Hemostatic Abnormalities in Patients with Closed Head Injuries and Their Role in Predicting Early Mortality

Abstract

Abnormalities of blood coagulation are frequently found in patients following traumatic brain injury. Exposure to thromboplastin, which is abundant in brain, plays an important role in initiating coagulopathy. Eighty patients of moderate-to-severe head injury were screened for platelet count, prothombin time (PT), activated partial thromboplastin time (a-PPTK), fibrinogen degradation product levels (FDP), D-dimer levels, and disseminated intravascular coagulation scores (DIC), calculated within the first 24 hours of injury. Increased consumptive coagulopathy at admission, as reflected by high DIC scores, predicted mortality in both moderate and severe head injury patients with a high degree of accuracy (p < 0.001). Similarly, increased PT, FDP, and D-dimer values correlated with higher mortality in both groups, but platelet counts and a-PPTK values correlated with mortality only in the severe head injury group. From this study we conclude that hemostatic abnormalities are independent predictors of early mortality in moderate-to-severe head injury patients.

Introduction

O ne of the major causes of mortality in patients of head injury is secondary insult to the brain. In the quest for elucidating factors responsible for secondary insult to brain, Goodnight and colleagues (1974) hypothesized that tissue thromboplastin is released from damaged brain. This action activates the extrinsic coagulation pathway, thereby producing a fibrin clot, as subsequently confirmed by several authors (Keimowitz and Annis, 1973; Vecht et al., 1975; van der Sande et al., 1978; Ueda et al., 1985; Olson et al., 1989; Enderson et al., 1991; Kerney et al., 1992; Hulka et al., 1996). In this intravascular phenomenon, fibrinolytic activity matches that of thrombus formation, making consumption of these coagulation factors, rather than microthrombosis, a primary event (Goodnight et al., 1974; van der Sande et al., 1978; Hoots, 1996; Hulka et al., 1996; Gando et al., 1999; Carrick et al., 2005; Halpern et al., 2008). Frequency of this event is higher in patients with penetrating injury than in those with closed head injury (Goodnight et al., 1974; Vecht et al., 1975; Ueda et al., 1985; Kerney et al., 1992; Hulka et al., 1996).

Increased risk of mortality is recognized in patients of head injury with hemostatic abnormalities (Vecht et al., 1975; Olson et al., 1989; Hulka et al., 1996). Several authors, including Goodnight and colleagues (1974), have suggested that reversal of these coagulation defects can occur with transfusion of cryoprecipitate or fresh frozen plasma, and this approach may be beneficial in selected groups of patients (Goodnight et al., 1974; Farringer et al., 1993). The purpose of this study was to examine various hemostatic abnormalities in patients of head injury and their role in predicting early mortality.

Materials and Methods

Patient selection

Of all the patients with moderate-to-severe head injuries admitted in our Department of Neurosurgery between June 2007 and March 2008, 80 patients, satisfying the following criteria, were included in this study (Table 1):

Patients with only closed head injuries;

Not taking anticoagulants or any such medication that interferes with the results of laboratory tests;

Not having hypertension, diabetes, or chronic illness, which adversely effect the outcome in patients of head injury;

Not harboring any other injury (like fractured long bones or fat embolism), which interferes with the results of the laboratory tests;

Not having significant blood loss requiring transfusion, thereby affecting the laboratory test results (Gando et al., 1999).

Table 1.

Distribution of Patients Based on GCS Score

	Expired (GOS 1)	Vegetative (GOS 2)	Discharged (GOS 3–5)
Moderate head injury patients, n = 30 (GCS 9–13)	6	4	20
Severe head injury patients, n = 50 (GCS_ < 8)	28	10	12

Total number of patients, 80.

DIC screen

All selected patients were subjected to a disseminated intravascular coagulation (DIC) screen, as depicted in Table 2, within 24 h of their injury. Each test value was awarded a score of 0 to 3, depending upon severity of derangement of the test; a final DIC score was calculated based on the individual scores. Thus a DIC score ranging from 0 to 15 was calculated individually for each patient (Table 2).

Table 2.

