European Journal of Trauma and Emergency Surgery (2021) 47:1517–1525

https://doi.org/10.1007/s00068-020-01460-8

ORIGINAL ARTICLE

Factors with the strongest prognostic value associated

with in‑hospital mortality rate among patients operated for acute
subdural and epidural hematoma
Bartłomiej Kulesza1 · Marek Mazurek1 · Adam Nogalski2 · Radosław Rola1

Received: 13 June 2020 / Accepted: 5 August 2020 / Published online: 10 August 2020
© The Author(s) 2020

Abstract
Introduction Traumatic brain injury (TBI) still remains a serious health problem and is called a “silent epidemic”. Each
year in Europe 262 per 100,000 individuals suffer from TBI. The most common consequence of severe head injuries include
acute subdural (SDH) and epidural hematomas (EDH), which usually require immediate surgically treatment. The aim of
our study is to identify factors which have the strongest prognostic value in relation to in-hospital mortality rate among of
patients undergoing surgery for EDH and SDH.
Patients and methods Cohort included 128 patients with isolated craniocerebral injuries who underwent surgery for EDH
(28 patients) and SDH (100 patients) in a single, tertiary care Department of Neurosurgery. The data were collected on
admission of patients to the Emergency Department and retrospectively analyzed. The following factors were analyzed:
demographic data, physiological parameters, laboratory variables, computed tomography scan characteristics and the time
between trauma and surgery. Likewise, we have investigated the in-hospital mortality of patients at the time of discharge.
Results We found that the factors with the strongest prognostic values were: the initial GCS score, respiratory rate, glycae-
mia, blood saturation, systolic blood pressure, midline shift and type of hematoma. Additionally, we proved that a drop by
one point in the GCS score almost doubles the risk of in-hospital death while the presence of coagulopathy increases the
risk of in-hospital death almost six times.
Conclusion Most of the factors with the strongest prognostic value are factors that the emergency team can treat prior to the
hospital admission. Coagulopathy, however that has the strongest influence on in-hospital death rate can only be efficiently
treated in a hospital setting.

Keywords Traumatic brain injury · Epidural hematoma · Subdural hematoma · In-hospital mortality rate

Introduction
* Bartłomiej Kulesza
[email protected]
Traumatic brain injury (TBI) still remains a serious health
Marek Mazurek
and socioeconomic problem being called a “silent epidemic”
[email protected]
because the effects of trauma are often not immediately vis-
Adam Nogalski
ible [1, 2]. Moreover, TBI usually affects young adults and
[email protected]
results in high mortality or severe disability. Each year in
Radosław Rola
Europe 262 per 100,000 individuals suffer from TBI [3, 4].
[email protected]
The most common consequences of severe head injuries are
1
Chair and Department of Neurosurgery and Paediatric extra-axial hemorrhagic lesions such as an acute subdural
Neurosurgery, Medical University in Lublin, Independent (SDH) and epidural hematoma (EDH), which usually require
Public Clinical Hospital No. 4 in Lublin, Jaczewskiego 8, immediate surgically treatment [4–6]. ‘No head injury is
20‑954 Lublin, Poland
2
too severe to despair of, nor too trivial to ignore’—this the
Chair and Department of Trauma Surgery and Emergency aphorism of Hippocrates shows that estimating the prognosis
Medicine, Medical University in Lublin, Independent
Public Clinical Hospital No. 1 in Lublin Poland, Stanisława after head injury has been and remains difficult [7]. Most of
Sztaszica 16, 20‑400 Lublin, Poland the factors associated with mortality, however were analyzed

13
Vol.:(0123456789)
1518 B. Kulesza et al.

on the basis of all types of TBI, not specifically in patients intraventricular hemorrhage (IVH), cerebral contusion,
with EDH and SDH. The aim of our study is to identify fac- maximum thickness of the hematoma, midline shift (MLS)
tors which have the strongest prognostic value in relation to and state of basal cisterns. The midline shift and maximum
in-hospital mortality rate among of patients who undergo thickness of the hematoma were calculated using the OsiriX
surgery for acute extra-axial hematoma, i.e. EDH and SDH. version 2.5 program based on cross sections from the pre-
operative CT scan. The last factor was the time between
the injury and the surgery. The in-hospitality mortality of
Patients and methods patients was assessed at the time of discharge. Statistical
analysis correlated all of the aforementioned factors with
The study cohort included 128 patients with isolated crani- in-hospital mortality rate.
ocerebral injuries. The patients were divided into two
groups, namely a group of 28 patients operated on due to Statistical analysis
epidural hematoma (EDH group) and a group of 100 oper-
ated on due to acute subdural hematoma (SDH group). All In univariate analysis, the C ­ hi2 homogeneity test was per-
of the patients were operated and treated in the Department formed to detect differences in unrelated qualitative charac-
of Neurosurgery and Pediatric Neurosurgery of the Inde- teristics between groups. In multivariate analysis, the logis-
pendent Public Clinical Hospital No. 4 (IPCH 4) in Lub- tic regression analysis was used to assess the factors with
lin, from 1.10.2014 to 31.08.2017. All patients underwent predominant prognostic value of the in-hospital mortality.
craniotomy and hematoma evacuation. During this period, If applicable, for comparison of two independent variables
162 patients underwent surgery for extra-axial hematoma the Mann–Whitney U test was implemented. A significance
whereas 34 patients were excluded from the study due to: the level of p < 0.05 was assumed indicating the existence of
lack of complete medical documentation, incomplete labora- statistically significant differences. In the logistic regres-
tory tests or the lack of formal description of the computed sion analysis, all the studied factors were considered. For the
tomography of the head and patients undergoing decompres- obtained model, the C ­ hi2 value for the difference between
sion craniectomy (Fig. 1). the current model and the model with only the free expres-
All of the data were collected retrospectively based on the sion was highly statistically significant (p < 0.00001). The
admission data from the Emergency Department (ED). All database and statistical research were based on the STATIS-
of the physiological factors were collected up to 10 min after TICA 13.0 computer software (StatSoft, Poland).
the patient’s arrival in ED. Blood was taken for laboratory
tests and computed tomography of the head was performed
up to 30 min after arrival in ED. The following variables Results
were analyzed: demographic data (gender and age), physi-
ological factors, laboratory variables, computed tomogra- The mean age of patients with EDH was 38.81 ± 13.37 years
phy scan characteristics and the time between trauma and and it was significantly lower than the mean age of patients
surgery. Physiological factors included initial GCS score, in SDH group—57.86 ± 18.26 years (p = 0.00001). Men
pupil reaction to light, saturation, systolic blood pressure were hospitalized most often than women. The patients
(SBP), heart rate (HR) and respiratory rate (RR). Laboratory with SDH had lower GCS scores than those in EDH group
variables included: the number of white blood cells (WBC), (p = 0.004). Death during hospitalization was more frequent
hemoglobin (HGB) value, number of platelets (PLT), gly- in patients with subdural hematoma (45%) when compared
cemia value, sodium concentration, and coagulopathy and to the group with epidural hematoma (10.71%), (p = 0.0009)
alcohol levels. Each patient included in the study had a com- (Table 1). The time to death (in-hospital mortality) was
puted tomography (CT) of the head on admission. The study slightly longer in the EDH group (median 4 days) in com-
aimed at particular characteristics from the CT, such as the parison to the SDH group (median 3 days). The difference
present skull fracture, subarachnoid hemorrhage (SAH), was not statistically significant (p = 0.71), though (Fig. 1).

