vitamin k-dependent coagulation factor deficiency in trauma: a comparative analysis between...

B L O O D M A N A G E M E N T

Vitamin K-dependent coagulation factor deficiency intrauma: a comparative analysis between international

normalized ratio and thromboelastography_3237 7..13

Bartolomeu Nascimento, Mohammed Al Mahoos, Jeannie Callum, Antonio Capone, Jennifer Pacher,

Homer Tien, and Sandro Rizoli

BACKGROUND: The use of international normalizedratio (INR) to diagnose vitamin K-dependent coagula-tion factor (VitK-CF) deficiency in trauma has limitations(inability to predict bleeding and long turnaround times).Thromboelastography (TEG) assesses the entirecoagulation process. With TEG, reaction time (TEG-R)is used to assess global coagulation factor activity andtakes less than 10 minutes. We assessed the ability ofTEG-R to detect VitK-CF deficiency in trauma, com-pared to the INR.STUDY DESIGN AND METHODS: A total of 219trauma patients with INR, TEG, and all VitK-CF mea-sured at admission were included. Demographics andlaboratory tests, drugs, blood transfusions, and severityscores were analyzed. Specificity, sensitivity, positivepredictive value (PPV), and negative predictive value(NPV) of INR (�1.3 and �1.5) and TEG-R (>8 min) todiagnose VitK-CF deficits (�50%) were calculated. Sec-ondary outcomes included time to INR and TEG results.RESULTS: Overall, TEG-R performed worse than INR.TEG-R had a sensitivity of 33% (95% CI, 16%-55%),specificity of 95% (95% CI, 91%-98%), PPV of 47%(95% CI, 23%-72%), and NPV of 92% (95% CI, 87%-95%). An INR of 1.5 or greater had a sensitivity of 67%(95% CI, 45%-84%), specificity of 98% (95% CI, 96%-99.7%), PPV of 84% (95% CI, 60%-97%), and NPV of96% (95% CI, 92%-98%). An INR of 1.3 or greater alsohad better sensitivity, PPV, and NPV. For patients onwarfarin, the times to INR results and TEG completionwere 58 (�23) and 92 (�40) minutes (p = 0.07),respectively. TEG-R was abnormal in only one patienton warfarin.CONCLUSION: Our study suggests that TEG-R is notsuperior at identifying VitK-CF deficiency compared toINR in trauma.

Hemorrhagic deaths are the main causes ofearly in-hospital mortality in trauma.1

Besides shock, tissue injury, and dilution, theincreasing use of anticoagulants and anti-

platelet (PLT) drugs might adversely contribute to theacute coagulopathy of trauma in some patients.2 Nowa-days, warfarin, an oral anticoagulant, is widely used forthe treatment and prophylaxis of many common throm-boembolic disorders such as atrial fibrillation, deep veinthrombosis, stroke, and myocardial infarction.3,4 It com-petitively inhibits vitamin K-dependent coagulation

ABBREVIATIONS: INR = international normalized ratio;

ISS = injury severity score; NPV = negative predictive value;

PPV = positive predictive value; PT = prothrombin time;

TEG = thromboelastography; TEG-R = thromboelastography test

reaction time; VitK-CF = vitamin K-dependent coagulation

factor.

From the Tory Regional Trauma Centre, Department of Surgery

& Department of Critical Care & Department of Clinical Pathol-

ogy, Sunnybrook Health Sciences Centre, University of Toronto;

and Canadian Forces Health Services, Toronto, Ontario, Canada.

Address reprint requests to: Sandro B. Rizoli, Sunnybrook

Health Sciences Centre, 2075 Bayview Avenue, Toronto, ON,

Canada M4N 3M5; e-mail: [email protected].

BN is a 2010 National Blood Foundation Grantee for the

conduct of research related to coagulopathy in trauma.

Funding for the Observational Trial was provided by the

Canadian Forces Health Services and Defense Research Devel-

opment Canada (DRDC), Government of Canada. The study

sponsor had no involvement in the study design; collection,

analysis, and interpretation of data; the writing of the

manuscript; or the decision to submit the manuscript for

publication.

Received for publication October 22, 2010; revision

received April 28, 2011, and accepted April 28, 2011.

doi: 10.1111/j.1537-2995.2011.03237.x

TRANSFUSION 2012;52:7-13.

