
Protamine sulphate can be used to reverse the
effects of heparin used in cardiac surgery utilising
cardiopulmonary bypass. Aprotinin, a protease in-
hibitor and potent antiplasmin, has been reported
to reduce blood loss in massive haemorrhage [29].
The mechanism of action is multifactorial with
 W.N. Erber / Transfusion and Apheresis Science 27 (2002) 83–92
effects on coagulation, fibrinolysis and platelet
function. Desmopressin causes a dose-dependent
increase in factor VIII and von Willebrand factor.
It has been used in massive surgical haemorrhage
(e.g. cardiac; prostatic) with some reports of suc-
cess [30]; other studies have shown no benefit
of desmopressin in massive transfusion. Oxygen-
carrying blood substitutes (e.g. perfluorocarbons;
recombinant haemoglobin solutions) are currently
in clinical trial. These agents may have a role in
massive transfusion.
11. Consequences of massive transfusion
A number of clinical and biochemical distur-
bances result from the transfusion of large vol-
umes of blood products. These are, in part, due to
the storage lesion of blood, the effects of the citrate
anticoagulant and hypothermia. Metabolic chan-
ges include:
(a) hyperkalaemia due to the high extracellular
potassium concentration in units of stored
red cells;
(b) hypocalcaemia secondary to the citrate antico-
agulant in stored blood products binding ion-
ised calcium;
(c) acidosis secondary to the acidic pH of stored
red cells (i.e. lactate accumulation; citric acid)
and a metabolic acidosis associated with hypo-
volaemia;
(d) hypernatraemia; and,
(e) a shift in the oxygen dissociation curve, due to
reduced 2,3-DPG in stored red cells (especially
with less than two weeks storage), resulting in
increased oxygen affinity and impaired oxygen
delivery [14].
A coagulopathy commonly occurs secondary to
massive transfusion and has multiple aetiologies.
These include the large volume blood product
administration, prolonged shock, DIC and hy-
pothermia. Multisystem organ failure (e.g. neuro-
logical, cardiac, respiratory and hepatic) may occur
secondary to hypoxia, DIC, the red cell storage
lesion, direct cytokine damage or microaggregates
in transfused blood products.
91
Massive transfusion has a poor overall survival.
Reports from large series range from 45–67%
mortality rates [11,31–33]. Patient age, duration
and magnitude of shock, the development of DIC
and the amount of blood transfused influence the
outcome. Higher mortality rates occur in patients
with pre-existing medical problems.
12. Conclusion
Massive transfusion is a complicated clinical
management problem. Control of haemorrhage by
securing haemostasis is critical because ongoing
blood loss and transfusion result in a number
of adverse clinical sequelae. The successful out-
come depends on understanding the physiologi-
cal changes that occur in massive haemorrhage,
particularly in haemostasis, and appropriate
blood product replacement to ensure platelets
and coagulation factors are maintained at hae-
mostatic levels. Blood component replacement
therapy should be guided by the results of labo-
ratory tests but informed clinical judgement is es-
sential. The differences between stored blood and
shed blood result in metabolic problems in massive
transfusion. Logistic issues such as the availability
of blood products and the supply of quality
products are critical to the management of these
patients.
Acknowledgements
I would like to thank Dianne Grey, Gary
Hoffman and Mark Schneider for their construc-
tive review of the manuscript. Sincere thanks to
Emma Robotham for her secretarial expertise in
the preparation of the manuscript.
References
[1] Peterson A. Massive transfusion. Int Anesthesiol Clin
1987;25:61–74.
[2] H€ocker P, Hartmann T. Management of massive transfu-
sion. Acta Anaesthesiol Scand 1997;111:205–7.
[3] Rutledge R, Sheldon GF, Collins ML. Massive transfu-
sion. Crit Care Clin 1986;2:791–805.
92 W.N. Erber / Transfusion and Apheresis Science 27 (2002) 83–92
[4] Stainsby D, MacLennan S, Hamilton PJ. Management of
massive blood loss a template guideline. Br J Anaesthes
2000;85:487–91.
[5] Blauhut B, Lundsgaard HPA. Blutverlust und Verbrenn-
ungen in der operativen Medizin. In: Transfusionmedizin.
C. Mueller-Eckhardt (Hrsg). Auflage-Verlag 1996;291–
320.
[6] Mannucci PM, Frederici AB, Sirchia G. Haemostasis
testing during massive blood replacement. A study of 172
cases. Vox Sanguinis 1982;42:113–23.
[7] Caleri CR, Cassidy G, Pivacek LE, Ragno G, Lieberthal
W, Crowley JP, et al. Anemia-induced increase in the
bleeding time: implication for treatment of nonsurgical
blood loss. Transfus Prac 2001;41:977–83.
[8] Reed RL, Ciavarella D, Heimbach DM, Baron L, Pavlin E,
Counts RB, et al. Prophylactic platelet administration
during massive transfusion: a prospective randomized
double-blind clinical study. Ann Surg 1986;203:40–8.