Evaluation of the factors showing the strongest

28 paents with prognostic value regarding in‑hospital mortality
EDH
128 paents rate
included
100 paents with
162 paents
SDSH The factors that entered the model and had a significant rela-
34 paents
excludes tionship with the mortality rate at the level of p < 0.05 were:
the initial GCS scale, type of hematoma, maximum thick-
Fig. 1  Patients inclusion scheme ness of hematoma, status of basal cisterns and coagulopathy.

13
Factors with the strongest prognostic value associated with in‑hospital mortality rate among… 1519

Table 1  Comparison of factors Factors EDH (n = 28) SDH (n = 100 Statistical analysis
with p < 0.05 in both groups
N % n %

Gender
Female 2 7.14 14 14.00 p = 0.00001
Male 26 92.86 86 86.00
Age
< 35 years old 10 35.71 14 14.00
36–60 years old 17 60.71 42 42.00
> 61 years old 1 3.57 44 44.00
GCS score
3–8 score GCS 9 32.14 60 60.00 p = 0.004
9–12 score GCS 7 25.00 25 25.00
13–15 score GCS 12 42.86 15 15.00
In-hospital mortality 3 10.71 45 45.00 p = 0.0009

p statistical value

The rest of the tested factors with a level of p ≥ 0.05 did not factors of in-hospital mortality, it was shown that the varia-
enter the model. The obtained logistic model is presented bles included in the table and the figures below are important
below: variables for the assessment of death and survival during

e4.834+0,551initial GCS scale−3.279type of hematoma−maximum thickness of hematoma−1.696status of basal cisterns+1.770coagulopathy

P(X)
1 + e4.834+0.551initial GCS scale−3.279type of hematoma−maximum thickness of hematoma−1.696status of basal cisterns+1.770coagulopathy

The model shows that a drop by four points in the GCS hospitalization. The factors with the strongest prognostic
score affects the increased risk of in-hospital death almost value are: the initial GCS score, respiratory rate (below 10
twice (1.73). The presence of epidural hematoma increases or above 25 breaths per min), hyperglycemia (blood glucose
the survival charter by 0.04 compared to patients with sub- level > 110 mg/dl), saturation (oxygen saturation < 96%),
dural hematoma. Similarly, the smaller the maximum thick- systolic blood pressure (below 90 or above 140 mmHg),
ness of hematomas by 10 mm, the death risk is almost one midline shift and type of hematoma (Table 3) and (Fig. 2).
time lower (0.88). Normal size basal cisterns increase the
survival chance by 0.18, while the presence of coagulopathy
Discussion
(INR > 1.2 or PT > 12.7 s) increases the risk of in-hospital
death (almost six times) (5.87). The results obtained are pre-
Demographic data
sented in the table below (Table 2).
As a result of the analysis using Data Mining, the selec-
Observations to date have often indicated the impact of
tion and the elimination of variables to assess the prognostic
demographic factors such as age and gender on the prognosis

Table 2  Logistic model for the Constant GCS scale Type of hematoma Maximum Status Coagulopathy
assessment of prognostic factors thickness of of basal
of in-hospital mortality in both hematoma cisterns
groups
Rating 4.8363 0.5506 − 3.2790 − 0.1305 − 1.6964 1.7697
Standard error 2.3359 0.1161 1.4588 0.0531 0.6379 0.8487
t (122) 2.0704 4.7408 − 2.2478 − 2.4580 − 2.6592 2.0852
p 0.0405 0.0000 0.0264 0.0154 0.0089 0.0392
− 95% CL 0.2118 0.3207 − 6.1670 − 0.2355 − 2.9593 0.0895
+ 95% CL 9.4608 0.7806 − 0.3910 − 0.0254 − 0.4334 3.4500
Chi2 Walda 4.2867 22.4751 5.0526 6.0420 7.0716 4.3480

Statistical analysis: ­Chi2 = 91.53; p < 0.00001

13
1520 B. Kulesza et al.