Volume 52, January 2012 TRANSFUSION 7

factors II, VII, IX, and X (VitK-CF) as well as regulatoryanticoagulant proteins C and S.4,5

The frequency of trauma patients on warfarin isthought to be on the rise due to the aging population andwidespread use of the drug.2,6,7 In the United States, thenumber of warfarin prescriptions increased 45% between1998 and 2004.3 A recent analysis of the National TraumaDatabank, including data on 1,230,422 trauma patientsfrom 402 American centers, reported that warfarin useincreased among all patients from 2.3% in 2002 to 4.0% in2006, and in patients older than 65 years, use increasedfrom 7.3% in 2002 to 12.8% in 2006.8 The drug-inducedcoagulation factor deficiency caused by warfarin is par-ticularly concerning in trauma patients. It worsens bloodloss, enhances internal bleeding particularly intracranialand retroperitoneal, and increases mortality.3,9,10 In trau-matic intracranial hemorrhage, it has been shown that themortality rate is significantly higher in patients on antico-agulation than in similarly injured controls.2,6,11 Because ofits narrow therapeutic range, warfarin requires regularlaboratorial monitoring with prothrombin time (PT)commonly expressed as international normalized ratio(INR).12,13 However, despite the lack of studies addressingthe issue, the use of INR for trauma patients on warfarinarguably has limitations. In a small series of anticoagu-lated trauma patients, INR could not reliably identifypatients with intracranial bleeding.11 Its use in trauma hasalso been criticized for the inability to predict bleeding14,15

and long turnaround times.16,17

Recently, there is growing interest in the use of throm-boelastography (TEG) and rotation thromboelastometry,which allows a fast assessment of the entire coagulationprocess, in the setting of trauma.18,19 On the TEG test, reac-tion time (R) is the variable used to study the initiation ofcoagulation, clotting factor activity and can take less than10 minutes to be performed.20,21 With the intent of assess-ing the ability of the R on TEG to detect VitK-CF deficiencyin trauma and perform a comparative analysis with INR,we proposed a post hoc analysis of a large observationalstudy recently completed at our institution.

MATERIALS AND METHODS

This study is a post hoc analysis of data collected for alarge observational cohort study in trauma performed atSunnybrook Health Sciences Centre, a tertiary Canadiantrauma center, between February and October 2007. Theoriginal study proposed to perform an early coagulationassessment using TEG in trauma. During the study period,628 adult (age � 16 years) trauma patients arriving alivehad additional blood samples collected for TEG every timeroutine coagulation tests were performed for the first 48hours of hospitalization. TEG results were not available tothe clinician. The Sunnybrook Research Ethics Boardapproved the study protocol. Informed consent was

obtained from all patients or substitute decision makersfor participation and continuation in the study. For thoseunable to consent and without a substitute decisionmaker, informed consent was delayed in accordance withthe Tri-Council Policy Agreement for Research in Emer-gency Health Situations (Article 2.8).

We analyzed data only on the patients who had acomplete set of coagulation testing, including INR, TEG,and all VitK-CF measured on arrival to the hospital. Addi-tional blood for the admission TEG and VitK-CF assayswere collected and transported to the laboratory by clini-cal staff. However, to prevent compromising patient careas per our research protocol, recommendations forobtaining research samples were in place as follows: 1)obtaining samples for research should not interfere withand/or delay patient care in any circumstances, particu-larly in unstable patients; and 2) priority should be givento clinical samples if insufficient blood is obtained.

Data on prehospital crystalloid use, demographics,admission laboratory tests, drugs, blood transfusions, andoutcomes were prospectively recorded by professionaldata collectors during the study period. Injury severityscore (ISS) and head-abbreviated injury severity wereobtained from the trauma registry. Severe head injury wasdefined as a head-abbreviated injury severity score of 3 ormore.

Definitions

VitK-CF deficiencyClotting factor activity is considered the gold standard forthe diagnosis of coagulation factor deficiencies.13 Toassess the ability of mild elevations of INR (�1.3 and �1.5)and TEG-R (>8 min) to detect VitK-CF deficits, any factoractivity of 50% or less for any VitK-CF (II, VI, IX, and X) wasused as the gold standard for this analysis.

Coagulation factor activity of 50% or less has beenshown to be associated with mild elevations of INR whentwo or more mild coagulation factor deficits are present.22

In trauma patients, Yuan and colleagues23 demonstratedthat an INR of 1.5 or more corresponds to levels of VitK-CFof 40% to 50%. In our cohort of bleeding trauma patients,multiple coagulation factor deficits are expected in thecontext of significant blood loss and/or warfarin use. Thethreshold of INR of 1.5 or more is used as a commonlyrecommended plasma transfusion trigger for the manage-ment of trauma-associated coagulopathy.24

TEG-ROn the TEG test, the reaction time is the variable used tostudy the initiation of coagulation representing clottingfactor activity. Without including the other TEG variables(PLTs participation and fibrinolysis data), this portion ofthe test can take less than 10 minutes to be performed.20,21

TEG-R is considered normal by the manufacturer if within4 to 8 minutes.25

NASCIMENTO ET AL.

