The Journal of TRAUMA威 Injury, Infection, and Critical Care

Indications for Early Fresh Frozen Plasma, Cryoprecipitate,

and Platelet Transfusion in Trauma
Lloyd Ketchum, MD, John R. Hess, MD, MPH and Seppo Hiippala, MD

Background: Massive blood transfu-
sion can be lifesaving in the treatment of
severe trauma. Guidelines for the use of
non-RBC blood components in the early
phase of trauma resuscitation are largely
based on extensions of expert recommen-
dations for general surgery.
Methods: The logic and evidence for
the use of plasma, platelets, and cryopre-
cipitate early in the course of massive
transfusion for trauma were reviewed.
Large series of consecutive patients were
sought.
Findings: Resuscitation of the most se-
verely injured and massively hemorrhaging
patients usually starts with crystalloid fluids
and progresses to uncross-matched RBC.
Low blood volume, insensible losses, con-
sumption, and resuscitation with plasma
poor RBC concentrates rapidly lead to
plasma coagulation factor concentrations of
less than 40%. This typically occurs before
10 U of RBC have been transfused. Early
initiation of plasma therapy is often delayed
by its lack of immediate availability in the
trauma center. Platelets usually fall to con-
centrations of 50 –100ⴛ 109̂/L after 10 –20
units of RBC have been given, but platelet
concentrations in individual patients are
quite variable and can decrease more
quickly. Ideal platelet concentrations in
trauma patients are not known, but are gen-
erally held to be greater than 50ⴛ 109/L.
Cryoprecipitate can rapidly increase the
concentrations of fibrinogen and von Wille-
brand’s factor, but the advantages of higher
than normal concentrations are speculative.
Conclusions: Early use of plasma
and platelets at the upper end of recom-
mended doses appears to reduce the incidence
of coagulopathy in massively transfused
individuals.
Key Words: Massive transfusion, Co-
agulopathy, Blood components.
J Trauma. 2006;60:S51–S58.

A
t the dawn of the age of blood component therapy, John
Collins wrote: “Volume, red cells, albumin . . . clotting
factors and platelets are needed in that order as the
volume of blood loss increases.”1 In a subsequent book,2
Collins also provided the following quantitative guidelines.
Volume
Young healthy patients can tolerate up to 30% loss of
blood volume but after that need significant volume replace-
ment to avoid severe shock.
Red Blood Cells (RBC)
At 50% loss, many patients will need RBC.
Albumin
After 150% loss, with continuous replacement of volume
with crystalloid and RBC, albumin will typically fall below
Submitted for publication November 18, 2005.
Accepted for publication November 28, 2005.
Copyright © 2006 by Lippincott Williams & Wilkins, Inc.
From the Walter Reed Army Institute of Research, Silver Spring,
Maryland, USA (L.K.); University of Maryland School of Medicine, Baltimore,
MD, USA (J.H.), and the University of Helsinki, Helsinki, Finland (S.H.).
The opinions or assertions contained herein are the private views of the
authors and are not to be construed as official or reflecting the views of the
Department of Defense or the United States Government. One of the authors
is an employee of the U.S. Government. This work was prepared as part of
his official duties and, as such, there is not copyright to be transferred.
Address for reprints: Lloyd H. Ketchum, MD, Chief, Department of
Blood Research, Walter Reed Army Institute of Research, 503 Robert Grant
Avenue—MCR, Silver Spring, MD 20910; email: Lloyd.Ketchum@na.
amedd.army.mil.
DOI: 10.1097/01.ta.0000199432.88847.0c
18 g/L, with significant loss of colloid oncotic activity and
the onset of edema, requiring colloid replacement.
Coagulation Factors
At 180% loss, with replacement as above, coagulation
factors typically fall below 20% of their normal concentra-
tions and the plasma coagulation times will be prolonged
more than 150% of normal.
Platelets
When more than 220% of blood volume has been re-
placed, platelet counts will usually have fallen to around
50,000/␮L; and frank coagulopathy will ensue.
These observations, distilled from calculation, animal
experiment and clinical practice, are still viewed as accurate.
However, their utility has been tempered by the recognition
that individual variability in response to injury is large.
Bleeding patients who are also frail, anemic or thrombocy-
topenic, or have underlying cardiac, lung, liver or kidney
disease, arrive at critical resuscitation thresholds sooner than
their healthy counterparts. Conversely, the young and athletic
may tolerate greater losses with fewer symptoms or signs.
Massively bleeding patients with reduced blood volumes,
lose a proportionately greater fraction of their remaining
blood with any given volume of loss.
