The Journal of TRAUMA威 Injury, Infection, and Critical Care
Hypercoagulability Is Most Prevalent Early after Injury and in
Female Patients
Martin A. Schreiber, MD, Jerome Differding, BS, Per Thorborg, MD, John C. Mayberry, MD, and
Richard J. Mullins, MD
Background: Hypercoagulability after injury is a major source of morbidity
and mortality. Recent studies indicate that
there is a gender-specific risk in trauma
patients. This study was performed to determine the course of coagulation after
injury and to determine whether there is a
gender difference. We hypothesized that
hypercoagulability would occur early after injury and that there would be no difference between men and women.
Methods: This was a prospective cohort study. Inclusion criteria were admission to the intensive care unit, Injury Severity Score > 4, and the ability to obtain
consent from the patient or a relative. A
Thrombelastograph (TEG) analysis was
performed and routine coagulation pa-
rameters and thrombin-antithrombin
complexes were measured within 24 hours
of injury and then daily for 4 days.
Results: Sixty-five patients met criteria for entry into the study. Their mean
age was 42 ⴞ 17 years and their mean
Injury Severity Score was 23 ⴞ 12. Forty
patients (62%) were men. The prevalence
of a hypercoagulable state by TEG was
62% on day 1 and 26% on day 4 (p <
0.01). Women were significantly more hypercoagulable on day 1 than men as measured by the time to onset of clotting
(women, 2.9 ⴞ 0.7 minutes; men, 3.9 ⴞ 1.5
minutes; p < 0.01; normal, 3.7– 8.3 minutes). Mean platelet counts, international
normalized ratios, and partial thrombo-
plastin times were within normal limits
throughout the study. Thrombin activation as measured by thrombin-antithrombin complexes decreased from 34 ⴞ 15
␮g/L on day 1 to 18 ⴞ 8 ␮g/L (p < 0.01)
on day 4, consistent with the prevalence of
hypercoagulability by TEG.
Conclusion: Hypercoagulability after injury is most prevalent during the
first 24 hours. Women are more hypercoagulable than men early after injury. The
TEG is more sensitive than routine coagulation assays for the detection of a hypercoagulable state.
Key Words: Hypercoagulable state,
Thrombelastogram, Gender, Thrombin
activation, Thrombosis.
J Trauma. 2005;58:475–481.
T
he hypercoagulable state that follows traumatic injury is
integral to limiting hemorrhage, but excessive thrombosis is associated with significant late morbidity and mortality. Untreated trauma patients undergoing routine venography have been found to have a 58% incidence of
thromboembolic events.1 The onset of adult respiratory distress syndrome and multiple organ failure after trauma have
also been associated with a hypercoagulable state characterized by elevated tissue factor levels.2,3 Fibrin deposition
within the alveolar compartment occurs in adult respiratory
distress syndrome.4,5 Activated protein C, a potent anticoagulant and profibrinolytic agent, is currently the only agent
that has been shown to improve outcome in septic patients
who develop multiple organ failure.6
Submitted for publication November 8, 2004.
Accepted for publication November 22, 2004.
Copyright © 2005 by Lippincott Williams & Wilkins, Inc.
From Oregon Health & Science University, Portland, Oregon.
Supported by Public Health Service grant 5 M01 RR000334.
Presented at the 63rd Annual Meeting of the American Association for
the Surgery of Trauma, September 29 –October 2, 2004, Maui, Hawaii.
Address for reprints: Martin A. Schreiber, MD, FACS, Oregon Health
& Science University, 3181 S.W. Sam Jackson Park Road, Mail Code
L223A, Portland, OR 97239; email: [email protected].