8 TRANSFUSION Volume 52, January 2012

Coagulation assessment

INRBlood samples for this test were collected using tubes con-taining 3.2% trisodium citrate. The assay was performed at37°C on a fully automated coagulation (ACL TOP analyzer,Instrumentation Laboratory, Bedford, MA), which hasits endpoint based on the optical density of formed clot.The PT reagent (HemosIL RecombiPlastin 2G, BeckmanCoulter, Miami, FL) was used according to manufacturer’sinstructions with an international sensitivity index ofapproximately 1.0 (geometric mean, 11.6 sec). Factor sen-sitivity analyses for this reagent were performed by ourcoagulation laboratory. The following factor sensitivitiesfor PT prolongation were observed: Factor (F)II, 30% to40%; FV, 40% to 50%; FVII, 40% to 50%; and FX, 40% to50%. A F IX sensitivity of 30% to 40% was observed for ourpartial thromboplastin time reagent.

At our institution, stat coagulation specimens are pri-oritized over routine samples, and rapid turnaround timesare achieved mainly by shortening labeling and spinningtimes (barcode-labeled tubes are used and placed in statspinner with tops off at 7200 rpm for only 2 minutes,instead of at 3500 rpm for 10 minutes used for routinesamples). Our laboratory INR turnaround time is only 20minutes.

TEGThe TEG assay was performed in our coagulation labora-tory by trained medical laboratory technologists. Beforestudy commencement, Haemoscope, Haemonetics Corp.(Niles, IL), the TEG manufacturer, was responsible fortraining our personnel for the study. Refreshment coursesand new training sessions were conducted as neededthroughout the study period.

Blood samples were collected on tubes containingcitrate to prevent the clotting process from starting imme-diately after venipuncture because any delays in transfer-ring the samples to the TEG machine would affect theprecision of the R variable measurement. As recom-mended by the manufacturer, samples were held for 45minutes before commencing the TEG assay. A hemostasisanalyzer (TEG 5000, Haemoscope, Haemonetics Corp.)was used for this assay. Its principles have been exten-sively described.20,21,25 Briefly, 500 mL of citrated samples ofwhole blood was transferred to a vial containing bufferedstabilizers and phospholipids (kaolin); 340 mL was thentransferred to a 37°C prewarmed disposable cup contain-ing 20 mL of calcium chloride and the measurement con-tinued for no less than 45 minutes to complete the wholeTEG exam.

Clotting factors assayBlood samples were centrifuged (1700 ¥ g for 15 min at4°C); plasma was separated and centrifuged for an addi-

tional 5 minutes to ensure PLT-free plasma (<10 ¥ 109/L).“Double”-spun plasma was frozen at -70°C until analysisby the Hemostasis Reference Laboratory of the McMasterUniversity in Hamilton, Ontario, Canada. FII, FVII, F IX,and FX activity assays were performed by mixing plasmawith controls deficient in these factors (Precision BioLog-ics, Dartmouth, Nova Scotia, Canada). Normal plasmashould have an activity of 100% or 1.0 U/mL. The degree ofcorrection of the PT (Dade-Behring Innovin, Deerfield, IL)is proportional to the level of the factor in the plasma. Thefactor activity data were expressed in percentages.

Study outcomesThe main study outcome was the ability of TEG-R and INRto diagnose VitK-CF deficiency in trauma, compared tothe gold standard coagulation factor assays. For thisoutcome, the specificity, sensitivity, positive predictivevalue (PPV), and negative predictive value (NPV) of INRand TEG to diagnose VitK-CF deficits were calculated. Sec-ondary outcome included time to first INR and TEGresults (from admission to release of results) and turn-around times for INR results.

Statistical analysisDemographics and baseline characteristics were pre-sented as means and standard deviations or medians andranges when appropriate. Categorical variables were sum-marized as frequency and percentages.

Sensitivity, specificity, PPV, and NPV of INR andTEG-R to detect VitK-CF deficiency were calculated withexact 95% binominal confidence intervals (CIs) to allowcomparison between both tests. For the turnaround timeanalysis, paired t test was used. Two-tailed tests wereapplied and significance was defined at p values of lessthan 0.05. Statistical analysis was performed using com-puter software (SPSS 15.0, SPSS, Inc., Chicago, IL).