A clinical approach to the problems of administering
plasma, platelets, and cryoprecipitate for the control of bleed-
ing needs to take into account four facts. First, 95% of all
injured patients do not need any of these components.3 Sec-
ond, a few patients, perhaps 1 in 20 of the remaining 5%, are
coagulopathic at the onset of resuscitation, usually as a result
of drugs or acquired diseases, and need immediate treatment

Volume 60 • Number 6 S51

 The Journal of TRAUMA威 Injury, Infection, and Critical Care
of coagulopathy as well as volume and oxygen carrying
capacity. Third, as resuscitation progresses to massive trans-
fusion, and plasma and platelet therapy are added to control
decreasing concentrations of these blood constituents, signif-
icant hemodilution is inevitable because of the composition
of banked blood components.4 Fourth, massively bleeding
patients have low blood volume and ongoing vigorous resus-
citation with crystalloids and colloids. The combination of
these factors means that data directly applying to every pos-
sible combination of circumstances is not available. Never-
theless, it is an excellent general rule that a determined effort
is needed to deliver sufficient amounts of all of the compo-
nents in a timely fashion to maintain adequate coagulation.
Hemorrhage is the second leading cause of death in
combat casualties.5 Severely injured soldiers generally have
the advantages of youth and physical fitness, but with the
disadvantage of high-energy penetrating weapons injury. The
hemostatic system, composed of platelets and coagulation
proteins, faces two distinct clinical challenges in dealing with
their injuries. The most obvious is to stop or limit blood loss
from the initial site of injury. The second occurs in the small
subset of patients requiring massive transfusion many of
whom will go on to develop a life-threatening trauma asso-
ciated coagulopathy in which hemostasis is reduced diffusely
and spontaneous bleeding occurs at uninjured or minimally
injured sites. Platelets and clotting factors must be present in
sufficient quantity and have adequate function to perform
their critical role of stopping bleeding.
Evolution of Coagulopathy in Severe Traumatic
Blood Loss
Following trauma accompanied by massive hemorrhage,
coagulation factors and platelets are lost and consumed, and
their activity is reduced by hypothermia, acidosis, and dilu-
tion. In addition, clots that are formed may be broken down
inappropriately by physical manipulation of wounds and fi-
brinolysis. The processes that lead to coagulopathy are dy-
namic and ongoing.
Most massively bleeding trauma patients have high blunt
trauma injury severity scores.6,7 They have often lost consid-
erable amounts of blood into internal spaces, either at the
scene of injury or during transport, without obvious signs of
hemorrhage. Depending on the age and physical capacity of
the injured person, some degree of volume homeostasis can
be achieved by tissue fluid shifts, and a significant amount of
plasma may be replaced with capillary refill. Of necessity
however, this autologous replacement is largely acellular and
contains little protein, beginning the process of dilution of the
components of coagulation. Dilution is then rapidly enhanced
by crystalloid resuscitation often started outside the hospital
and continued, even more vigorously, at the emergency de-
partment. As volume replacement is achieved, blood pressure
rises, and the patient bleeds more profusely. This additional
blood loss requires more volume to maintain acceptable
blood pressure, and so on. The sequence of recurring hypo-
S52
volemia, followed by therapeutic interventions that in turn
lead to further blood loss, leads to a rapid dilution of plasma,
all while the patient is still in the ED undergoing primary
assessment and is still without any definitive treatment plan.
This unpredictable and often poorly controlled sequence
of events in already hypovolemic patients leads to a deficient
coagulation profile much sooner than is usually anticipated.
In addition, the extensive exposure of subendothelial matrix
of the injured tissues from millions of endothelial microtears
generates excess, and usually unappreciated, consumption of
coagulation factors and platelets. Reduced clearance of inac-
tivated coagulation factors further interferes with the assem-
bly of active factor complexes. The emergence of coagulation
disturbances can be further accelerated by some solutions
used for volume replacement. Hypertonic saline preparations
can affect platelet function, and hydroxyethyl starch and
dextran solutions impair platelet adhesion and aggregation by
decreasing the von Willebrand factor activity in plasma.8,9
Monitoring of Coagulation in Trauma
The coagulation profile, prothrombin time (PT), platelet
count and plasma fibrinogen concentration, should be deter-
mined as soon as the blood volume has been restored to a
more acceptable level. Unfortunately, standard laboratory
testing processes are often too slow to provide results in a
timely fashion. The use of small portable point-of-care de-
vices can prove helpful in this situation, but calibration is a
problem. Several devices of this type measure PT, the most
sensitive and practical coagulation test to detect imminent
coagulation disturbances. Most devices now available also
provide the results as a percentage of the activity of a refer-
ence plasma, a more readily understood measure of dilution
than the PT in seconds. A PT activity remaining above
30 – 40% generally ensures normal coagulation with a safe
margin.10
The hemoglobin concentrations obtained before transfu-
sion of red cells can also be used, with some reservations, as
a surrogate to estimate the degree of plasma dilution and
coagulation factor deficiency. Assuming the patient to have a
normal hemoglobin before trauma, a value of 40 –50 g/L after
volume resuscitation, but before RBC replacement, strongly
suggests that the patient has already lost at least two thirds of the
original blood elements, including plasma and coagulation fac-
tors, and therefore should be given fresh frozen plasma.
The decline of platelet count is a highly individual phe-
nomenon. Some patients are able to maintain high platelet
counts despite ongoing blood loss by recruiting platelets from
spleen and possibly mobilizing new ones from bone marrow.