Table 3  Dominant factors for the dependent variables of the in-hos- coexistence of other diseases in this age group [18–20].
pital mortality However, our observations did not indicate age among the
Factors Chi2 P factors with the strongest predictive value in relation to the
in-hospital mortality rate. Some authors pointed to the exist-
Initial GCS scale 45.05 0.000000
ence of a difference in the prognosis of traumatic patients
RR 27.88 0.000000
depending on the patient’s sex [21]. However, there is strong
Glycaemia 22.50 0.000002
evidence that gender did not affect the prognosis in TBI [11,
Saturation 20.19 0.000007
22–26]. This is also confirmed by the results of our observa-
SBP 13.03 0.0003
tions (Fig. 3).
MLS 25.45 0.0006
Type of hematoma 10.97 0.0009
Physiological factors
Pupil reactive 10.80 0.001
IVH 7.65 0.01
The Glasgow Coma Scale (GCS) is one of the basic tools
SAH 7.07 0.01
used to assess the condition of patients after head injuries.
Status of basal cisterns 4.65 0.03
It is used not only to assess the victim’s state of conscious-
ness, but it can also be useful in predicting the condition of
patients in the following days after injury [27]. Likewise, our
data prove a strong impact of GCS and the pupil response on
the outcome and mortality rate of patients operated on for
extra-axial hematoma [28–30]. This relationship was also
strongly marked in the observations of other authors [11, 15,
31–34]. Importantly, the multivariate analysis placed GCS
score and pupil response to light among the factors with the
strongest prognostic value and our studies showed that a
drop by one point in the GCS score almost doubles the risk
of in-hospital death. However, it should be noted that some
authors did not notice similar trends [26].
The Glasgow scale is a very useful tool, but the strategy
of dealing with trauma patients cannot depend only on this
indicator. It has been proved that the score is related to a
number of parameters such as drug and alcohol intoxication,
medical sedation, hypotension or hypoxia, often independent
of the injury [12]. Hypotension and hypoxia following TBI
are recognized as a significant secondary disorder associ-
ated with a poor outcome [35]. Petroni et al. found a very
Fig. 2  Time in-hospital mortality

of patients with TBI. Analysis carried out by Pozzato et al.

on 6827 people hospitalized due to TBI showed that the
risk of severe trauma is highest at age 15–19 and over 75.
The authors estimated that for the latter group it is three
times larger than for the general population [8]. A meta-
analysis by Hukkelhoven et al. [9] showed that only 15% of
patients over 65 years of age have a positive prognosis after
severe head injury. Similar conclusions were made by Perel
et al. [10] stating that increasing age above 40-years old was
associated with higher mortality rate, creating approximately
linear function. The relationship between age, unfavorable
prognosis and the mortality rate were also seen in the works
of other authors [11–17]. It is probably associated with
lower regenerative abilities and greater sensitivity of the Fig. 3  Dominant factors for the dependent variables of the in-hospital
brain tissue of older people to ischemia as well as frequent mortality

13
Factors with the strongest prognostic value associated with in‑hospital mortality rate among… 1521

strong relationship between low blood pressure and mor- hematocrit (Ht), and red blood cell count (RBC) are an
tality. Hypotension (SBP < 90 mmHg) was associated with independent risk factor for unfavorable prognosis. Interest-
90% mortality rate [36]. The values of SBP higher than 135 ingly, the authors also emphasized the role of red blood cell
or even 150 or lower than 90 mmHg were associated with distribution width (RDW) and platelet count disorders in
poorer outcomes [22, 37]. Similarly, our Data Mining analy- predicting patient mortality. A similar analysis was carried
sis placed SBP and saturation amongst the dominant factors out by Corbett et al. In this study, the association of abnor-
that influence mortality with a high statistical significance malities of hemoglobin on admission with a higher risk of
(p < 0.0003). adverse prognosis was present during the first 18 months
The impact of other vital signs on the prognosis of after major TBI. No such relationship was found for the
patients with TBI was also analyzed. Respiratory rate higher parameters of the white blood cell system (neutrophil and
than 25 and lower than 10 increased mortality rates. Simi- lymphocyte counts), which was previously presented by
larly, heart rates beyond a normal range are associated with other authors [42, 44]. Our analysis did not reveal any rela-
a poor outcome in TBI [37]. The role of respiratory failure tionship between abnormalities in blood morphotic systems
was also emphasized in the work of other authors. Osterman and in-hospital mortality.
et al. [11] showed that mortality among patients after TBI Interestingly, Corbett et al. [42] also showed a relation-
with coexisting respiratory failure reaches 79.3%. It should ship between fibrinogen metabolism changes, activated
be emphasized that our research placed RR among the fac- partial thromboplastin time (APTT), international normal-
tors closely correlating with mortality (p < 0.0000001). ized ratio (INR) and disseminated intravascular coagula-
Oxygen saturation is yet another indicator directly related tion score (DIC score) and a greater risk of unsuccessful
to mortality rates following TBI. Kalayci et al. [38] studied prognosis within 18 months after the injury. The authors
the patients undergoing craniectomy for SDH and proved attributed a special predictive value to INR [42]. Likewise,
that the saturation less than or equal to 96% was significantly Fuji et al. [45] studying patients who underwent surgery for
associated with higher mortality rates (p = 0.004). intracranial hematomas, found that lower values of INR and
PTT ensured remarkably better outcomes than higher ones.
Laboratory variables Coagulopathies, especially changes in prothrombin time
and platelet counts, are major determinants of disability and
Laboratory variable in TBI patients have also been fre- death among the patients with traumatic intracranial hem-
quently analyzed in the literature. One of them is stress orrhage [39, 40]. Epstein et al. [46] in their meta-analysis
hyperglycemia, a common finding after the injury [39]. High covering 22 studies determined the average percentage of
glucose is a cause of secondary insults for the patients after patients with TBI who develop coagulopathy at 35.2%. A
TBI, and it is associated with a poorer outcome [40]. This convergent correlation was seen in the work of Yuan et al.
trend was also noted in observations carried out by Bobeff They estimated that this problem affects 18.6% of patients
et al. The authors showed that patients with glucose levels after isolated TBI, but increases to 30.4% in serious inju-
exceeding 160 mg/dL are at a higher risk of complications ries. The authors also showed that patients with severe TBI
after treatment [41]. Similarly, the work of Corbett et al. were characterized by higher INR, prothrombin time (PT),
[42] showed that disorders in this area are associated with APTT, D-dimer level and lower PLT and fibrinogen levels
higher risk of unfavorable outcome at 18 months after severe [31]. Earlier studies defined PT as the indicator most often
TBI. Likewise, in our study hyperglycemia placed among disturbed in the case of TBI, but APTT seems to correlate
the factors with the strongest prognostic value concerning better with prognosis and mortality among patients [47–50].
in-hospital mortality (p = 0.000002). Importantly, our multivariate analysis showed that the pres-
Many authors also considered the relationship between ence of any coagulopathy increases the risk of in-hospital
electrolyte balance and the condition of patients after injury. death almost six times.
It has been shown that both hypo- and hypernatremia are Alcohol consumption was also found to be an important
associated with a poorer outcome, thus sodium revealed a risk factor for TBI, with the prevalence of alcohol intoxica-
U shaped relationship with the outcome [5, 40, 43]. Impor- tion between 20 and 55% at the time of the injury [51, 52].
tantly, hypo-natremia is a relatively infrequent occurrence In our study 31.25% patients were under the influence of
on admission following TBI. Contrariwise, our study did alcohol. Alcohol intoxication was associated with a poorer
not confirm the effects of hypo- and hypernatremia on in- outcome after a severe TBI [15, 52]. On the other hand, it is
hospital mortality. associated with a decreased mortality [53], thus the relation-
TBI-associated changes may also be seen in patients’ ship between alcohol and the outcome after TBI remains
blood counts. Bobeff et al. [41] showed that abnormali- uncertain [53–55]. Our results did not include alcohol in the
ties in hematological parameters such as hemoglobin (Hg), factors affecting mortality.