DOI: 10.1097/01.TA.0000153938.77777.26
Volume 58 • Number 3
Data from animal studies indicate that male gender is a
risk factor for death after hemorrhage and sepsis.7,8 Human
studies of gender dimorphism after trauma have produced
conflicting results. Although some studies have shown no
gender differences in outcomes, other studies have shown
survival advantages for women below the age of 50.9 –12
Although the cause of improved outcome among women is
not known, the presence of relatively high levels of estrogen
results in a hypercoagulable state that might be beneficial
early after injury in bleeding patients.13–16
Despite the critical impact of a hypercoagulable state on
trauma patients, routine coagulation assays are not useful in
its detection. The Thrombelastograph (TEG) analyzer provides a comprehensive functional evaluation of overall coagulation status and has the added advantage that it can be
performed at the bedside. Kaufmann et al. reported that the
majority of trauma patients undergoing TEG analysis during
their initial evaluation are hypercoagulable.17 Serial coagulation analyses using the TEG and comparisons between men
and women have not been previously published. The purpose
of this study was to determine the time course of coagulation
changes over 4 days in critically injured trauma patients and
to determine whether there is a difference between men and
women. We also sought to compare routine coagulation parameters to the TEG. We hypothesized that hypercoagulability would occur early after injury and there would be no
gender differences.
475
The Journal of TRAUMA威 Injury, Infection, and Critical Care
Table 1 Baseline Patient Characteristics and Initial Resuscitationa
No.
Age (yr)
Injury Severity Score
Time to first blood draw (h)
Fluid (L) during first 24 h
Packed red blood cells (L)
transfused during first 24 h
a
All Patients
Men
Women
p Value
65
42 ⫾ 17
23 ⫾ 12
17.3 ⫾ 5.7
5.2 ⫾ 3.8
0.8 ⫾ 1.6
45 (69%)
40.0 ⫾ 16.3
23 ⫾ 13
17.0 ⫾ 5.4
5.0 ⫾ 4.0
0.5 ⫾ 1.4
20 (31%)
45.5 ⫾ 18.3
23 ⫾ 10
18.1 ⫾ 6.4
5.5 ⫾ 3.4
1.2 ⫾ 1.7
⬎0.1
⬎0.1
⬎0.1
⬎0.1
⬎0.1
Mean values are shown with their standard deviations.
PATIENTS AND METHODS
This was a prospective cohort study performed at one of
only two Level I trauma centers in the state of Oregon. The
Oregon Health & Science University Institutional Review
Board approved the protocol. Trauma patients admitted to the
intensive care unit with a minimum Injury Severity Score
(ISS) of 4 were eligible for the study. Patients whose initial
laboratory assays could not be obtained within 24 hours of
injury and those with isolated head injury were excluded.
Informed consent was obtained from the patient or a legal
representative.
Clinical data collected included age, gender, ISS, time of
initial blood draw relative to the time of injury, thromboembolic complications, and outcome. Deep vein thrombosis
(DVT) prophylaxis consisted of sequential compression devices and enoxaparin 30 mg subcutaneously twice a day when
not contraindicated because of a bleeding risk. Duplex was
used to diagnose deep vein thrombosis and computed tomography was used to diagnose pulmonary embolus. These tests
were obtained on the basis of the following trauma service
protocols. Ambulatory patients underwent daily physical examinations. Patients who developed unilateral leg swelling
were evaluated with duplex scanning. Patients who underwent prolonged bedrest were evaluated with weekly duplex
scans. Patients who developed acute shortness of breath or
hypoxia and in whom the pulmonary findings were not explained by chest radiography underwent chest computed tomography to exclude pulmonary embolus.
Patients underwent once-daily laboratory assessments
for 4 days. A single blood draw was performed for all analyses. Platelet counts, international normalized ratios (INRs),
partial thromboplastin times (PTT), and fibrinogen levels
were performed in the clinical laboratory at Oregon Health &
Science University. Blood for thrombin-antithrombin complexes (TATs) and d-dimer assays was centrifuged and the
plasma was separated and frozen in a ⫺80°C freezer for batch
analysis in the General Clinical Research Center. Thrombinantithrombin complexes were performed using commercially
obtained enzyme-linked immunosorbent assay kits from
Dade Behring (Newark, DE) and d-dimers were performed
using enzyme-linked immunosorbent assay kits from Diagnostica Stago (Asnieres-Sur-Seine, France).
Daily TEG assessments were performed on a 5000 series
Thrombelastograph Hemostasis Analyzer (Hemoscope Cor476
poration, Niles, IL). Level I and Level II controls were run
and confirmed to be within specified limits each day of the
study. Assays were performed by a trained technologist.
Fresh whole blood from the original syringe was activated
with kaolin within 3 minutes of being drawn and assays were
run at 37°C. Analyzed parameters included the r-time (r),
angle (␣), and maximum amplitude (MA). The r represents
the time from the start of a sample run until clot is detectable.