Most patients, however, approach the critical platelet count of
50,000/␮L after losing two blood volumes.11 Because of the
high variability, platelet count should be determined early on
after admission and repeated frequently to assess the individ-
ual course of the evolving thrombocytopenia.
By the time the clinical manifestations of severe coagu-
lopathy become apparent, as pathologic bleeding from skin
Supplement 2006

lacerations, cannulation sites or mucous membranes, the PT
activity will be well below the safety margin of 30 to 40% of
normal. These are late signs denoting significant coagulopa-
thy and usually demand exceptional interventions to regain
control.
Replacement Therapy
FFP Supplementation
Staying ahead of the coagulopathy of massive bleeding
in severe trauma requires anticipation. As noted, relying on
washout equations to guide replacement almost assures un-
derestimation of losses in this situation. The basic assump-
tions of the washout equation are simply not met in the care
of bleeding trauma patients.10 If they were, the hypothetical
best-case scenario, the patient would still retain more than
one third of plasma with all coagulation factors after losing
more than one blood volume. Obviously this is not the case in
the massively bleeding trauma patient. The best-case scenerio
does not take into account the insensible losses of whole
blood into tissue compartments, the greater proportional
losses of clotting factors and platelets because of low blood
volume, the consumption of clotting factors at sites of injury,
the inhibitory effects of colloid resuscitation fluids and inac-
tive coagulation factors, and the loss of clotting activity to
hypothermia and acidosis. Every violation of the best-case
scenario demands earlier supplementation of fresh frozen
plasma to avoid coagulopathy.
Most bleeding trauma patients will need fresh frozen
plasma well before losing one blood volume. This is different
from the recommendations for more controlled circumstances
of blood loss. In controlled elective surgery situations, the
loss of one and a half blood volumes justifies, in most cases,
the transfusion of FFP. One can make the argument that the
“early” start of FFP transfusion in trauma, at or before one
volume loss, is not far away from the “legitimate” starting
point. Starting “early” normally means that an adult trauma
patient will get four to six more FFP units than an equivalent
patient experiencing blood loss in an elective surgery. How-
ever, considering the difficulty in estimating true blood loss
in traumatic injury, there is probably very little actual differ-
ence in the proportions of blood loss to FFP units between
these two groups of patients. If clinically evident coagulopa-
thy is prevented by the early use of FFP, subsequent blood
product consumption is likely to be less.
The above recommendation, for early FFP supplementa-
tion in the course of resuscitation for massive bleeding in
trauma patients, is based on computer simulation models
taking into account the interaction of circulatory changes,
blood loss and the effects of fluid resuscitation. In one such
model, two units of FFP had to be given for each 3 units of
red blood cells to avoid excessive plasma dilution compro-
mising hemostasis, well before the patient had lost one blood
volume.10 This ratio is significantly higher than advocated by
the majority of transfusion protocols, and the model has been
blamed for overestimating the need of FFP. However, in the
Volume 60 • Number 6
Early FFP, Cryo, and Platelets
treatment of true massive bleeding requiring pressurized in-
fusions, clinical experience suggests that it is exceedingly
difficult to over transfuse such patients with any of the blood
components available, possibly excluding specific small vol-
ume preparations such as cryoprecipitate or fibrinogen. Fur-
thermore, it is impractical and often impossible to resort
exclusively to blood products, because lines are primed and
kept open with saline and delivery of blood is sometimes
delayed. Therefore the hypotensive episodes are frequently
treated with crystalloids and colloids, which rapidly dilute
plasma, red blood cells and platelets.
The use of plasma carries certain caveats. Fresh frozen
plasma contains most of the citrate anticoagulant added to the
collection bag and is therefore approximately 20% diluted
compared with normal plasma. As the blood volume drawn is
constant, the degree of dilution mainly depends on the he-
matocrit of the donor. A unit of FFP contains approximately
0.5 g of fibrinogen and all the pro- and anti-coagulant pro-
teins within normal range.
In Europe, solvent-detergent treated plasma (SD-plasma,
not available in the US) is widely used instead of normal FFP.
The average volume of normal FFP unit is approximately 250
mL while the volume of an SD-plasma unit is just 200 mL. In
addition, up to 20% of the coagulation factor activity of
SD-plasma is lost in the processing.12 In terms of total coag-
ulation factor content, one unit of SD-plasma equals just two
thirds of normal FFP unit. Therefore, if SD-plasma is used for
supplementation in massive blood loss, the ratio should be
one unit for each unit of red cells.
Platelets
A declining platelet count usually requires intervention
much later than the deficit of plasma. Although we can
determine the platelet count easily, we have no practical
methods to rapidly assess the function of native platelets. Nor
do we have practical ways to assess and assure the viability
and activity of transfused platelets. Platelet function assays
exist but are too cumbersome to be used in the emergency or
operation room. Furthermore, these functional assays are not
validated for low platelet counts and therefore usually have
no use in hemorrhagic thrombocytopenia. The possible excep-
tion, thromboelastography (TEG), is slow, suffers in accu-
racy, and is expensive, but can occasionally suggest problems
not evident from the PT and platelet count.