13
1522 B. Kulesza et al.

Computer tomography scan characteristics showed that in-hospital mortality was significantly lower in
the group of patients operated on within up to 200 min after
Imaging studies currently play a key role in the management of the arrival at the emergency department (p = 0.03). Seelig
head injuries. Strong evidence was found for the midline shift et al. [71] studying the patients undergoing surgery for SDH,
[27, 56, 57] and a greater increase of the midline shift associ- found that the surgery would reduce mortality from 90 to
ated with a higher mortality [56, 58]. Ostermann et al. [11] in 30% within 4 h. On the other hand, there are still a few stud-
their observations of 265 elderly patients after TBI showed that ies that did not associate a shorter period of time between
midline shifts affect 25% of them. At the same time, they stated the injury and the surgery with the outcome [26, 72, 73].
that shift over 15 mm is associated with a significantly higher Our statistical analysis shown no association between time
risk of death. In a large analysis of 861 patients after TBI Nel- to surgery and in-hospital mortality, but it was reasonable
son et al. [59] identified midline shift as the most important to perform a surgery as soon as possible.
risk factor for adverse prognosis. The authors studying patients
with EDH and SDH found a higher mortality associated with a
greater midline shift, as well as the thickness of hematoma [26, Conclusion
29, 30]. Our multivariate analysis confirms these relationships
between MLS and mortality rate (p = 0.0006). Amongst the contempororary literature on the patients with
The presence of a traumatic subarachnoid hemorrhage a traumatic brain injury, only a few of the studies analyzed a
and intraventricular hemorrhage predicts higher mortality selected group of patients operated on due to the extra-axial
[5, 27, 56, 60]. Our multivariate analysis using Data Min- hematomas. We were unable to find a study which would
ing demonstrates that SAH and IVH are both significantly collectively analyze all of the factors which we examined in
associated with the mortality rate (p = 0.01). The next fac- our cohort of patients operated on due to epi and subdural
tor associated with in-hospital mortality were the type of hematomas. Accordingly, we identified a group of the most
hematoma and status of basal cisterns in multivariate analy- important factors that highly significantly (p ≤ 0.000007)
sis (p = 0.0009 and p = 0.03). In-hospital mortality in EDH correlate with the in-hospital mortality rate such as: the
group was 10.71% while in SDH group—45%. Grigorakos initial GCS scale, respiratory rate, glycaemia, and satura-
et al. found the highest mortality rate in SDH (43.75%) in tion. Apart from the GCS scale, these are the factors that
comparison with other post-traumatic changes seen in com- can be effectively treated outside the hospital by the ambu-
puted tomography [61]. Khaled et al. found mortality rate lance team, which may eventually result in decreased mor-
among patients with EDH at 10.66% [24]. Status of basal tality rate. The presence of coagulopathy, however, which
cisterns was closely associated with mortality [10, 22, 58, increases the risk of in-hospital mortality rate almost six
62]. Our research showed that all these factors significantly times can be effectively treated only in a hospital setting.
correlated with mortality; in addition the presence of EDH Therefore, a prompt transportation to the hospital is also of
increases the survival chances when compared to patients the paramount importance since the treatment for coagulopa-
with SDH. Similarly, the smaller the maximum thickness thy should be introduced as soon as possible.
of hematomas, the death risk is lower. Normal size basal Our results require confirmation in other studies on
cisterns increase the survival chance by 0.18. a larger group of patients because this study have signifi-
The presence of skull fractures may also be important. cant limitations. First, it included relatively small number
On the one hand, thanks to mechanical damage, some of the of patients operated on for epidural hematoma. Second, no
energy is absorbed, which would otherwise be transferred patients conservatively treated due to TBI were included in
directly to the sensitive brain tissue. On the other hand, the the study, while a large group of patients after TBI such as
presence of fractures directly testifies to the large force caus- those with brain contusions and small intracranial hemato-
ing the injury [63]. The skull fracture among TBI patients is mas with no mass effect do not require surgery in the first
associated with an increased risk of neurosurgically-relevant line treatment. We also did not consider other factors such
intracranial lesion [64]. Bobeff et al. showed a higher inci- as blood gas, hematoma volume, C-reactive protein levels,
dence of complications during treatment in patients with lin- D-dimer and thyroid-stimulating hormone in the study.
ear skull fracture [41]. Our results did not include fractures
as factors of unfavorable prognosis. Funding This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.