RESULTS

TEG was performed on 628 adult trauma patients over 9months in 2007. A total of 219 patients had INR, TEG, andVitK-CF measured on arrival and were included in thisanalysis. Of 219 patients, 29 (13%) and 19 (9%) had INRvalues of 1.3 and 1.5 or greater, respectively; 24 (11%) hadVitK-CF of 50% or less and 14 (6%) had TEG-R of morethan 8 minutes. The demographics and baseline charac-teristics for the whole cohort and for patients with abnor-mal coagulation tests are described (Table 1). Except forthe patients on warfarin, patients with abnormal coagula-tion tests had higher ISSs, and a higher proportion ofpatients sustaining severe head injury (head-abbreviatedinjury severity � 3) were more acidotic and received moretransfusions than patients with normal coagulation

VitK-CF DEFICIENCY BY INR AND TEG


factors (VitK-CF > 50%). In contrast to the patients withabnormal coagulation not on warfarin, patients on war-farin were older, suffered only severe blunt injuries withlower ISS, were less acidotic, and received fewer transfu-sions (Table 1). The individual factor activities for allVitK-CF are reported in Table 2. Lower activities of FII andFX were seen in patients on warfarin.

Overall, TEG-R performed worse than INR withrespect to the ability to detect VitK-CF deficiency, regard-less of the INR cutoff adopted (Tables 3-5). The INR had ahigh specificity and a high NPV for VitK-CF deficiency.

Eight patients were taking warfarin during the studyperiod before injury. They were older and mostly femalewith blunt trauma and less severe injuries (Table 1). Theyall had an abnormal INR on admission to the hospital(2.3 � 4). Six of eight patients on warfarin had at least aVitK-CF deficit. Two patients on warfarin had receivedvitamin K and fresh-frozen plasma by a transferring facil-ity and did not have a critically low VitK-CF but still pre-sented with abnormal INRs. TEG-R was abnormal in onlyone patient anticoagulated by warfarin.

For the patients on warfarin, the time to INR results(from the admission time to the release of results) was onaverage shorter than the time to TEG completion (fromadmission to completion of the whole assay), 58 minutes(�23 min) versus 92 minutes (�40 min) but it did notreach significance (p = 0.07). The laboratory INR turn-around time (from arrival in the laboratory to release ofresults) for patients on warfarin was 20 minutes(�10 min). Had TEG-R been used clinically, its turn-around time would have been a 45-minute wait time plus10 minutes for completion of the R portion, thus 55minutes.

DISCUSSION

Our study found that INR was substantially better thanTEG-R in ruling out VitK-CF deficiency in trauma. Thisobservation was valid for both a more conservative and amore liberal cutoff of INR, 1.3 and 1.5, respectively. Inaddition, INR had similar ability to rule in VitK-CF defi-ciency of TEG-R. For the clinician treating major bleedingthat can have devastating consequences, this analysis sug-gests that INR, due to its higher sensitivity, is a more reli-able test than TEG-R to prevent missing any case ofVitK-CF abnormality. TEG-R was abnormal in only onepatient (out of eight) anticoagulated by warfarin. Finally,our findings also suggested that, when TEG-R is per-formed using citrated blood, the potential advantage offaster turnaround time is lost.

TEG is an appealing test for trauma. It provides a real-time functional evaluation of the whole coagulationprocess including the intrinsic, extrinsic, and commonpathways as well as the initial PLT-fibrin interaction, PLTaggregation, clot strengthening, fibrin cross-linkage, and

TAB

LE

1.D

emo

gra

ph

ics,

adm

issi

on

lab

ora

tory

valu

es,

and

tran

sfu

sio

nd

ata

for

the

enti

reco

ho

rtan

dp

atie

nts

wit

hab

no

rmal

dif

fere

nt

coag

ula

tio

nte

sts*

Var

iabl

esTo

talc

ohor

t(n

=21

9)IN

R�

1.3

(n=

29)

VitK

-CF

�50

%(n

=24

)T

EG

-R>

8m

in(n

=17

)P

atie

nts

onw

arfa

rin(n

=8)

Age

(yea

rs)

39(�

18)

44(�

21)

43(�

22)

33(�

16)

72(�

11)

Mal

e(%

)15

4(6

9)19

(65)

17(7

1)13

(76)

2(2

5)P

enet

ratin

gtr

aum

a(%

)32

(15)

4(1

4)3

(12)

2(1

2)0

ISS

26(�

12)

35(�

14)

34(�

15)

35(�

15)

25(�

15)

Sev

ere

head

inju

ry(%

)96

(44)

19(6

5)15

(62)

11(6

5)5

(62)

INR

1.1

(1-1

.2)

1.7

(1.3

-2)

1.7

(1.3

-2.2

)1.