We do know that the recovery rate of five-day old plate-
lets after transfusion is about 50% and that nonviable platelets
are sequestered into the spleen.13 However, we have no data
on how fast these useless platelets are removed from the
circulation in unstable clinical settings like massive trauma.
Platelet counts taken shortly after platelet transfusion may be
giving us a falsely optimistic perception of the efficacy of
platelet replacement if a significant proportion of those plate-
lets included in such counts are nonviable and in fact fated for
eventual clearance by the spleen.
S53
 The Journal of TRAUMA威 Injury, Infection, and Critical Care
Nor do we know if the generally accepted threshold of
50,000/␮L for platelet transfusion is correct and applicable to
all types of trauma. In one small, retrospective study, the lowest
platelet count was around 40,000/␮L in both the coagulopathic
and non-coagulopathic groups. However the temporary rela-
tionship of the lowest platelet count, transfusions and possi-
ble coagulopathy were not disclosed.14 In blunt trauma the
consumption of coagulation factors and platelets is often
accelerated and may require interventions sooner compared
with penetrating injuries.
In contrast to the above, the effects of hypothermia on
platelet function are well known, and contribute significantly
to the coagulopathy of trauma. Mild hypothermia, core tem-
perature above 33°C, appears to affect platelet adhesion more
than the coagulation enzymes. At core temperatures under
33°C, both platelet function and coagulation enzyme activi-
ties are reduced.15 The effects of hypothermia should be
taken into account when estimating platelet function. Active
measures to warm trauma patients at risk for coagulopathy
may improve hemostasis more than platelet transfusion and
have become standard-of-care in advanced trauma care.
Massive blood loss requires rapid decisions. The abso-
lutely right timing of a platelet transfusion is rarely the issue
of utmost importance in the minds of the attending team.
Considering the long delivery time typical for platelet orders
from the blood bank and the rapidity with which massive
trauma situations evolve in the ED, the early placement of
platelet orders seems wise. There is, further, at least some
evidence suggesting that early platelet transfusion may im-
prove outcome in trauma.
Cosgriff and his colleagues reported on a cohort of 58
massively transfused (⬎10 units red blood cells) non-head
injury trauma patients to determine risk factors for develop-
ing life-threatening coagulopathy.16 Life-threatening coagu-
lopathy was defined as both a prothrombin time and a partial
thromboplastin time of greater than two times normal controls.
Based on these criteria 27 (47%) of the patient developed
life-threatening coagulopathy. Univariant analysis identified
Injury Severity Score, lowest systolic blood pressure in the
first 24 hours, pH and temperature as predictive of develop-
ing life-threatening coagulopathy. Interestingly the volume of
PRBC in 6 hours and 24 hours were not predictive. Of the 27
patients who developed life-threatening coagulopathy 15 pa-
tients (56%) survived. A comparison of blood product usage
in those who survived versus died demonstrated platelets/
PRBC unit transfusion ratio 0.79 versus 0.48 ( p ⫽ 0.01).
Cinat and his colleagues reported on a cohort of 45
massively transfused (⬎50 units of red blood cells) divided
into 2 groups for comparison (early [1988 –1992] and late
[1993–1997]).17 Survival following massive transfusion in-
creased over this 10 year time period (16% versus 45%, early
versus late period, p ⫽ 0.03). Factors associated with poor
outcome included male sex, major vascular injury, high In-
jury Severity Score, severe acidosis, prolonged hypotension,
refractory hypothermia and decreased use of platelet transfu-
S54
sion. A comparison blood product usage in survivors versus
nonsurvivors demonstrated a RBC/FFP ratio of 1.8 versus 2.5
( p ⫽ 0.06) and RBC-platelet transfusion ratio of 7.7 versus
11.9 ( p ⫽ 0.03) which if we assume the average platelet
transfusion contained approximately 6 units; translates to
platelet-RBC unit transfusion ratio of 0.78 versus 0.50 which
is remarkably similar to the ratio seen in the Cosgriff cohort.
While there are assuredly various changes in patient manage-
ment between these time periods the decrease in the RBC-
platelet ratio (12.7 versus 8.0, early versus late period, p ⫽
0.01) points to a potentially important role for aggressive
platelet administration in massive transfusion.
While a randomized controlled trial comparing prophy-
lactic platelet transfusion to FFP failed to show a benefit of
platelet transfusion, after a variety of exclusions there were
33 patients evaluated in this study and only 6 of them devel-
oped abnormal microvascular bleeding.18 In this study 6 units
of platelet were administered after 12 units of blood, giving a
platelet to PRBC ratio of 0.5, which is the platelet transfusion
ratio of the non-survivors in the cohort studies. Given the
small number of patients with microvascular bleeding, this
study does not exclude the possibility that prophylactic plate-
let administration may confer a large benefit in patients at
significant risk of developing microvascular bleeding.