This is the time point at which most traditional assays such as
INR and PTT reach their endpoints. The angle represents the
rate of clot formation. It is primarily affected by fibrinogen
function and, to a lesser degree, platelet function. The MA is
a measure of maximum strength of the clot. It is primarily
determined by platelet function, although fibrinogen function
also contributes.
Statistical analysis was performed using SPSS software,
version 12 (SPSS, Inc. Chicago, IL). Categorical data were
analyzed using ␹2 analysis or Fisher’s exact test when the
value in any cell was less than 5. Independent continuous data
were analyzed with analysis of variance, and serial continuous data were analyzed with repeated-measures analysis of
variance. Parameters that were not normally distributed were
analyzed with the Wilcoxon signed-ranks test. A two-tailed
Pearson’s correlation coefficient was calculated to determine
whether two values correlated. A value of p ⬍ 0.05 was
considered significant.
RESULTS
Sixty-five patients were eligible for the study and underwent further analysis. Sixty-three patients (97%) had a blunt
mechanism of injury. Demographics, time to first blood draw,
and resuscitation volumes during the first 24 hours are shown
in Table 1. There was no difference between men and women
with respect to age, ISS, time to first blood draw, fluid given,
or blood transfused. Twenty patients (31%) had an ISS between 4 and 15. These were proportionally distributed between men and women.
The daily prevalence of a hypercoagulable state, as measured by a shortened time to detectable clot (r value), is
shown for all patients in Figure 1 and by gender in Figure 2.
Patients with an r value below the normal limits of the test
(⬍3.7 minutes) were considered hypercoagulable. The greatest prevalence of a hypercoagulable state occurred on day 1
and was progressively less on days 2 and 3. Women were
March 2005
Hypercoagulability after Trauma
Fig. 1. Prevalence of a hypercoagulable state as defined by r ⬍ 3.7
minutes. r, time from start of assay until clot is detectable. ap ⫽ 0.03
for day 1 versus day 2; bp ⫽ 0.04 for day 2 versus day 3; cp ⬍ 0.01
versus day 1.
Fig. 2. Prevalence of a hypercoagulable state, comparing men and
women. r, time from start of assay until clot is detectable. ap ⫽ 0.01
for men on day 1 versus women on day 1; bp ⫽ 0.02 for women on
day 1 versus women on day 2; cp ⬍ 0.01 versus women on day 1;
d
p ⬍ 0.01 versus men on day 1.
significantly more likely to be hypercoagulable on day 1 than
men. The prevalence of hypercoagulability in women significantly decreased by day 2, and there was no difference
between men and women after day 1.
Mean values for TEG parameters over the 4-day period
are shown in Table 2. The mean time to clot formation (r) was
equal to the lower limit of normal on day 1 and increased
significantly on days 3 and 4. The mean rate of clot formation
(␣) and the maximum strength of the clot (MA) were within
normal limits throughout the study and did not change
significantly.
Comparison of TEG values between men and women
reveals that the mean r value for women was significantly less
than for men on day 1 (men, 3.9 ⫾ 1.5 minutes; women, 2.9
⫾ 0.7 minutes; p ⬍ 0.01). This indicates that the mean time
to clot formation was 1 minute earlier in women than in men.
There were no other significant differences between men and
women with respect to other TEG variables.
Standard clotting parameters are shown in Table 3. All
mean INR, PTT, and platelet count values were within normal limits for our laboratory. Mean INR values were significantly greater on days 1 and 2 than on days 3 and 4, but these
differences do not appear to be clinically relevant. Mean PTT
values did not change throughout the course of the study.
Mean platelet count values on days 2 and 3 were slightly
greater than the lower limit of normal, and they were significantly lower than on days 1 and 4. Mean fibrinogen levels
increased significantly on each day of the study and exceeded
normal limits on days 3 and 4. Mean d-dimer levels were
markedly elevated throughout the course of the study. Similar
to the mean platelet values, these levels were greater on days
1 and 4 than on days 2 and 3. Standard clotting parameter
values did not differ between men and women.