6(1

.1-2

.1)

1.98

(1.8

-2.7

)P

LTs

(¥10

9 /L)

245

(�24

5)16

0(�

88)

163

(�10

6)15

4(�

96)

260

(�97

)F

ibrin

ogen

(g/L

)2.

5(�

0.9)

1.4

(�1.

2)1.

7(�

1.3)

1.5

(�1.

2)3.

7(�

1.4)

pH7.

3(�

0.14

)7.

14(�

0.2)

7.1

(�0.

2)7.

1(�

0.2)

7.2

(�0.

2)P

reho

spita

lcry

stal

loid

(mL)

600

(155

-130

0)10

50(5

75-2

050)

950

(550

-160

0)10

00(5

50-1

600)

1000

(625

-115

0)24

-hr

red

bloo

dce

ll(u

nits

)4

(2-1

0)11

(5-2

6)13

(5-3

0)9

(5-1

3)3

(2.5

-3.5

)24

-hr

froz

enpl

asm

a(u

nits

)6

(6-4

0)8

(4-9

)8

(4-1

2)6

(5-9

)4

(3.5

-5)

24-h

rP

LTs

(uni

ts)

9(5

-15)

10(5

-20)

15(5

-22)

9(5

-17)

5(5

-5)

24-h

rcr

yopr

ecip

itate

(uni

ts)

14(8

-19)

17(9

-36)

17(8

-42)

13(8

-18)

0

*V

alue

sar

eex

pres

sed

asm

ean

(�S

D)

and

med

ian

(inte

rqua

rtile

rang

es)

whe

nap

prop

riate

.N

umbe

rof

mis

sing

varia

bles

per

grou

p:P

LTs

(tot

alco

hort

,11

;IN

R�

1.3,

3;V

itK�

50%

,1)

;fib

rinog

en(t

otal

coho

rt,

12;

INR

�1.

3,10

;V

itK�

50%

,3;

TE

G-R

>8

min

,2;

patie

nts

onw

arfa

rin,

2);

and

pH(t

otal

coho

rt,

106;

INR

�1.

3,10

;V

itK�

50%

,7;

TE

G-R

>8

min

,4)

.

NASCIMENTO ET AL.


fibrinolysis.19,26 It is a simple test, easy to perform andinterpret, may be done at the bedside, and may guidemanagement including blood transfusion.27,28 Some of theTEG limitations, however, were apparent in this study.While each TEG variable is often described as being deter-mined by specific elements or portions of the coagulationprocess, in fact multiple factors cooperate and may affectthe measurements. For example TEG-R was of specialinterest since it assesses the early phases of coagulation,which mainly involves clotting factors. However, clottingfactors are not the sole determinant of R and all elementsof the initial fibrin formation may also affect it.

TEG results could potentially be available earlier thanfor INR. The time to result for TEG could be shorter if only

the R variable was measured (8 min) using fresh (not anti-coagulated) whole blood. The concern then is that done inthis way, TEG would underestimate the real R time sincecoagulation starts at the moment of venipuncture and notwhen the blood sample is placed on the TEG machine andthe recording initiated. To avoid this error, for the presentstudy the blood samples were anticoagulated at the timeof sampling, which added another 45-minute wait beforethe TEG test could be initiated as recommended by themanufacturer. Modifications such as substituting kaolinfor tissue factor (i.e., rapid TEG) reduce the time to resultby eliminating the R variable measurement.16 A recentstudy reported that the results for rapid TEG were avail-able in 19 minutes, for TEG using kaolin and nonantico-

TABLE 2. VitK-CF activity (%) for the entire cohort and patients with abnormal different coagulation tests*Coagulationfactor Total cohort (n = 219) INR � 1.3 (n = 29) VitK-CF � 50 (n = 24)

TEG-R >8 minutes (n = 17) Patients on warfarin (n = 8)

II 80 (67-92) 49 (30-56) 42 (27-49) 55 (41-82) 31 (20-44)VII 96 (77-118) 61 (44-81) 52 (33-69) 85 (66-104) 36 (26-53)IX 109 (90-126) 62 (37-94) 51 (34-71) 76 (47-110) 62 (45-86)X 88 (72-105) 49 (30-61) 38 (23-53) 61 (49-73) 27 (21-33)

* Values are expressed as median (interquartile range). VitK-CF of 50% or less group = defined as patients with any VitK-CF (II, VII, IX, X) of50% or less.