In summary there are two publications of patient cohorts
requiring massive transfusion in which there appears to be a
survival advantage of receiving approximately 0.8 units of
platelets per unit of RBC. In civilian clinical practice this
could be accomplished by administering a standard 6 unit
pheresis pack of platelet for every 7.5 units of RBCs. In the
military setting where platelet availability is a tremendous
challenge the only alternative currently available is the ad-
ministration of fresh whole blood. An ongoing evaluation of
massive transfusion in Iraq by Perkins and colleagues during
the current conflict includes distinct patient cohorts who
received either fresh whole blood or pheresis platelets may
provide additional insight into this question.
Cryoprecipitate
Cryoprecipitate was developed and used for the treatment
of hemophilia A and von Willebrand disease before it was
replaced in the pharmacopoeia by specific preparations. Cryo-
precipitate contains abundant fibrinogen, von Willebrand-factor/
VIII complex, and fibrin stabilizing factor/XIII and has been
included in all transfusion protocols since the whole blood
era.19 Initially, the main indication for cryoprecipitate was the
restoration of plasma labile factor-VIII, as deficit of this
factor was thought to be one of the vital defects caused by
massive whole blood transfusion. Later, both the need for
supplementation of fibrinogen and the usefulness of cryopre-
cipitate as a source for this factor were recognized as well. In
addition, the boost of von Willebrand factor may enhance
adhesion and aggregation of platelets. The effect of surplus
factor XIII remains disputable.
Supplement 2006

Presently, the strategy to preserve adequate coagulation
is strongly based on fresh frozen plasma and platelets. The
need of prophylactic cryoprecipitate should be reevaluated.
Obviously, the crucial role and small body stores of plasma
fibrinogen, and its early depletion in massive trauma, requires
an efficient plan for replacement. However, the administra-
tion of cryoprecipitate by rule of thumb as supplement to FFP
may not be necessary for all trauma patients exposed to
massive transfusion. One unit of FFP contains approximately
0.5 g of fibrinogen and all other pro- and anticoagulant
proteins in balanced proportions. A unit of cryoprecipitate
contains 0.25 g of fibrinogen, but in 4% of the volume of a
unit of FFP, 10 mL compared with 250 mL respectively. Ten
units will contain 2.5 g in 100 mL of plasma. However,
provided the replacement volume is sufficient, FFP should be
able to ensure hemostatic concentrations of all coagulation
factors, including fibrinogen. Therefore, although prophylac-
tic cryoprecipitate supplementation is probably not harmful,
it may not be routinely necessary. The benefits of higher than
normal fibrinogen concentrations or higher than normal von
Willebrand’s factor in trauma patients are unknown. In Eu-
rope cryoprecipitate is no longer universally available and has
been widely replaced by specific plasma preparations, such as
fibrinogen, which are used to correct suspected or confirmed
deficiencies.
Practical Considerations
Trauma centers keep warm fluids readily available and
frequently have small numbers of units of uncrossmatched
group O RBC as well. Units of FFP take 20 –30 minutes to
thaw (each one is a 250 g block of ice in a plastic bag at –30
to ⫺80°C), so transfusion services supporting trauma centers
often keep thawed plasma (which can be kept for 5 days)
available. However, plasma is kept in the blood bank and
takes time to issue and transport. It is often further delayed
waiting for a blood type. Platelets must be kept in the blood
bank because of their stringent storage requirements (20 –
24°C on a shaker or rotator) and short outdate. Cryoprecipi-
tate is stored frozen so it must be thawed, and is frequently
pooled in the blood bank before issue. It can be obtained
faster if it is specifically requested as unpooled.
The willingness of blood bank directors and transfusion
committees to allow the storage and use of uncrossmatched
RBC in trauma resuscitation units is critically dependent of
the ability of those units to not violate the conditions of blood
bank licensure. A trauma unit director must enforce appro-
priate blood use and documentation.
CONCLUSIONS
Early administration of plasma on a 2:3 or 1:1 ratio with
units of RBC is advisable in the most seriously injured and
Volume 60 • Number 6
Early FFP, Cryo, and Platelets
massively bleeding individuals. It is constrained by delays in
thawing FFP or transporting thawed plasma to the trauma
center. Platelets can generally be delayed until 10 –20 units of
RBC have been given and in less urgent situations should be
guided by platelet counts or evidence of coagulopathic bleed-
ing. However, cohort data on survival in massive transfusion
suggest that administering a standard 6 unit platelet transfu-
sion for each 7– 8 units of RBC may be associated with
improved survival. Cryoprecipitate, usually given in 10 U
increments can rapidly raise the concentration of fibrinogen.
The benefits of giving additional cryoprecipitate to raise
fibrinogen and von Willebrand’s factor are unknown.
REFERENCES
1. Collins JA. Massive transfusion and current blood – banking
practices. In Chaplin H Jr., Jaffe ER, Lenfant C, Valeri CR (eds).
Preservation of red blood cells. NAS, Washington, DC. 1973:37– 41.