Time between injury and operation

Compliance with ethical standards
Most researchers agree that delaying the moment of surgical Conflict of interest Bartłomiej Kulesza, Marek Mazurek, Adam
treatment implementation negatively affects the survival and Nogalski and Radosław Rola declare that they have no conflict of in-
prognosis of patients [29, 65, 68, 69]. Matsushima et al. [70] terest.

13
Factors with the strongest prognostic value associated with in‑hospital mortality rate among… 1523

Consent ethics commission The study obtained the consent of the Bio- Surg Res. 2018;13(1):277. https​://doi.org/10.1186/s1301​8-018-
ethics Commission located at Medical University of Lublin. Number 0975-y (Published 2018 Nov 3).
of consent KE-0254/313/2016. 12. Fu TS, Jing R, McFaull SR, Cusimano MD. Recent trends in
hospitalization and in-hospital mortality associated with trau-
matic brain injury in Canada: a nationwide, population-based
Open Access This article is licensed under a Creative Commons Attri- study. J Trauma Acute Care Surg. 2015;79:449–54.
bution 4.0 International License, which permits use, sharing, adapta- 13 Susman M, DiRusso SM, Sullivan T, Risucci D, Nealon P, Cuff
tion, distribution and reproduction in any medium or format, as long S, Haider A, Benzil D. Traumatic brain injury in the elderly:
as you give appropriate credit to the original author(s) and the source, increased mortality and worse functional outcome at discharge
provide a link to the Creative Commons licence, and indicate if changes despite lower injury severity. J Trauma. 2002;53:219–23 (dis-
were made. The images or other third party material in this article are cussion 223–214).
included in the article’s Creative Commons licence, unless indicated 14. Mosenthal AC, Livingston DH, Lavery RF, Knudson MM, Lee
otherwise in a credit line to the material. If material is not included in S, Morabito D, Manley GT, Nathens A, Jurkovich G, Hoyt DB,
the article’s Creative Commons licence and your intended use is not Coimbra R. The effect of age on functional outcome in mild
permitted by statutory regulation or exceeds the permitted use, you will traumatic brain injury: 6-month report of a prospective multi-
need to obtain permission directly from the copyright holder. To view a center trial. J Trauma. 2004;56:1042–8.
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. 15. Einarsen CE, van der Naalt J, Jacobs B, Follestad T, Moen
KG, Vik A, Håberg AK, Skandsen T. Moderate traumatic
brain injury: clinical characteristics and a prognostic model
of 12-month outcome. World Neurosurg. 2018;114:e1199–
References: e1210210. https​://doi.org/10.1016/j.wneu.2018.03.176 (Epub
2018 Mar 31).
1. Ghajar J. Traumatic brain injury. Lancet. 2000;356:923–9. 16. Roe C, Skandsen T, Manskow U, Ader T, Anke A. Mortality and
2. Koskinen S, Alaranta H. Traumatic brain injury in Finland one-year functional outcome in elderly and very old patients with
1991–2005: a nationwide register study of hospitalized and fatal severe traumatic brain injuries: observed and predicted. Behav
TBI. Brain Inj. 2008;22:205–14. Neurol. 2015;2015:845–9.
3. Peeters W, van den Brande R, Polinder S, Brazinova A, Stey- 17. de Guise E, LeBlanc J, Dagher J, Tinawi S, Lamoureux J, Mar-
erberg EW, Lingsma HF, Maas AI. Epidemiology of traumatic coux J, Maleki M, Feyz M. Traumatic brain injury in the elderly:
brain injury in Europe. Acta Neurochir. 2015;157:1683–96. a level 1 trauma centre study. Brain Inj. 2015;29:558–64.
4. Dewan MC, Rattani A, Gupta S, Baticulon RE, Hung YC, 18. Ushewokunze S, Nannapaneni R, Gregson BA, Stobbart L, Cham-
Punchak M, Agrawal A, Adeleye AO, Shrime MG, Rubiano bers IR, Mendelow AD. Elderly patients with severe head injury
AM, Rosenfeld JV, Park KB. Estimating the global incidence in coma from the outset—has anything changed? Br J Neurosurg.
of traumatic brain injury. J Neurosurg. 2018. https​: //doi. 2004;18:604–7.
org/10.3171/2017.10.JNS17​352. 19. Mina AA, Knipfer JF, Park DY, Bair HA, Howells GA, Bendick
5. Kulesza B, Litak J, Grochowski C, Nogalski A, Rola R. The ini- PJ. Intracranial complications of preinjury anticoagulation in
tial factors with strong predictive value in relation to six-month trauma patients with head injury. J Trauma. 2002;53:668–72.
outcome among patients operated due to extra-axial hematomas. 20. Mohindra S, Mukherjee KK, Gupta R, Chhabra R. Continuation
Diagnostics. 2020;10(3):174. https​://doi.org/10.3390/diagn​ostic​ of poor surgical outcome after elderly brain injury. Surg Neurol.
s1003​0174. 2008;69:474–7.
6. Kulesza B, Mazurek M, Rams L, Nogalski A. Acute epidural 21 Kada AY, Bouyoucef KA. La mort encéphalique post traumatique:
and subdural hematomas after head injury: clinical distinguish- épidémiologie et facteurs de risque [Post traumatic brain death:
ing features. Indian J Surg. 2020. https​://doi.org/10.1007/s1226​ epidemiology and risk factors]. Pan Afr Med J. 2018;31:29. https​
2-020-02304​-w. ://doi.org/10.11604/​ pamj.2018.31.29.16354​ (Published 2018 Sep
7. Hukkelhoven CW, Steyerberg EW, Habbema JD, Farace E, 13).
Marmarou A, Murray GD, Marshall LF, Maas AI. Predicting 22. Murray GD, Butcher I, McHugh GS, Lu J, Mushkudiani NA,
outcome after traumatic brain injury: development and inter- Maas AI, Marmarou A, Steyerberg EW. Multivariable prognostic
national validation of prognostic scores based on admission analysis in traumatic brain injury: results from the IMPACT study.
characteristics. J Neurotrauma. 2005;22(10):1025–39. J Neurotrauma. 2007;24(2):329–37.
8. Pozzato I, Tate RL, Rosenkoetter U, Cameron ID. Epidemiol- 23. Mushkudiani NA, Engel DC, Steyerberg EW, Butcher I, Lu J,
ogy of hospitalised traumatic brain injury in the state of New Marmarou A, Slieker F, McHugh GS, Murray GD, Maas AI.
South Wales, Australia: a population-based study. Aust N Z J Prognostic value of demographic characteristics in traumatic
Public Health. 2019;43(4):382–8. https​://doi.org/10.1111/1753- brain injury: results from the IMPACT study. J Neurotrauma.
6405.12878​(Epub 2019 Mar 4). 2007;24(2):259–69.
9. Hukkelhoven CW, Steyerberg EW, Rampen AJ, Farace E, 24. Khaled CN, Raihan MZ, Chowdhury FH. Surgical management
Habbema JD, Marshall LF, Murray GD, Maas AI. Patient age of traumatic extradural hematoma: experiences with 610 patients
and outcome following severe traumatic brain injury: an analy- and prospective analysis. Ind J Neurotrauma. 2008;2:75–9.
sis of 5600 patients. J Neurosurg. 2003;99:666–73. 25. Khan KA, Choudhary M, Sinha VD, Gora N, Bairwa M. Predic-
10. MRC CRASH Trial Collaborators, Perel P, Arango M, Clayton tors of outcome after traumatic brain injuries: experience of a ter-
T, Edwards P, Komolafe E, Poccock S, Roberts I, Shakur H, tiary health care institution in Northwest India. World Neurosurg.
Steyerberg E, Yutthakasemsunt S. Predicting outcome after trau- 2019;126:e699–e705. https:​ //doi.org/10.1016/j.wneu.2019.02.126
matic brain injury: practical prognostic models based on large (Epub 2019 Mar 4).
cohort of international patients. BMJ. 2008;336(7641):425–9. 26. Baucher G, Troude L, Pauly V, Bernard F, Zieleskiewicz L, Roche
11. Ostermann RC, Joestl J, Tiefenboeck TM, Lang N, Platzer P, PH. Predictive factors of poor prognosis after surgical manage-
Hofbauer M. Risk factors predicting prognosis and outcome of ment of traumatic acute subdural hematomas: a single-center
elderly patients with isolated traumatic brain injury. J Orthop series. World Neurosurg. 2019;126:e944–e952952. https​://doi.
org/10.1016/j.wneu.2019.02.194 (Epub 2019 Mar 12).