TABLE 3. Ability of INR of 1.3 or greater to detect VitK-CF deficiency in trauma*INR cutoffs VitK-CF � 50% VitK-CF > 50% Total

INR � 1.3 18 11 29INR < 1.3 6 184 190Total 24 195 219Sensitivity, 75%

(95% CI, 53%-90%)Specificity, 94% (95% CI, 90%-97%) PPV, 62% (95% CI, 42%-79%) NPV, 97% (95% CI, 93%-99%)

* VitK-CF deficiency defined by any factor of 50% or less.

TABLE 4. Ability of INR of 1.5 or greater to detect VitK-CF deficiency in trauma*INR cutoffs VitK-CF � 50% VitK-CF > 50% Total

INR � 1.5 16 3 19INR < 1.5 8 192 200Total 24 195 219Sensitivity, 67%

(95% CI, 45%-84%)Specificity, 98% (95% CI, 96%-99.7%) PPV, 84% (95% CI, 60%-97%) NPV, 96% (95% CI, 92%-98%)


TABLE 5. Ability of TEG-R to detect VitK-CF deficiency in trauma*TEG-R cutoffs VitK-CF � 50% VitK-CF > 50% Total

TEG-R > 8 minutes 8 9 17Normal TEG-R 16 186 202Total 24 195 219Sensitivity, 33%

(95% CI, 16%-55%)Specificity, 95% (95% CI, 91%-98%) PPV, 47% (95% CI, 23%-72%) NPV, 92% (95% CI, 87%-95%)




agulated samples in 29.9 minutes while for INR it took 34minutes.16 Thus the time differences between those twoTEG methods (rapid TEG and kaolin-nonanticoagulatedTEG) and INR were of 15 and 4 minutes, respectively.16

Furthermore, in our experience, the R time varies widely(3.8-9.8 min), even when healthy volunteers were tested.26

Therefore, normal range values for the test might varyeven more than what is assumed by the manufacturers. Rvariable is thus the most susceptible to technical errorsand for this reason it is occasionally ignored.16

PT or INR was designed for and continues to be theprimary method for monitoring patients on warfarin andmanaging its anticoagulation effects. INR assesses theintegrity of the extrinsic and common pathways (FII, FV,FVII, and FX) through measurement of clotting time ofrecalcified plasma in the presence of tissue thromboplas-tin. Despite its many limitations, INR and activated partialthromboplastin time continue to be extensively used asthe standard test to diagnose and guide management oftraumatic coagulopathy.24 Some important limitationsinclude its inability to account for hypothermia (samplesare warmed during processing), lack of association withbleeding, or need for blood transfusion; it may fail todetect significant hemostatic defects and long turnaroundtimes in some centers.14-17

INR is a routine laboratory test done worldwide forthe diagnosis of warfarin toxicity, at a relatively low costand according to established standards while TEG is rela-tively unknown and is not standard equipment in mostlaboratories in the world. Thus, for the limited indicationof evaluating trauma patients on warfarin or harboring acoagulopathy due to deficits of VitK-CF, both tests can beused, but INR alone is sufficient and arguably better.Similar to our results, INR of 1.5 or more was found to bea good test (sensitivity, 84%; specificity, 88%; PPV, 89%;and NPV, 82%) to rule out or in multiple mild VitK-CFdeficits in a small cohort of trauma patients.23

By reviewing the data of a large cohort of injuredpatients enrolled in a prospective observational study atour trauma center, where a sophisticate coagulation panel(TEG and coagulation factor activity) was performed, wehad a unique opportunity to compare INR versus TEG-Ragainst coagulation factor assays performed withinminutes of the patient’s arrival to hospital. The coagul-opathy of trauma is complex and poorly understood. Mul-tiple causes have traditionally been implicated in thistrauma coagulopathy, including dilution, hypothermia,acidosis, and clotting factor consumption. In addition,shock and tissue destruction have been lately suggested tobe involved with the development of an early coagulopa-thy that is independent of dilution.24 Finally, the increas-ing use of warfarin might adversely contribute to the acutecoagulopathy of trauma in some patients at the presenta-tion to trauma centers.2,10 Therefore, one of our studystrengths was the opportunity to study this early acute

trauma coagulopathy in such a complex and uniquecohort of patients within minutes of their admission tohospital.