2. Collins JA. The pathophysiology of hemorrhagic shock. In Collins
JA, Murawski K, and Shafer AW, (eds). Massive Transfusion in
Surgery and Trauma. Progress in Clinical and Biological Research,
Vol 108. A.R. Liss, New York. 1982:5–29.
3. Como JJ, Dutton RP, Scalea TJ, Edelman BB, Hess JR. Blood
transfusion rates in the care of acute trauma. Transfusion. 2004;
44:809 – 813.
4. Armand R, Hess JR. Treating coagulopathy in trauma patients.
Transfus Med Rev. 2003;17:223–231.
5. Bellamy RF, Maningas PA, Veyer JS. Epidemiology of trauma:
military experience. Ann Emer Med. 1986;15:1384 –1388.
6. Wudel JH, Morris JA, Yates K, Wilson A, Bass SM. Massive
transfusion: Outcome in blunt Trauma Patients. J Trauma. 1991;
31:1–7.
7. Velmahos GC, Chan L, Chan M, et al. Is there a limit to massive
blood transfusion after trauma? Arch Surg. 1998;133:947–952.
8. Wilder DM, Reid TJ, Bakaltcheva IB. Hypertonic resuscitation and
blood coagulation. In vitro comparison of several hypertonic
solutions for their action on platelets and plasma coagulation.
Thromb Res. 2002;107:255–261.
9. Treib J, Haass A, Pindur G. Coagulation disorders caused by
hydroxyethyl starch. Thromb Haemost. 1997;78:974 – 83.
10. Hirshberg A, Dugas M, Banez EI, Scott BG, Wall MJ Jr, Mattox
KL. Minimizing dilutional coagulopathy in exsanguinating
hemorrhage: a computer simulation. J Trauma. 2003;54:454 – 463.
11. Hiippala S, Myllylä G, Vahtera E. Hemostatic factors and
replacementof major blood loss with plasma-poor red cell
concentrates. Anesth Analg. 1995;81:360 –365.
12. Doyle S, O’Brien P, Murphy K, Fleming C, O’Donnell J.
Coagulation factor content
of solvent/detergent plasma compared with fresh frozen plasma.
Blood Coag Fibrinol. 2003;14:283–287.
13. Slichter SJ. Platelet transfusion: future directions. Vox Sang. 2004;
87(Suppl):47–51.
14. Phillips TF, Soulier G, Wilson RF. Outcome of massive transfusion
exceeding two blood volumes in trauma and emergency surgery.
J Trauma. 1987;27:903–910.
15. Wolberg AS, Meng ZH, Monroe DM, Hoffman M. A sytematic
evaluation of the effect of temoerature on coagulation enzyme
activity and platelet function. J Taruma. 2004;56:1221–1228.
16. Cosgriff N, Moore E, Sauaia A, Kenny-Moynihan M, Burch JM,
Galloway B. Predicting life-threatening coagulopathy in the massively
transfused trauma patient; Hypothermia and acidoses revisited.
J Trauma. 1997;42:857– 862.
S55
 The Journal of TRAUMA威 Injury, Infection, and Critical Care
17. Cinat ME, Wallace WC, Nastanski F, et al. Improved survival
following massive transfusion in patients who have undergone
trauma. Arch Surg. 1999;134:964 –968 discussion 968 –970.
18. Reed RL, 2nd, Ciavarella D, Heimbach DM, et al. “Prophylactic
platelet administration during massive transfusion. A prospective,
randomized, double-blind clinical study.” Ann Surg. 1986;203:
40 – 8.
19. British Committee for Standards in Haematology, Blood Transfusion
Task Force. Guidelines for the use of fresh frozen plasma,
cryoprecipitate and cryosupernatant. Br J Haematol. 2004;126:11–28.
DISCUSSION
Dr. Fred Moore: When thinking about the issues dis-
cussed in this manuscript, I see three problems. First, we
don’t understand the basic pathophysiology of the coagulopa-
thy that we’re trying to treat. As a trauma surgeon I’ve been
taught that coagulopathy occurs because of hemodilution,
consumption, acidosis and hypothermia. But actually, if you
test for coagulopathy as severely injured patients come off the
helicopter, they will be coagulopathic before any of these
events have happened. So there’s something else going on
and we don’t understand it. Second, we don’t know how to
monitor coagulopathy early when it’s most important. By
default, we use the PT. Third, we have empiric guidelines for
FFP and platelet administration that were derived in the
1980s when whole blood was commonly used. And as we’ve
heard, these were based on wash-out equations and grossly
underestimate what we should be giving patients who require
massive transfusions.
So what do the authors recommend? They recommend
that we aggressively use FFP. So my first question to the
authors is, we need to be more aggressive, but when do we
start? One of the comments in the paper was after the patient
loses one blood volume. However, as a trauma surgeon, I’m
not quite sure how I would figure that out in a patient who is
potentially bleeding from multiple sites. I believe that as soon
as you think somebody is bleeding a lot, you ought to give
fresh frozen plasma.