13
1524 B. Kulesza et al.

27. Kulesza B, Litak J, Mazurek M, Nogalski A. The initial factors after traumatic brain injury: a retrospective cohort study. J Neu-
affecting 6-month outcome of patients undergoing surgery for rol Surg A Cent Eur Neurosurg. 2019;80(6):460–9. https​://doi.
acute post-traumatic subdural and epidural hematoma. Folia Med org/10.1055/s-0039-16926​72 (Epub 2019 Aug 29).
(Plovdiv). 2020;62(1):94–104. https:​ //doi.org/10.3897/folmed​ .62. 42. Corbett JM, Ho KM, Honeybul S. Prognostic significance of
e4774​3. abnormal hematological parameters in severe traumatic brain
28. Taussky P, Widmer HR, Takala J, Fandino J. Outcome after acute injury requiring decompressive craniectomy. J Neurosurg.
traumatic subdural and epidural haematoma in Switzerland: a 2019;132(2):545–51. https​://doi.org/10.3171/2018.10.JNS18​
single-centre experience. Swiss Med Wkly. 2008;138:281–5. 2293.
29. Gurer B, Kertmen H, Yilmaz ER, Dolgun H, Hasturk AE, Sekerci 43. Rejeb I, Chakroun O, Chtara K, Boujelbene M, Ksibi H, Chaari
Z. The surgical outcome of traumatic extraaxial hematomas caus- A, Bahloul M, Rekik N. Factors predicting early outcome in
ing brain herniation. Turk Neurosurg. 2017;27(1):37–52. patients admitted at emergencydepartment with severe head
30. Hamed M, Schuss P, Daher FH, Borger V, Güresir A, Vatter H, trauma. J Acute Dis. 2015;4(1):68–72.
Güresir E. Acute traumatic subdural hematoma: surgical manage- 44. Gürkanlar D, Lakadamyali H, Ergun T, Yilmaz C, Yücel E,
ment in the presence of cerebral herniation—a singlecenter series Altinörs N. Predictive value of leucocytosis in head trauma.
and multivariate analysis. World Neurosurg. 2016;94:501–6. Turk Neurosurg. 2009;19(3):211–5.
31. Yuan Q, Yu J, Wu X, Sun Y, Li Z, Du Z, Wu X, Hu J, correspond- 45. Fujii T, Moriel G, Kramer DR, et al. Prognostic factors of early
ing author. Prognostic value of coagulation tests for in-hospital outcome and discharge status in patients undergoing surgical
mortality in patients with traumatic brain injury. Scand J Trauma intervention following traumatic intracranial hemorrhage. J Clin
Resusc Emerg Med. 2018;26(1):3. https​://doi.org/10.1186/s1304​ Neurosci. 2016;31:152–6.
9-017-0471-0 (Published 2018 Jan 5). 46. Epstein DS, Mitra B, O’Reilly G, Rosenfeld JV, Cameron
32. Steyerberg EW, Mushkudiani N, Perel P, Butcher I, Lu J, PA. Acute traumatic coagulopathy in the setting of isolated
McHugh GS, Murray GD, Marmarou A, Roberts I, Habbema traumatic brain injury: a systematic review and meta-analy-
DF, Maas AI. Predicting outcome after traumatic brain injury: sis. Injury. 2014;45:819–24. https​: //doi.org/10.1016/j.injur​
development and international validation of prognostic scores y.2014.01.011.
based on admission characteristics. PLoS Med. 2008;5:e165. 47. MacLeod JB, Lynn M, McKenney MG, Cohn SM, Murtha M.
33. Jacobs B, Beems T, van der Vliet TM, van Vugt AB, Hoede- Early coagulopathy predicts mortality in trauma. J Trauma.
maekers C, Horn J, Franschman G, Haitsma I, Naalt J, Andries- 2003;55:39–44. https​: //doi.org/10.1097/01.TA.00000​7 5338​
sen TM, Borm GF, Vos PE. Outcome prediction in moderate .21177​.EF.
and severe traumatic brain injury: a focus on computed tomog- 48. Lustenberger T, Talving P, Kobayashi L, Inaba K, Lam L, Plurad
raphy variables. Neurocrit Care. 2013;19:79–89. D, Demetriades D. Time course of coagulopathy in isolated
34. Roozenbeek B, Lingsma HF, Lecky FE, Lu J, Weir J, Butcher I, severe traumatic brain injury. Injury. 2010;41:924–8. https​://
et al. Prediction of outcome after moderate and severe traumatic doi.org/10.1016/j.injur​y.2010.04.019.