Our study has important limitations. First, due to thesmall numbers of cases for the sensitivity and PPV calcu-lations, the CIs demonstrated a possibility of wide varia-tion of the test ability in detecting VitK-CF deficiency.However, the limitation applies to both tests (INR andTEG) and TEG-R had clearly a worse performance thanINR. Second, due to the known low prevalence of traumapatients on warfarin, the sole analysis of these patients inour cohort (eight patients anticoagulated by warfarin)would be significantly limited. Third, the original studydesign did not allow us to retrieve information on theturnaround time for TEG-R only. It would certainly bevaluable to compare the laboratory turnaround time forthis part of the TEG test to the INR. Nevertheless, we wereable to present the time taken from admission to INRresults or completion of the whole TEG tests as well as theturnaround time for INR (from the time the samplearrived in the laboratory to release of results). Contrarily tolonger INR turnaround times in previous studies, ourlaboratory INR turnaround time is only 20 minutes.Finally, our study did not assess the outcomes associatedwith the use of INR and TEG-R with respect to INR correc-tion times, bleeding, and mortality.

Our study suggests that TEG-R variable is not superiorto identify patients with coagulopathy due to VitK-CFdeficit compared to INR in trauma. The TEG assay usingcitrated blood might produce longer turnaround timethan INR. Considering the widespread use of INR, its cost,its easy interpretation, and its time to obtain the results, itappears that TEG-R does not offer an advantage over INRwhen dealing with this cohort of trauma patients. Thus, inthe acute trauma setting, INR remains a useful and reli-able indicator of trauma-induced coagulopathy associ-ated with reduced levels of VitK-CF.

ACKNOWLEDGMENTS

The authors thank Dr Marciano Reis, Clinical Pathology Depart-

mental Chairman, for the laboratory support to conduct the

observational study on thromboelastography. The authors also

thank Cyndy Rogers and Bill Sharkey for their contribution in

providing trauma registry data. Finally, we are in debt with Dr Rita

Selby, Helena Brnjac, and Hugh Hoogendoorn for their technical

assistance with coagulation assay descriptions.

CONFLICT OF INTEREST

SR has received honorarium and speaker’s fees (as a member of

the Scientific Advisory Board) from NovoNordisk S/A, manufac-

turer of NovoSeven (recombinant factor VII). The other authors

have no conflict of interest relevant to the subject matter of this

publication.

NASCIMENTO ET AL.


REFERENCES

1. Sauaia A, Moore FA, Moore EE, Moser KS, Brennan R, Read

RA, Pons PT. Epidemiology of trauma deaths: a reassess-

ment. J Trauma 1995;38:185-93.

2. Mina AA, Bair HA, Howells GA, Bendick PJ. Complications

of preinjury warfarin use in the trauma patient. J Trauma

2003;54:842-7.

3. Wysowski DK, Nourjah P, Swartz L. Bleeding complications

with warfarin use: a prevalent adverse effect resulting in

regulatory action. Arch Intern Med 2007;167:1414-9.

4. Freedman MD. Oral anticoagulants: pharmacodynamics,

clinical indications and adverse effects. J Clin Pharmacol

1992;32:196-209.

5. Ansell J, Hirsh J, Hylek E, Jacobson A, Crowther M, Palareti

G. Pharmacology and management of the vitamin K

antagonists: American College of Chest Physicians

evidence-based clinical practice guidelines (8th edition).

Chest 2008;133(6 Suppl):160S-98S.

6. Ivascu FA, Janczyk RJ, Junn FS, Bair HA, Bendick PJ, Howells

GA. Treatment of trauma patients with intracranial hemor-

rhage on preinjury warfarin. J Trauma 2006;61:318-21.

7. Palareti G, Cosmi B. Bleeding with anticoagulation

therapy—who is at risk, and how best to identify such

patients. Thromb Haemost 2009;102:268-78.

8. Dossett LA, Riesel JN, Griffin MR, Cotton BA. Prevalence

and implications of preinjury warfarin use: an analysis of

the national trauma databank. Arch Surg 2011;146:565-70.

9. Mountain D, Sistenich V, Jacobs IG. Characteristics, man-

agement and outcomes of adults with major trauma taking

pre-injury warfarin in a Western Australian population

from 2000 to 2005: a population-based cohort study. Med J

Aust 2010;193:202-6.

10. Howard JL, Cipolle MD, Horvat SA, Sabella VM, Reed JF III,

Fulda G, Tinkoff G, Pasquale MD. Preinjury warfarin

worsens outcome in elderly patients who fall from stand-

ing. J Trauma 2009;66:1518-22.