Thrombin activation as measured by TATs is shown in
Figure 3. Thrombin-antithrombin complexes were also markedly elevated during the 4-day period. Thrombin activation
was greatest on day 1 and diminished over time. These values
did not differ between the genders.
The correlation between TEG parameters and corresponding routine coagulation parameters is shown in Figure
4. Significant correlations occurred between r and PTT and
MA and platelets on day 1. There were no significant correlations after day 1.
Table 2 Comparison of Thrombelastrograph Analyzer Parameters Over the 4-Day Period
Day
1
2
3
4
r (min)
(Normal, 3.7–8.3)
␣ (degrees)
(Normal, 46.8–73.6)
MA (mm)
(Normal, 54.5–72.5)
3.7 ⫾ 1.5a
3.9 ⫾ 1.5a
4.8 ⫾ 1.7
4.7 ⫾ 1.7
71.5 ⫾ 4.5
70.5 ⫾ 6.8
68.6 ⫾ 8.7
69.3 ⫾ 8.2
63.2 ⫾ 5.9
65.3 ⫾ 10.9
66.3 ⫾ 10.3
68.1 ⫾ 11.6
a
p ⬍ 0.01 vs. day 3 and day 4.
r, time from start of assay until clot is detectable; ␣, rate of clot formation; MA, maximum amplitude.
Volume 58 • Number 3
477
The Journal of TRAUMA威 Injury, Infection, and Critical Care
Table 3 Comparison of Standard Coagulation Parameters
Day
INR
(Normal, 0.9–1.2)
PTT (s)
(Normal, 26–36)
Platelet Count ⫻ 103/mm3
(Normal, 150–400)
Fibrinogen (␮g/L)
(Normal, 200–450)
d-Dimer (ng/mL)
(Normal, ⬍400)
1
2
3
4
1.11 ⫾ 0.02a,b
1.13 ⫾ 0.02a,b
1.07 ⫾ 0.02
1.06 ⫾ 0.04
29.4 ⫾ 0.6
30.7 ⫾ 0.8
31.6 ⫾ 1.2
32.7 ⫾ 1.7
182 ⫾ 8a,c
151 ⫾ 6b
158 ⫾ 7b
185 ⫾ 9
305 ⫾ 11*
442 ⫾ 17*
552 ⫾ 20*
587 ⫾ 22*
9,275a,c ⫾ 3,974
3,226b ⫾ 387
3,515b ⫾ 436
7,230 ⫾ 2,672
p ⬍ 0.01 vs. day 3.
p ⬍ 0.01 vs. day 4.
c
p ⬍ 0.01 vs. day 2.
* p ⬍ 0.01 for each value compared to the others.
INR, international normalized ratio; PTT, partial thromboplastin time.
a
b
Chemical DVT prophylaxis was used in 19 patients
(29%) during the course of the study and in 4 additional
patients after conclusion of the 4-day study period. Four
patients (6.2%) suffered a thromboembolic event. Three patients were found to have a DVT and one patient had a
pulmonary embolus. All four of these patients were men and
were hypercoagulable by r value on day 1. The prevalence of
a thromboembolic complication was 4 of 40 in hypercoagulable patients versus 0 of 24 in nonhypercoagulable patients
(p ⬎ 0.1).
DISCUSSION
Multiple intrinsic mechanisms enable patients to stop
bleeding after injury, including thrombin-mediated clot formation, platelet aggregation, and vasoconstriction. Failure of
one or more coagulation mechanisms exacerbates blood loss,
and exsanguination continues to be a significant source of
early death after injury.18 Coagulopathy, as defined by elevations of the INR and PTT, occurs in approximately 25% of
seriously injured patients and is associated with increased
mortality.19,20 After the initial resuscitation and correction of
coagulopathy in these patients, complications resulting from
Fig. 3. Thrombin-antithrombin complexes. Normal range is 1 to 4.1
␮g/L. ap ⬍ 0.01 versus days 2, 3, and 4; bp ⬍ 0.01 versus day 4.
478
hypercoagulability and dysfunctional inflammation become
dominant.