DIC Screen Components, Values, and Scoring System

	Platelet count LAKHS/CUMM.	PT (s)	a-PTTK (s)	FDP (μg/mL)	D-dimer (μg/dL)	DIC score (0–3)
Normal	>1.5	13.5	26–34	<10	<10000	0
Mild derangement	1–1.5	13.5–15	>34, <26	10–20	1000–2000	1
Moderate derangement	0.60–1	15–18	>39, <22	20–40	2000–4000	2
Severe derangement	<0.60	>18	>54	>40	>4000	3

Overall DIC score was calculated as sum of individual scores of each of the tests of DIC screen. Overall score of 0–3 was taken to be normal; score 4–6 was considered mild derangement; score 7–10 was moderate derangement; and score >10 was severe derangement of overall DIC scoring.

All patients were divided into two groups: moderate head injury with GCS 9–13, and severe head injury with GCS <8, based on first post-resuscitation evaluation of GCS score of the patient. Outcome in each group was measured as discharged with GOS 3–5, discharged in vegetative dependent state GOS 2, or expired (Table 1). The DIC scores of expired (GOS 1) and those discharged in vegetative state (GOS 2) were compared with those of discharged with GOS 3–5 in both head injury groups, individually as well as in combined form (Tables 3 –6).

Table 3.

Comparison of Total DIC Scores and Individual Lab Tests of Expired and Discharged Patients in Severe Head Injury Group

	Mean ± Sd
Score	Expired (GOS 1) (n = 28)	Discharge (GOS 3–5) (n = 12)	P value	Significance
DIC	8.78 ± 1.29	4.83 ± 1.07	<0.001	HS
Pt. count	1.0 ± 0.75	0.33 ± 0.74	<0.001	HS
PT	2.28 ± 0.59	1.33 ± 0.47	<0.001	HS
PTTK	1.0 ± 0.75	1.67 ± 0.47	<0.001	HS
FDP	2.00 ± 0.00	0.83 ± 0.37	<0.001	HS
D-dimer	2.43 ± 0.49	0.67 ± 0.47	<0.001	HS

Table 4.

Comparison of Total DIC scores and Individual Lab Tests of Expired (GOS 1) and Discharged (GOS 3–5) Patients in Moderate Head Injury Group

	Mean ± Sd
Score	Expired (GOS 1) (n = 6)	Discharge (GOS 3–5) (n = 20)	P value	Significance
DIC	7.67 ± 0.47	3.00 ± 1.18	<0.001	HS
Pt. count	0.33 ± 0.47	0.40 ± 0.49	>0.05	NS
PT	1.67 ± 0.47	1.00 ± 0.45	<0.02	Sig
PTTK	0.67 ± 0.47	0.70 ± 0.78	>0.05	NS
FDP	2.33 ± 0.47	0.50 ± 0.50	<0.001	HS
D-dimer	2.67 ± 0.47	0.40 ± 0.49	<0.001	HS

Table 5.

Comparison of Total DIC Scores and Individual Lab Tests of Expired (GOS 1) and Discharged (GOS 3–5) Patients in Both Groups Combined

	Mean ± Sd
Score	Expired (n = 34)	Discharged (n = 32)	P value	Significance
DIC	8.59 ± 1.26	3.69 ± 1.44	<0.001	HS
Pt. count	0.88 ± 0.76	0.37 ± 0.60	<0.001	HS
PT	2.17 ± 0.62	1.12 ± 0.48	<0.001	HS
PTTK	0.94 ± 0.72	1.06 ± 0.83	<0.05	Sig
FDP	2.06 ± 0.23	0.62 ± 0.48	<0.001	HS
D-dimer	2.47 ± 0.50	0.50 ± 0.50	<0.001	HS

Table 6.

Comparison of Total DIC Scores and Individual Lab Tests of Vegetative and Discharged (GOS 3–5) Patients in Both Groups Combined

	Mean ± Sd
Score	Veg. state (GOS 2) (n = 14)	Discharged (GOS 3–5) (n = 32)	P value	Significance
DIC	8.00 ± 2.13	3.69 ± 1.44	<0.001	HS
Pt. count	0.71 ± 0.79	0.37 ± 0.60	<0.001	HS
PT	1.78 ± 0.77	1.12 ± 0.48	<0.001	HS
PTTK	1.86 ± 0.83	1.06 ± 0.83	<0.001	HS
FDP	1.71 ± 0.45	0.62 ± 0.48	<0.001	HS
D-dimer	1.78 ± 0.41	0.50 ± 0.50	<0.001	HS

Statistical analysis

The data were analyzed using Z test and t-test.