Now, the second intervention the authors talked about
was platelets, but there were some conflicting statements.
One was that we should start platelets after 10 to 20 units of
packed red blood cells. On the other hand, we should be
giving these platelets at a very high ratio of 0.8 units of
platelets per unit of packed red blood cells. So if you’re not
going to start until 20 units, then you must be going to give
a lot of platelets late. Why not start up front?
The last issue is the use of cryoprecipitate. We don’t give
a lot of early cryoprecipitate. Our patients arrive into the ICU
with a mean fibrinogen level of about 160 mg/dl (only 1% are
less than 50 mg/dl). Over the next 24 hours, fibrinogen levels
quickly increase into the low 400s. These data are very tight.
We believe this is due to the acute phase response and that the
liver is producing a lot of fibrinogen.
Dr. Lloyd Ketchum: I’ll take the question of platelets. I
agree with Dr. Hess that only the small group of patients
destined to require massive transfusion need any platelets.
S56
But when can you make that call? I think you can make that
call pretty early when somebody is going to be a massive
transfusion patient. Once you believe you have a massive
transfusion case, you should be going to a one-to-one-to-one
replacement of FFP, red cells and platelets. Admittedly,
based on the dilutional calculations, you’re never going to get
completely back to normal concentrations because the addi-
tive solutions dilute all of the blood products, but I think you
need to be headed in that direction.
Dr. Angel Delgado: I really enjoyed the paper because it
gave me more questions than answers and as a scientist doing
research, that’s exciting. Dr. Hess talked about vitamin K-
dependent pathways which are affected on their pH condi-
tions. And he talked about platelet activation problems with
hypothermia. Dr. Lawson talked about platelet phospholipid
on membranes that have to be expressed in order for coagu-
lation cascades to occur. And all these things are affected by
hypothermia, acidosis and Hemodilution. The data about tissue
factor expression in monocytes also has to be understood. These
factors contribute to making clinical decisions on when to use
different products. Understanding will give us an aid on earlier
intervention and helping in preventing this derangement.
I have two main questions for the authors. One, we’re
trying to find a way to measure and come up with a test that
will give an indication early on of coagulopathy, and we
don’t have that. Right now, the best thing that I believe we
have is the TEG. Certain labs are doing TEG, in the OR and
at the bedside. And there may be some parameters within the
TEG that may elucidate on when to use FFP or platelets or
stop the use of all those. Also, what is your impression about
using some of these new drugs, like platelet derivatives,
rFVIIa, earlier in the therapy of massive hemorrhage and
trauma to ameliorate the trauma induced coagulopathy and
improve the outcome of mortality?
Dr. Ketchum: As far as your question about FFP, when
do we need it and where do we need it, I think that there’s a
lot of consumption going on. And there’s two parts of the
consumption. There is the consumption of the clotting fac-
tors, but there is also the consumption of the clotting inhib-
itors and the fibrinolytic pathway components, which shut
down clotting. So I think you can reasonably expect that if
you put FFP in, you’re infusing some fibrinogen and you’re
infusing both sides of the hemostatic system which are being
consumed because of the tissue injury. Admittedly the co-
agulopathy is compounded by cold and by dilution, but fun-
damentally, particularly in massive injury, you’re just using
everything up.
Dr. John Owings: One of the things that we’ve seen is,
even within that first four-hours after injury, is a dramatic
increase in monocyte expression of tissue factor. I think the
upshot is that in the very initial period of time they are
hypercoagulable. It’s a question if giving all these blood
products is maintaining an organized coagulation that will
then stave off the late complications that are immune.
Supplement 2006

Dr. Jeff Lawson: I just want to add some perspective as
a guy who spent five or six years isolating clotting proteins
from plasma, from cryo, and from FFP. It’s important to
remember that the lowest concentration of factors on a molar
basis is factor VII and factor V. They go away really fast
because they live in concentrations that are between one and
ten nanomolar. Things like factor IX and factor X live at
concentrations that are at 100 to 200 nanomolar. And pro-
thrombin lives at a concentration that is micromolar. You can
isolate grams of prothrombin from FFP if you want to. Where
I give cryoprecipitate is not for fibrinogen, but for the things
that I think got burned out. Cryoprecipitate is what falls out
of solution when you freeze and thaw plasma. So if I’m using
cryo, it’s because I think that V or VII deficient or I’m giving
platelets because the platelets have factor V. So I think there
are benefits to these things that are not subtle and we sort of
lump them into these one-to-one-to-one kind of ratios and it’s
really not that way at all. It’s what do you think you’re out of
and what’s the best source to get it? If you’ve burned a lot of
resources, you’re likely out of factors V and VII pretty early
on in the game.
Dr. Mike Dubick: If you had a preference, would you
like to use these blood products as your first fluid, or would
it not really matter if you give crystalloid first?