The TEG is a bedside test that provides comprehensive
information concerning coagulation. This test differs from
other laboratory measures of coagulation because it is performed on fresh whole blood and it is used to rapidly assess
the interaction of platelets with the protein coagulation cascade from initial platelet-fibrin interaction, through platelet
aggregation, clot strengthening, and fibrin cross-linking, to
eventual clot lysis. The TEG has been used to assess coagulation in numerous clinical settings that include trauma, abdominal surgery, cardiac surgery, sepsis, transplant surgery,
and the assessment of the efficacy of low-molecular-weight
heparin.21–25 These studies demonstrate the reliability and
validity of TEG for identifying both hypercoagulable and
hypocoagulable states.
This observational cohort study shows that, on the basis
of TEG r values, the majority of trauma patients are hyper-
Fig. 4. Correlation between TEG values and corresponding routine
coagulation variables. r, time from start of assay until clot is
detectable; INR, international normalized ratio; PTT, partial thromboplastin time; MA, maximum amplitude. *p ⬍ 0.01 for the correlation between r and PTT and MA and platelets on day 1.
March 2005
Hypercoagulability after Trauma
coagulable early after injury. Our observation in this group of
predominantly blunt trauma patients is consistent with the
observations of Kaufmann et al., who showed that 65% of
trauma patients are hypercoagulable in the emergency department based on their TEG profile.17
The current study adds the additional observation that the
majority of patients correct their coagulation status to normal.
The prevalence of hypercoagulability sequentially decreases
on day 2 and day 3 and stabilizes on day 4. Thrombin
activation as measured by TATs was also greatest on day 1
and gradually decreased over the course of the study, mirroring the TEG findings. This is likely explained by the fact that
trauma causes exposure and deencryption of tissue factor,
resulting in activation of the coagulation cascade, thrombin
production, and production of inflammatory cytokines that
further accelerate clotting.26,27 This process is greatest immediately after injury and decreases as tissue factor levels
decrease.
The prevalence and severity of hypercoagulability detected in women on day 1 was an unexpected finding of this
study. Greater than 80% of the women were hypercoagulable,
and the mean time to onset of clotting by TEG was 1 minute
less than in men. The prevalence of hypercoagulability on
day 2 in women was significantly less than on day 1 and was
essentially equivalent to men throughout the remainder of the
study. The greater propensity to develop a hypercoagulable
state early after injury and rapidly resolve it after hemorrhage
has stopped could convey a survival advantage for women
after trauma and partially explain the gender dimorphism that
has been described. Theoretically, this would result in reduced hemorrhage early and decreased complications related
to hypercoagulability later. Although all of the thromboembolic complications recorded in this study occurred in men,
the difference did not reach statistical significance. Further
research is needed to define the factors that may contribute to
this gender bias toward initial hypercoagulability followed by
rapid resolution, including menopausal status, pharmaceutical
hormone use, and individual endogenous hormonal levels.
Routine coagulation parameters including the INR, PTT,
and platelet count were within normal limits and failed to
detect early hypercoagulability. The differences noted in
mean INR and platelet values over the course of the study are
unlikely to be clinically relevant. Mean fibrinogen values
increased each day. This increase was not associated with an
increase in ␣ or MA, suggesting that the increase in fibrinogen levels after trauma does not produce a functional
difference.
In general, TEG and corresponding coagulation parameters correlated poorly. Significant correlations occurred between r and PTT and MA and platelets on day 1. There were
no significant correlations after day 1 and there was no
correlation between r and INR. The poor correlation between
TEG and routine coagulation parameters suggests that TEG
provides additional information regarding the thrombosis status of trauma patients compared with that provided by conVolume 58 • Number 3
ventional tests. A major limitation of commonly used laboratory tests of coagulation function is a failure to identify a
hypercoagulable state.
The clinical significance of the transient hypercoagulable
state described in this work is yet to be determined. Enhanced
coagulation may be beneficial in patients with active hemorrhage, but patients with substantial tissue injury and minimal
risk of hemorrhage may suffer pathologic consequences including occlusion of the microcirculation leading to organ
injury and an excessive inflammatory response. Further research is needed to determine whether prophylactic anticoagulation therapy would be beneficial in trauma patients who
are identified to be hypercoagulable by TEG.
Patients with isolated head injury were excluded from
this study because they differ from trauma patients with
multiple injuries with respect to coagulation. Tissue factor
levels in head-injured patients exceed those in non– headinjured trauma patients.28 Marked activation of coagulation
pathways occur, as evidenced by diminished fibrinogen levels and platelet counts as well as elevated fibrin degradation
products, TATs, and prothrombin fragments F1 ⫹ 2.29,30 This
patient population merits a separate study.