Results

Of the 80 patients included in this study, 50 patients had severe head injury, of which 20 expired (GOS 1), 12 were discharged with GOS 3–5, and 10 were finally discharged with GOS 2 (Table 1).

The mean DIC score in those discharged (GOS 3–5) was 4.83 ± 1.07, and, of those expired, was 8.78 ± 1.29. On analyzing this result using t-test, this difference was found to be highly significant (p < 0.001). Similarly, when individual scores of each of the laboratory tests of DIC screen in both the groups were compared, the differences of the individual laboratory scores were found to be independently significant (p < 0.001 for all the tests) (Table 3).

Thirty patients in this study had moderate head injury. Of these, 20 were discharged with GOS 3–5, four were discharged with GOS 2, and six patients expired (Table 1). The mean DIC score of those discharged with GOS 3–5 was 3.00 ± 1.18, and of those expired (GOS 1) was 7.67 ± 0.47. On analyzing these data using z-test, the difference between the two was found to be statistically significant (p < 0.001). On comparing the individual laboratory test scores of the DIC screen of the two groups, the difference was highly significant for FDP and D-dimer levels (p < 0.001), significant for PT values (p < 0.02), and not significant at all for platelet count and a-PTTK values (Table 4). Of the 80 patients included in this study, 34 expired, 14 were discharged with GOS 2, and 32 were discharged with GOS 3–5. The mean DIC score in expired patients was 8.59; in those with GOS 2, the mean score was 8, and in those discharged with GOS 3–5, it was 3.69. The difference was statistically significant between the expired and those discharged with GOS 3–5 (p < 0.001), as well as between vegetative (GOS 2) and those discharged with GOS 3–5 (p < 0.001). Similarly the individual scores of laboratory tests of DIC screen were compared, and results are given in Tables 5 and 6.

Discussion

During the course of this study spanning over a period of 10 months, patients were carefully selected after strictly applying the inclusion and exclusion criteria. Of the 80 patients included in this study, 34 had moderate derangement of their DIC scores (DIC score 7–9); of these patients, 26 expired and eight were discharged with GOS 2. Twelve patients had severe derangement of their DIC scores, out of which eight expired and four were discharged with GOS 2.

Thirty-four patients had mild derangements of their DIC scores (4–6), of which 32 were discharged with GOS 3–5 state, two were discharged with GOS 2 (Table 1). Thus the mean DIC score of the patients who expired was 8.52; of those who were discharged with GOS 2, it was 8; and of those discharged with GOS 3–5 it was 3.88 (Tables 5 and 6). This difference when analyzed was found to be statistically significant between expired and those discharged with GOS 3–5 (p < 0.001), as well as with those with GOS 2 and those with GOS 3–5 (p < 0.001).

Blunt injury activates the coagulation process, as postulated by Goodnight et al. in 1974. This process is severe only in patients in the expired group and those discharged with GOS 2 compared to those discharged with GOS 3–5 in our study, as evidenced by higher DIC scores in the first two groups. This difference was statistically significant when taken together for all 80 patients, as well as separately for moderate and severe head injury groups (Tables 3 –6).

Thus based on our study we can not only say that patients with higher DIC scores are more likely to have a poor outcome; we can also predict which of the patients in the severe head injury group (which is generally associated with higher mortality and morbidity) are more likely to expire and which are likely to survive because the difference in DIC scores in both subgroups in the severe head injury group is statistically significant (p < 0.001).

Similarly, in the moderate head injury group, six patients expired and 20 were discharged with GOS 3–5, and the difference between their mean DIC scores was statistically significant (p < 0.001). Thus although moderate head injury is traditionally associated with lesser mortality and morbidity, DIC scores can be used to predict those patients with higher mortality in this group.

Both head injury groups had few patients discharged with GOS 2 and their mean DIC scores were higher compared with those discharged with GOS 3–5; the difference was statistically significant in both groups (Table 6).

Thus we can safely conclude that patients with higher DIC scores are more likely to have either higher mortality or vegetative state regardless of whether the patient had moderate or severe head injury, and that patients with lower DIC scores are more likely to have a better outcome irrespective of their GCS scores.