11. Ivascu FA, Howells GA, Junn FS, Bair HA, Bendick PJ,

Janczyk RJ. Rapid warfarin reversal in anticoagulated

patients with traumatic intracranial hemorrhage reduces

hemorrhage progression and mortality. J Trauma 2005;59:

1131-7.

12. Fihn SD, Gadisseur AA, Pasterkamp E, van der Meer FJ,

Breukink-Engbers WG, Geven-Boere LM, van Meegen E, de

Vries-Goldschmeding H, Antheunissen-Anneveld I, van’t

Hoff AR, Harderman D, Smink M, Rosendaal FR. Compari-

son of control and stability of oral anticoagulant therapy

using acenocoumarol versus phenprocoumon. Thromb

Haemost 2003;90:260-6.

13. Quick AJ. [New concepts on blood coagulation and hemor-

rhagic diseases.] Resen Clin Cient 1952;21:219-24.

14. Mintz PD. Outstanding scientist of the year 2005: William

Spotnitz, gifted scientist, inventor, and cardiothoracic and

vascular surgeon. J Long Term Eff Med Implants 2005;15:

339-46.

15. O’Connor SD, Taylor AJ, Williams EC, Winter TC. Coagula-

tion concepts update. AJR Am J Roentgenol 2009;193:1656-

64.

16. Frankel HL, Rozycki GS, Ochsner MG, McCabe JE, Harviel

JD, Jeng JC, Champion HR. Minimizing admission labora-

tory testing in trauma patients: use of a microanalyzer.

J Trauma 1994;37:728-36.

17. Jeger V, Zimmermann H, Exadaktylos AK. Can RapidTEG

accelerate the search for coagulopathies in the patient with

multiple injuries? J Trauma 2009;66:1253-7.

18. Carroll RC, Craft RM, Langdon RJ, Clanton CR, Snider CC,

Wellons DD, Dakin PA, Lawson CM, Enderson BL, Kurek

SJ. Early evaluation of acute traumatic coagulopathy by

thrombelastography. Transl Res 2009;154:34-9.

19. Schochl H, Nienaber U, Hofer G, Voelckel W, Jambor C,

Scharbert G, Kozek-Langenecker S, Solomon C. Goal-

directed coagulation management of major trauma

patients using thromboelastometry (ROTEM)-guided

administration of fibrinogen concentrate and prothrombin

complex concentrate. Crit Care 2010;14:R55.

20. Mallett SV, Cox DJ. Thrombelastography. Br J Anaesth

1992;69:307-13.

21. Chandler WL. The thromboelastography and the throm-

boelastograph technique. Semin Thromb Hemost 1995;21

Suppl 4:1-6.

22. Burns ER, Goldberg SN, Wenz B. Paradoxic effect of mul-

tiple mild coagulation factor deficiencies on the prothrom-

bin time and activated partial thromboplastin time. Am J

Clin Pathol 1993;100:94-8.

23. Yuan S, Ferrell C, Chandler WL. Comparing the prothrom-

bin time INR versus the APTT to evaluate the coagulopathy

of acute trauma. Thromb Res 2007;120:29-37.

24. Nascimento B, Callum J, Rubenfeld G, Neto JB, Lin Y,

Rizoli S. Clinical review: fresh frozen plasma in massive

bleedings—more questions than answers. Crit Care 2010;

14:202.

25. Kheirabadi BS, Crissey JM, Deguzman R, Holcomb JB. In

vivo bleeding time and in vitro thrombelastography mea-

surements are better indicators of dilutional hypothermic

coagulopathy than prothrombin time. J Trauma 2007;62:

1352-9.

26. Scarpelini S, Rhind SG, Nascimento B, Tien H, Shek PN,

Peng HT, Huang H, Pinto R, Speers V, Reis M, Rizoli SB.

Normal range values for thromboelastography in

healthy adult volunteers. Braz J Med Biol Res 2009;42:

1210-7.

27. Spiess BD, Gillies BS, Chandler W, Verrier E. Changes in

transfusion therapy and reexploration rate after institution

of a blood management program in cardiac surgical

patients. J Cardiothorac Vasc Anesth 1995;9:168-73.

28. Spalding GJ, Hartrumpf M, Sierig T, Oesberg N, Kirschke

CG, Albes JM. Cost reduction of perioperative coagulation

management in cardiac surgery: value of “bedside” throm-

belastography (ROTEM). Eur J Cardiothorac Surg

2007;31:1052-7.



vitamin k-dependent coagulation factor deficiency in trauma: a comparative analysis between...

Documents