In summary, trauma results in a hypercoagulable state
that occurs early after injury. This state is more prevalent and
more profound in women during the first 24 hours. Hypercoagulability after trauma is not detected by routine coagulation assays. The TEG is an alternative test of coagulation
that may prove useful in identifying trauma patients who are
at risk for developing complications related to a hypercoagulable state.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
Geerts WH, Code KI, Jay RM, et al. A prospective study of venous
thromboembolism after major trauma. N Engl J Med. 1994;
331:1601–1606.
Gando S, Nanzaki S, Morimoto Y, et al. Systemic activation of
tissue-factor dependent coagulation pathway in evolving acute
respiratory distress syndrome in patients with trauma and sepsis.
J Trauma. 1999;47:719 –723.
Gando S, Kameue T, Nanzaki S, et al. Participation of tissue factor
and thrombin in posttraumatic systemic inflammatory syndrome. Crit
Care Med. 1997;25:1820 –1826.
Idell S. Extravascular coagulation and fibrin deposition in acute lung
injury. New Horiz. 1994;2:566 –574.
Hasegawa N, Husari AW, Hart WT, Kandra TG, Raffin TA. Role of
the coagulation system in ARDS. Chest. 1994;105:268 –277.
Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and safety of
recombinant human activated protein C for severe sepsis. N Engl
J Med. 2001;344:699 –709.
Wichmann MW, Zellweger R, De Maso CM, Ayala A, Chaudry IH.
Enhanced immune responses in females, as opposed to decreased
responses in males following hemorrhagic shock and resuscitation.
Cytokine. 1996;8:853– 863.
Diodato MD, Knoferl MW, Schwacha MG, Bland KI, Chaudry IH.
Gender differences in the inflammatory response and survival
following hemorrhage and subsequent sepsis. Cytokine. 2001;
14:162–169.
479
The Journal of TRAUMA威 Injury, Infection, and Critical Care
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
Napolitano LM, Greco ME, Rodriguez Z, et al. Gender differences
in adverse outcomes after blunt trauma. J Trauma. 2001;50:274 –
280.
Croce MA, Fabian TC, Malhotra AK, Bee TK, Miller PR. Does
gender difference influence outcome? J Trauma. 2002;53:889 – 894.
Wohltmann CD, Franklin GA, Boaz PW, et al. A multicenter
evaluation of whether gender dimorphism affects survival after
trauma. Am J Surg. 2001;181:297–300.
George RL, McGwin G Jr, Windham ST, et al. Age-related gender
differential in outcome after blunt or penetrating trauma. Shock.
2003;19:23–32.
Olivieri O, Friso S, Manzato F, et al. Resistance to activated protein
C in healthy women taking oral contraceptives. Br J Haematol.
1995;91:465– 470.
Quehenberger P, Loner U, Kapiotis S, et al. Increased levels of
activated factor VII and decreased plasma protein S activity and
circulating thrombomodulin during use of oral contraceptives.
Thromb Haemost. 1996;76:729 –734.
A multicenter study of coagulation and haemostatic variables during
oral contraception: variations with four formulations—Task Force on
Oral Contraceptives. WHO Special Program of Research,
Development and Research Training in Human Reproduction, World
Health Organization, Geneva, Switzerland. Br J Obstet Gynaecol.
1991;98:1117–1128.
Meade TW, Haines AP, North WR, et al. Haemostatic, lipid, and
blood-pressure profiles of women on oral contraceptives containing
50 microgram or 30 microgram oestrogen. Lancet. 1977;2:948 –951.
Kaufmann CR, Dwyer KM, Crews JD, et al. Usefulness of
thrombelastography in assessment of trauma patient coagulation.
J Trauma. 1997;42:716 –722.
Stewart RM, Myers JG, Dent DL, et al. Seven hundred fifty-three
consecutive deaths in a level I trauma center: the argument for injury
prevention. J Trauma. 2003;54:66 –71.
Macleod JB, Lynn M, McKenney MG, et al. Early coagulopathy
predicts mortality in trauma. J Trauma. 2003;55:39 – 44.