This conclusion has both therapeutic as well as prognostic implications as we can predict which patients in the severe head injury group are likely to survive or have better outcomes and which patients in spite of moderate head injury are likely to have fatal or poor outcomes in the form of persistent vegetative state.

Conclusions

We have drawn the following conclusions based on the findings of our study:

Hemostatic evaluation of patients with head injury can be used to independently predict mortality irrespective of GCS scores.

Patients with higher DIC scores are more likely to have higher mortality regardless of whether they are in moderate or severe head injury groups.

Since DIC scores and other hemostatic abnormalities are associated with higher early mortality, there may be a case for formulating a therapeutic intervention directed at these abnormalities with an aim toward improving the final outcome.

Evaluation of these abnormalities in patients with higher GCS scores is of special importance because patients with these abnormalities are likely to deteriorate and have higher mortality in spite of their better GCS scores and are the ones who require aggressive management of their hemostatic profile in view of improving the outcome.

Of the various components of the DIC screen, only in FDP and D-dimer levels was the difference statistically significant between expired and discharged (GOS 3–5) groups in both moderate and severe head injury patients. Hence their measurement could independently predict mortality.

Footnotes

Author Disclosure Statement

No competing financial interests exists.

References

Carrick

M.M.

, Tyroch

A.H.

, Youens

C.A.

, Handley

2005. Subsequent development of thrombocytopenia and coagulopathy in moderate and severe head injury: support for serial laboratory examination. J. Trauma, 58:725–730.

Enderson

B.L.

, Chen

J.P.

, Robinson

, Maull

K.I.

1991. Fibrinolysis in multisystem trauma patients. J. Trauma, 31:1240–1246.

Faringer

P.D.

, Mullins

R.J.

, Johnson

R.L.

, Trunkey

D.D.

1993. Blood component supplementation during massive transfusion of AS-1 red cells in trauma patients. J. Trauma, 34:481–487.

Gando

, Nanzaki

, Kemmotsu

1999. Disseminated intravascular coagulation and sustained systemic inflammatory response syndrome predict organ dysfunctions after trauma: application of clinical decision analysis. Ann. Surg., 229,1:121–127.

Goodnight

S.H.

, Kenoyer

, Rapaport

S.I.

, Patch

M.J.

, Lee

J.A.

, Kruze

1974. Defibrination after brain tissue destruction. N. Engl. J. Med., 290:1043–1047.

Halpern

C.H.

, Reilly

P.M.

, Turtz

A.R.

, Stein

S.C.

2008. Traumatic coagulopathy: the effect of brain injury. J. Neurotrauma, 25,8:997–1001.

Hoots

W.K.

Narayan

R.K.

, Wilberger

J.E.

, Povlishock

J.T.

1996. Coagulation disorders in the head-injured patient. Neurotrauma. McGraw-Hill: New York, 673–688.

Hulka

, Mullins

R.J.

, Frank

E.H.

1996. Blunt brain injury activates the coagulation process. Arch. Surg., 131:923–928.

Keimowitz

R.M.

, Annis

B.L.

1973. Disseminated intravascular coagulation associated with massive brain injury. J. Neurosurg., 39:178–180.

10.

Kerney

T.J.

, Bentt

, Gorde

, Lee

, Hiatt

J.R.

, Shabot

M.M.

1992. Coagulopathy and catecholamines in severe head injury. J. Trauma, 32:608–612.

11.

Olson

J.D.

, Kaufman

H.H.

, Moake

, Goreman

T.W.O.

, Hoots

, Wagner

, Kice Brown

, Gildenberg

P.L.

1989. The incidence and significance of haemostatic abnormalities in patients with head injuries. Neurosurgery, 24:825–832.

12.

Ueda

, Fujitsu

, Fujino

, Sekino

, Kuwabara

1985. Correlation between plasma fibrin-fibrinogen degradation product values and CT findings in head injury. J. Neurol. Psyichiatry, 48:58–60.

13.

van der Sande

J.J.

, Velkamp

J.J.

, Boekhout-Mussert

R.J.

, Bouwhuis-Hoogerwerf

M.L.

1978. Head injury and coagulation disorders. J. Neurosurg., 49:357–365.

14.

Vecht

C.J.

, Smit Sibinga

C.T.

, Minderhoud

J.M.

1975. Disseminated intravascular coagulation and head injury. J. Neurol. Psychiatry, 38:567–571.