brain injury: external validation of the International Mission on 49. Allard CB, Scarpelini S, Rhind SG, Baker AJ, Shek PN, Tien
Prognosis and Analysis of Clinical Trials (IMPACT) and Cor- H, Fernando M, Tremblay L, Morrison LJ, Pinto R, Rizoli SB.
ticoid Randomisation After Significant Head injury (CRASH) Abnormal coagulation tests are associated with progression of
prognostic models. Crit Care Med. 2012;40:1609–17. traumatic intracranial hemorrhage. J Trauma. 2009;67:959–67.
35. McHugh GS, Engel DC, Butcher I, Steyerberg EW, Lu J, https​://doi.org/10.1097/TA.0b013​e3181​ad5d3​7.
Mushkudiani N, Hernández AV, Marmarou A, Maas AI, Gor- 50 Talving P, Benfield R, Hadjizacharia P, Inaba K, Chan LS.
don DM. Prognostic value of secondary insults in traumatic Demetrios Demetriades. Coagulopathy in severe traumatic brain
brain injury: results from the IMPACT study. J Neurotrauma. injury: a prospective study. J Trauma. 2009;66:55–61.
2007;24(2):287–93. 51. Bernier R, Hillary F. Trends in alcohol use during moderate
36. Petroni G, Quaglino M, Lu jan S, Kovalevski L, Rondina C, and severe traumatic brain injury: 18 years of neurotrauma in
Videtta W, Carney N, Nancy T, Randall C. Early prognosis of Pennsylvania. Brain Inj. 2016;30(4):414–21.
severe traumatic brain injury in an urban argentinian trauma 52. Pandit V, Patel N, Rhee P, Kulvatunyou N, Aziz H, Green DJ,
center. J Trauma. 2010;68:564–70. O’Keeffe T, Zangbar B, Tang A, Gries L, Friese RS, Joseph B.
37. Saadat S, Akbari H, Khorramirouz R, et al. Determinants of Effect of alcohol in traumatic brain injury: is it really protec-
mortality in patients with traumatic brain injury. Ulus Travma tive? J Surg Res. 2014;190(2):634–9.
Acil Cerrahi Derg. 2012;18(3):219–24. 53. Plurad D, Demetriades D, Gruzinski G, Preston C, Chan L,
38. Kalayci M, Aktunç E, Gül S, Hanci V, Edebali N, Cagavi F, Gaspard D, Margulies D, Cryer G. Motor vehicle crashes: the
Açikgöz B. Decompressive craniectomy for acute subdural association of alcohol consumption with the type and severity
haematoma: an overview of current prognostic factors and a of injuries and outcomes. J Emerg Med. 2010;38:12–7.
discussion about some novel prognostic parametres. J Pak Med 54. Raj R, Skrifvars MB, Kivisaari R, Hernesniemi J, Lappalainen
Assoc. 2013;63:34–49. J, Siironen J. Acute alcohol intoxication and long-term out-
39. Joseph B, Aziz H, Zangbar B, Kulvatunyou N, Pandit V, come in patients with traumatic brain injury. J Neurotrauma.
O’Keeffe T, Tang A, Wynne J, Friese RS, Rhee P. Acquired 2015;32:95–100.
coagulopathy of traumatic brain injury defined by routine labo- 55. Alexander S, Kerr ME, Yonas H, Marion DW. The effects
ratory tests: Which laboratory values matter? J Trauma Acute of admission alcohol level on cerebral blood flow and out-
Care Surg. 2014;76(1):121–5. comes after severe traumatic brain injury. J Neurotrauma.
40. Van Beek JG, Mushkudiani NA, Steyerberg EW, Butcher I, 2004;21:575–83.
McHugh GS, Lu J, Marmarou A, Murray GD, Maas AI. Prog- 56. Maas AI, Steyerberg EW, Butcher I, Dammers R, Lu J, Marmarou
nostic value of admission laboratory parameters in traumatic A, Mushkudiani NA, McHugh GS, Murray GD. Prognostic value
brain injury: results from the IMPACT study. J Neurotrauma. of computerized tomography scan characteristics in traumatic
2007;24(2):315–28. brain injury: results from the IMPACT study. J Neurotrauma.
41. Bobeff EJ, Posmyk BJ, Bobeff KŁ, Fortuniak J, Wiśniewski K, 2007;24(2):303–14.
Stawiski K, Stefańczyk L, Jaskólski DJ. Predicting outcome and 57. Hiler M, Czosnyka M, Hutchinson P, Balestreri M, Smielewski
conservative treatment failure in patients with skull fracture P, Matta B, Pickard JD. Predictive value of initial computerized