Dr. Seppo Hiippala: We all know that it’s impossible in
that situation. You always start with an intravenous line and
you start with crystalloids. You have to maintain the blood
pressure in an acceptable level, even though you are some-
times intentionally avoiding normovolemia. I⬘m referring to
tactical hypotensive cases. Rarely are blood products given
from the hip. It unfortunately takes some time.
Dr. Tom Repine: I did extensive reading about ratios
and prophylactic component replacing while trying to come
up with a massive transfusion protocol for use at our combat
support hospital. There isn’t any data that says giving early
prophylactic component therapy, basically of any kind, de-
creases the number of transfusions, impacts survival, or the
like. The reason we held off on platelets was a very com-
monsense reason. It’s the same kind of commonsense reason
why early FFP in these patients seems to make sense. A
casualty who was brought in with no legs was obviously
going to require a lot of blood products and was most likely
in hemorrhagic shock, no matter how effective field therapy
had been to that point. In these patients, we gave FFP right up
front, but it was as much for volume and to control the chaos
of the resuscitation as it was for any other reason. The
intuitive reason to give it up front makes sense with the
clotting factors as well, just hasn’t panned out in clinical
study. On the other hand, a patient rarely shows up with
thrombocytopenia unless there is some congenital thrombo-
cytopenia. Giving platelets right away is probably going to
flush the platelets right out of the bleeding patient and you’re
not getting much benefit. So waiting until there is a dilutional
thrombocytopenia, which is reliably after 10 to 20 units of
packed red blood cells, makes sense intuitively. Again, there
Volume 60 • Number 6
Early FFP, Cryo, and Platelets
are certain things that you know you start off behind with. If
we say every trauma patient shows up and they’re coagulo-
pathic, I think starting with FFP makes excellent sense. Plate-
lets make sense to wait, because the patient should not be
thrombocytopenic when they show up and you can catch up
with that. The other thing that we did was, we gave cryopre-
cipitate early without an indication. The literature doesn’t
support that, either. But we gave it because it thawed very
quickly and we expected the labile factors to be down in
stored units. Overall, there is no prophylactic transfusion data
that says it’s any good. All these ratios in varying studies are
based on retrospective analysis of what was given prospec-
tively. Managing a problem of this complexity under the
pressure of time, especially where getting behind can have
disastrous consequences, warrants being proactive, even if
only for simplicity’s sake.
Dr. Harvey Klein: I just wanted to comment again on
the fixed ratio issue. If you look at the cardiac surgery
literature, granted, a different clinical setting, but there’s
really no evidence that fixed ratios decrease bleeding, use of
red cells, morbidity or mortality. It just isn’t there. So again,
I would toss out that either we need to study this prospec-
tively or else we need a way to figure out which of these
patients needs it, because otherwise, giving everybody a fixed
ratio is going to use a lot of blood component and put a lot of
people at risk for probably a very small benefit.
Dr. John Holcomb: I’m going to speak for the surgeons
in the group. We’re all thinking the same thing right now.
Charlie, you said this earlier. There are papers that address
fixed ratios and say that they don’t work and waste products.
The problem is when we’re down there in the ED or in the
OR, we know that if we get behind, the patient is going to die.
You know that and I think everybody here would agree with
that. So you’ve got to stay out of trouble because once you’re
in trouble you can’t get out. And our solution to that is a fixed
ratio. While we would all agree a prospective study would be
optimal, every one of the clinicians here would use a fixed
ratio.
Dr. Owings: Every one of us, though, would use a
different ratio. I have good results using fresh frozen plasma,
almost no platelets and no cryo, and I give them calcium.
Dr. Rick Dutton: Two comments. Staying out of trou-
ble: if you’re in a massive transfusion situation, I don’t know
exactly what the patient needs, but I’m sure it’s not crystal-
loid. So it’s one-to-one-to-one blood products until I think of
something else. Second, in terms of getting out of trouble,
John Hess is right about our cryo use in 2000. We’ve changed
that now, because we observed in the rFVIIa era the occa-
sional patient who was really bleeding very heavily for a long
period and we can’t get to stop. And we now give them a
cocktail of cryo, platelets, and rFVIIa, and we’ve had some
good success with that.
Dr. Holcomb: One of primary observations of clinicians
who resuscitate trauma patients is that if you can
S57
 The Journal of TRAUMA威 Injury, Infection, and Critical Care
identify patients who are sick when they come in, why would
you further dilute that patient with crystalloid? It makes
absolutely no sense. They have a hemorrhagic problem. As
Dr Moore said they present coagulopathic. The problem is
bleeding, yet we resuscitate them with a solution that exac-
erbates coagulopathy. Why not resuscitate them in a hemo-
static fashion? We and others are starting to develop and
study a concept called hemostatic resuscitation for the very
S58
small group of patients that are going to bleed a lot. Person-
ally, I take down the crystalloid and hang the products (FFP,
cryo and platelets) that will help promote intravascular
hemostasis while we work on stopping the bleeding from
large holes in vessels and organs with externally applied
modalities. Big bleeding is their problem, so we feel that
giving more of the products that help stop bleeding makes
a lot of sense.
Supplement 2006