[9] Counts RB, Haisch C, Simon TL, Maxwell NG, Heimbach
DM, Carrico CJ. Haemostasis in massively transfused
trauma patients. Ann Surg 1979;190:91–9.
[10] Harvey MP, Greenfield TP, Sugrue ME, Rosenfeld D.
Massive blood transfusion in a tertiary referral hospital:
clinical outcomes and haemostatic complications. Med J
Austr 1995;163:356–9.
[11] Parker RI. Etiology and treatment of acquired coagulop-
athies in the critically ill adult and child. Crit Care Clin
1997;13:591–609.
[12] Hewson JR, Neame PB, Kumar N. Coagulopathy related
to dilution and hypotension during massive transfusion.
Crit Care Med 1985;13:387–91.
[13] Reiss RF. Haemostatic defects in massive transfusion:
rapid diagnosis and management. Am J Crit Care
2000;3:158–65.
[14] Marsaglia G, Thomas ED. Mathematical consideration of
cross circulation and exchange transfusion. Transfusion
1971;11:216–9.
[15] Lundsgaard-Hansen P. Treatment of acute blood loss. Vox
Sanguinis 1992;63:241–6.
[16] Leslie S, Toy P. Laboratory haemostatic abnormalities in
massively transfused patients given red blood cells and
crystalloid. Transfus Med 1991;96:770–3.
[17] Ciavarella D, Reed RL, Counts RB, Baron L, Pavlin E,
Heimbach DM, et al. Clotting factor levels and the risk of
diffuse microvascular bleeding in the massively transfused
patient. Br J Haematol 1987;67:365–8.
[18] Hiippala S. Replacement of massive blood loss. Vox
Sanguinis 1998;74(Suppl 2):399–407.
[19] Rohrer MJ, Natale AM. Effect of hypothermia on the
coagulation cascade. Crit Care Med 1992;20:1402–5.
[20] Hiippala ST, Myllya GJ, Vahtera EM. Hemostatic factors
and replacement of major blood loss with plasma-poor red
cell concentrates. Anesthes Analges 1995;81:360–5.
[21] Mallett SV, Cox DJA. Thromboelastography. Br J Anaes-
thes 1992;69:307–13.
[22] Blauhut B. Indications for prothrombin complex concen-
trates in massive transfusions. Thrombosis Res 1999;95:
S63–9.
[23] Kenet G, Walden R, Eldad A, Martinowitz U. Treatment
of traumatic bleeding with recombinant factor VIIa.
Lancet 1999;354:1879.
[24] Al Douri M, Shafi T, Al Khudairi D, Al Bokhari E, Black
L, Akinwale N, et al. Effect of the administration of
recombinant activated factor VII (rFVIIa; NovoSeven) in
the management of severe uncontrolled bleeding in patients
undergoing heart valve replacement surgery. Blood Coag
Fibrinol 2000;11(Suppl 1):S121–7.
[25] Moscardo F, Perez F, de la Rubia J, Balerdi B, Lorenzo JI,
Senent ML, et al. Successful treatment of severe intra-
abdominal bleeding associated with disseminated intravas-
cular coagulation using recombinant activated factor VII.
Br J Haematol 2001;114:174–6.
[26] de Wolf JT, Hendriks HG, Meijer K, Klompmaker IJ,
Hagenaars AA, Melsen T, et al. Impressive reduction of
blood requirements in orthotopic liver transplantation due
to recombinant factor VIIA (RFVIIA, NovoSeven).
Transfusion 1999;39(S):87S.
[27] Hendriks HG, Meijer K, de Wolf JT, Klompmaker IJ,
Porte RJ, de Kam PJ, et al. Reduced transfusion require-
ments by recombinant factor VIIa in orthotopic liver
transplantation: a pilot study. Transplantation
2001;71:402–5.
[28] Erber WN, Tan J, Grey D, Lown JAG. Use of unrefrig-
erated fresh whole blood in massive transfusion. Med J
Austr 1996;165:11–3.
[29] Laupacis A, Fergusson D. Drugs to minimise perioperative
blood loss in cardiac surgery: meta-analyses using periop-
erative blood transfusion as the outcome. Anaesthes
Analges 1997;85:1258–67.
[30] Lethagen S. Desmopressin––A haemostatic drug: state of
the art review. Eur J Anaesthes 1997;14(Suppl):1–9.
[31] Miller RD, Robbins TO, Tng MJ, Barton SL. Coagulation
defects associated with massive transfusions. Ann Surg
1971;174:794–801.
[32] Sawyer PR, Harrison CR. Massive transfusion in adults.
Diagnosis, survival and blood bank support. Vox Sangu-
inis 1990;58:199–203.
[33] Wudel JH, Morris JJ, Yates K, Wilson A, Bass SM.
Massive transfusion: outcome in blunt trauma patients.
J Trauma 1991;31:1–7.