13
Factors with the strongest prognostic value associated with in‑hospital mortality rate among… 1525

tomography scan, intracranial pressure, and state of autoregu- 65. Park JH, Park JE, Kim SH, Lim YC, You NK, Ahn YH, Choi H,
lation in patients with traumatic brain injury. J Neurosurg. Cho J. Outcomes of ultra-early decompressive craniectomy after
2006;104(5):731–7. severe traumatic brain injury-treatment outcomes after severe TBI.
58. Bhaskar Naidu P, Truman Srinivas M, Selvakumar K. Computed Korean J Neurotrauma. 2014;10:112–8.
tomography predictors for in-hospital mortality in severe and 66. Haselsberger K, Pucher R, Auer LM. Prognosis after acute
moderate head injury patients. Inter Sur J. 2016;3(3):1306–9. subdural or epidural haemorrhage. Acta Neurochir (Wien).
59. Nelson DW, Nystrom H, MacCallum RM, Thornquist B, Lilja A, 1988;90:111–6.
Bellander BM, Rudehill A, Wanecek M, Weitzberg E. Extended 67. Wilberger JE, Harris M, Diamond DL. Acute subdural hema-
analysis of early computed tomography scans of traumatic toma: morbidity, mortality, and operative timing. J Neurosurg.
brain injured patients and relations to outcome. J Neurotrauma. 1991;74:212–8.
2010;27:51–64. 68. Sakas DE, Bullock MR, Teasdale GM. One-year outcome fol-
60. Fabbri A, Servadei F, Marchesini G, Stein SC, Vandelli A. Early lowing craniotomy for traumatic hematoma in patients with fixed
predictors of unfavourable outcome in subjects with moderate dilated pupils. J Neurosurg. 1995;82:961–5.
head injury in the emergency department. J Neurol Neurosurg 69. Zafrullah Arifin M, Gunawan W. Analysis of presurgery time as
Psychiatry. 2008;79:567–73. a prognostic factor in traumatic acute subdural hematoma. J Neu-
61. Grigorakos L, Alexopoulou A, Tzortzopoulou K, Stratouli S, rosurg Sci. 2013;57:277–80.
Chroni D, Papadaki E, Alamanos I, Sakellaridis N. Predictors of 70. Matsushima K, Inaba K, Siboni S, Skiada D, Strumwasser AM,
outcome in patients with severe traumatic brain injury. J Neurosci Magee GA, Sung GY, Benjaminm ER, Lam L, Demetriades D.
Clin Res. 2016;1:1–4. Emergent operation for isolated severe traumatic brain injury:
62. Toutant SM, Klauber MR, Marshall LF, et al. Absent or com- Does time matter? J Trauma Acute Care Surg. 2015;79:838–42.
pressed basal cisterns on first CT scan: ominous predictors of 71. Seelig JM, Becker DP, Miller JD, Greenberg RP, Ward JD, Choi
outcome in severe head injury. J Neurosurg. 1984;61:691–4. SC. Traumatic acute subdural hematoma: major mortality reduc-
63 Wood RL, Eames R. Application of behaviour modifi cation in the tion in comatose patients treated within four hours. N Engl J Med.
rehabilitation of traumatically brain-injured patients. In: Davey G, 1981;304:1511–8.
editor. Applications of conditioning theory. New York: Methuen; 72. Sergides IG, Whiting G, Howarth S, Hutchinson PJ. Is the recom-
1981. p. 81–101. mended target of 4 hours from head injury to emergency crani-
64. Munoz-Sánchez MA, Murillo-Cabezas F, Cayuela-Domínguez otomy achievable? Br J Neurosurg. 2006;20:301–5.
A, Rincón-Ferrari MD, Amaya-Villar R, León-Carrión J. Skull 73. Kotwica Z, Brzezinski J. Acute subdural hematoma in adults:
fracture, with or without clinical signs, in mTBI is an independ- an analysis of outcome in comatose patients. Acta Neurochir.
ent risk marker for neurosurgically relevant intracranial lesion: a 1993;121:95–9.
cohort study. Brain Inj. 2009;23:39–44.

13