Brohi K, Singh J, Heron M, et al. Acute traumatic coagulopathy.
J Trauma. 2003;54:1127–1130.
Mahla E, Lang T, Vicenzi MN, et al. Thromboelastography for
monitoring prolonged hypercoagulability after major abdominal
surgery. Anesth Analg. 2001;92:572–577.
Grant HW, Hadley GP. Prediction of neonatal sepsis by
thromboelastography. Pediatr Surg Int. 1997;12:289 –292.
Klein SM, Slaughter TF, Vail PT, et al. Thromboelastography as a
perioperative measure of anticoagulation resulting from low
molecular weight heparin: a comparison with anti-Xa concentrations.
Anesth Analg. 2000;91:1091–1095.
Royston D, von Kier S. Reduced haemostatic factor transfusion
using heparinase-modified thrombelastography during
cardiopulmonary bypass. Br J Anaesth. 2001;86:575–578.
Findlay JY, Rettke SR, Ereth MH, et al. Aprotinin reduces red blood
cell transfusion in orthotopic liver transplantation: a prospective,
randomized, double-blind study. Liver Transpl. 2001;7:802– 807.
Kaneko T, Fujii S, Matsumoto A, et al. Induction of tissue factor
expression in endothelial cells by basic fibroblast growth factor and
its modulation by fenofibric acid. Thromb J. 2003;1:6 –14.
Bokarewa MI, Morrissey JH, Tarkowski A. Tissue factor as a
proinflammatory agent. Arthritis Res. 2002;4:190 –195.
Gando S, Nanzak S, Kemmotsu O. Coagulofibrinolytic changes after
isolated head injury are not different from those in trauma patients
without head injury. J Trauma. 1999;46:1070 –1076.
van der Sande JJ, Veltkamp JJ, Boekhout-Mussert RJ, BouwhuisHoogerwerf ML. Head injury and coagulation disorders.
J Neurosurg. 1978;49:357–365.
Scherer RU, Spangenberg P. Procoagulant activity in patients with
isolated severe head trauma. Crit Care Med. 1998;26:149 –156.
480
DISCUSSION
Dr. Gregory A. Timberlake (Jackson, Mississippi):
Thank you, Dr. Trunkey. I was very pleased to review this
excellent paper from Dr. Schreiber and his colleagues. I want
to thank them for getting a “smooth rough” copy of the
manuscript to me in plenty of time for review.
Why do I think it’s an excellent paper? Because it raises
in my mind more questions for future research than it answers. The authors told us that they had three questions with
two underlying null hypotheses which they presented in their
study.
In this preliminary presentation, based on this prospective cohort study, they appear to have answered these questions at least preliminarily. But before accepting all of their
answers at face value, I would like the authors to answer
several questions of mine related to both their materials and
methods as well as their results.
In their materials and methods, how many patients with
low ISS—and I’ll arbitrarily define that as 4 to 15—were
included and was there a gender difference?
Why were patients with isolated brain injury included?
Why did the authors evaluate patients only when “clinically
indicated” and not every patient for DVT? And what were the
trauma service protocols that they used to make these clinical
decisions?
As far as the results, this was a 97 percent blunt trauma
population. Do they think it would be different with penetrating trauma, or is this bias just a reflection of their population?
The authors state they found no difference in the demographics, time to blood draw, et cetera, but the women were
on average five years older. Additionally, the women seem to
have a higher fluid and blood requirement, and when I asked
someone who knows something about statistics, which I
don’t, to check that, they said that the data seemed to come
awful close to being statistically significant with the higher
fluid and blood requirement between men and women. Do
they think with more patients, this would become significant?
And would it change their conclusions?
My other questions relate to the “more questions raised
then answered” issue and perhaps areas for future research. I
have already mentioned my question on what could be the
effect of penetrating trauma on these factors.
What would be the result if you looked at isolated traumatic brain injury? Why did you not screen or what would
you see if you screened all patients with DVT?
Although women seem to be protected from thromboembolic disease in this study, is this a hormonal effect or is
this related to body surface area, muscle mass, et cetera?
Why were only 50 percent of the men hypercoagulable,
and why did all of the thromboembolic disease occur in this
group? The women were much more likely to have a hypercoagulable state but had no observed TEDs.
March 2005
Hypercoagulability after Trauma
Finally, what is the cost of doing TEG? Again I’d like to
commend the authors on what I think is a very fine piece of
preliminary scholarship. I look forward to future reports from
this group. Thank you very much for the opportunity to
discuss this paper.
Dr. John T. Owings (Sacramento, California): I’d also
like to applaud you on a nice presentation and a good study.
Thromboelastography is not something new, and as we look
at the hypercoagulable state that follows trauma, which we
have been exploring for the past 15 years, it’s a combination
of up-regulation of procoagulants and then also lack of the
down-regulators of coagulation such as protein C and
antithrombin.
My concern about the study that you’ve just presented is
that my third-grade teacher told me don’t ever use a word in
its own definition. In this study, it appears to be that you’ve
described thromboelastography as a more sensitive method of
testing for hypercoagulability because in the study the patients had thromboelastograms that looked more hypercoagulable. It seems like that violates my third-grade teacher’s
rule, and I’d like you to comment on that.
Ultimately, I do think that we need a better method of
describing who is going to have thromboembolic complications. I am not sure that looking at thromboelastography up
front is going to help us to describe that. Thank you.
Dr. Frederick B. Rogers (Burlington, Vermont): Just a
quick question: How many of your female trauma patients
were using birth control pills?
Dr. Hiroshi Tanaka (Osaka, Japan): I want to know the
effect of the sex hormone. In other words, is there any
difference between ages? I mean, maybe there is a difference
in those women younger than 40 years old or older than 50
years old, because there is some menopausal effect. Thank
you.
Dr. Kenneth Proctor (Miami, Florida): At last year’s
meeting, we presented an article that showed that putting a
catheter in a trauma patient increased the operating time. I
wonder whether you can tell me whether there were more
catheters in the women versus the men?
Dr. Martin A. Schreiber (closing): I would like to thank
everybody for their insightful comments and especially Dr.
Timberlake for his kind comments. I’ll attempt to answer all
your questions. In terms of the low ISS patients, there were
20 of the 65 patients that had an ISS between 4 and 15, who
were proportionately distributed between men and women.
We excluded the brain injury population, because we see
this as a separate population because of high tissue thrombo-
Volume 58 • Number 3
plastin in the brain. We feel that these should be studied
separately. We believe that had we studied them, they would
have been more hypercoagulable initially, and that they
would have had more fibrinolysis as well.
In terms of our protocol for thromboembolic screening,
ambulatory patients are examined by physical examination. If
they have evidence of a DVT or pulmonary embolus, they are
studied.
Patients who are at high risk and who are at significant
bedrest undergo weekly duplex scans. In terms of the question about penetrating injuries, these patients are different
from blunt patients because they have proportionately less
tissue injury compared with bleeding. However, we have
shown in our laboratory that bleeding alone results in a
hypercoagulable state, and bleeding and resuscitation, depending on the resuscitation fluid, also result in a hypercoagulable state.
In terms of our statistical analysis, we did not see a trend
toward differences between the men and women in any of the
epidemiologic parameters or in the use of blood. There were a
few outliers in the women that did affect the mean numbers. We
believe that these differences between men and women result
from hormonal differences and their affects on both the inflammatory mediators as well as coagulation parameters. We did not
intend to find this, and this indeed was a surprise to us, so we did
not look at hormonal concentrations. We did not take a history
about birth control pills. We did not ask about menopause.
However, we do plan to do this in the future.
In terms of the cost of a TEG, a TEG costs $24, and that
includes the two levels of control. You do the two levels of
controls once a day, and after that, the TEG would be $18.
There is a Common Procedural Terminology code for TEG
that allows you to bill for it.
Dr. Owings, I appreciate your comments and also all the
work that you’ve done in this area. I agree with you that using
the term hypercoagulability could be debated. What we can
say for sure from these data is that women clot faster than
men, and whether or not you want to call that hypercoagulability, I think we have shown clearly that women do clot
faster than men.
Again, we don’t know about birth control pills or hormonal status. There was no difference by age in coagulation
status. We ran a correlation coefficient, and age did not
correlate with coagulation status. Basically, I think that answers all the questions. Thank you.
481