Acta Neurochirurgica

https://doi.org/10.1007/s00701-018-3656-z

ORIGINAL ARTICLE - BRAIN TRAUMA

Frequency and characteristics of traumatic brain injury in restrained

drivers involved in road traffic accidents
Joji Inamasu 1 & Msasato Nakaya 1 & Dai Kujirai 2 & Keita Mayanagi 1 & Masashi Nakatsukasa 1

Received: 16 May 2018 / Accepted: 13 August 2018

# Springer-Verlag GmbH Austria, part of Springer Nature 2018

Abstract
Background While seatbelt is an important device protecting drivers from traumatic brain injury (TBI), it has rarely been reported
how often and in what circumstances restrained drivers sustain TBI after road traffic accident (RTA). Whole-body computed
tomography (WBCT) for blunt trauma patients may provide a unique opportunity to investigate the frequency and characteristics
of TBI sustained by restrained drivers.
Methods A single-center, retrospective observational study was conducted using prospectively acquired data. Between January
2013 and December 2017, 409 restrained drivers (284 men/125 women, mean age of 45.0 ± 19.1 years) whose vehicle had been
severely damaged in RTAs underwent WBCT for evaluation of injuries. Multivariate regression analysis was performed to
identify variables predictive of TBI. Influence of collision patterns (frontal, lateral or rollover) on the frequency and severity
of TBI was evaluated. Relationship between collision patterns and CT findings was also reviewed.
Results Thirty-one restrained drivers (7.6%) sustained TBI after RTA. The distribution of Glasgow Coma Scale (GCS) scores
among the 31 drivers was 15 in 9, 13–14 in 9, 9–12 in 4, and ≤ 8 in 9, indicating that the majority of TBIs were classifiable as
mild. The frequency of TBI in alert and oriented drivers, i.e., those with a GCS score of 15, was 2.9%. Multivariate regression
analysis showed that both altered mental status (OR, 4.933; 95% CI, 1.135–21.431) and loss of consciousness (OR, 6.492; 95%
CI, 1.669–25.249) were associated with TBI. The frequency of TBI tended to be higher in drivers with rollover collision than
those with frontal collision (6 vs. 13%, p = 0.07). Interhemispheric acute subdural hematoma and subcortical petechial hemor-
rhage seemed to be characteristic CT findings in drivers with frontal and lateral collision, respectively.
Conclusions The key finding of this study, i.e., that (1) TBI was observed in 7.6% of restrained drivers with severe vehicular
damage, may provide useful information to neurosurgeons who take care of RTA victims. The majority of the TBIs were mild
without need for neurosurgical intervention. While association may exist between type of collision and type of brain injury,
further studies with prospective design are warranted.

Keywords Driver . Traumatic brain injury . Road traffic accident . Seatbelt . Whole-body computed tomography

Abbreviations WBCT Whole-body computed tomography

TBI Traumatic brain injury EMS Emergency medical service
RTA Road traffic accident ED Emergency department
GCS Glasgow Coma Scale
AIS Abbreviated Injury Scale
This article is part of the Topical Collection on Brain Trauma GOS Glasgow Outcome Scale
SD Standard deviation
* Joji Inamasu AMS Altered mental status
inamasu@fujita-hu.ac.jp LOC Loss of consciousness
SAH Subarachnoid hemorrhage
1
Department of Neurosurgery, Saiseikai Utsunomiya Hospital, 1-98 ASDH Acute subdural hematoma
Takebayashi, Utsunomiya 321-0974, Japan OR Odds ratio
2
Department of Emergency Medicine, Saiseikai Utsunomiya CI Confidence interval
Hospital, Utsunomiya, Japan ICH Intracerebral hemorrhage

Introduction
The clinical characteristics of TBI among drivers involved in
RTAs have been studied extensively. While a seatbelt is the most
important device protecting drivers from TBI, with unrestrained
drivers being more likely to sustain severe injury [11, 23], it has
rarely been reported how often and in what circumstances re-
strained drivers sustain TBI after RTAs. WBCT for patients
suffering from blunt trauma may provide a unique opportunity
to investigate the frequency and characteristics of TBI sustained
by restrained drivers [10, 15, 21]. The objective of this study
was to investigate the frequency and characteristics of TBI
sustained by restrained drivers after severe RTAs.
Materials and methods
This was a single-center, retrospective observational study
using prospectively acquired data. All procedures performed
in studies involving human participants were in accordance
with the ethical standards of the institutional research commit-
tee and with the 1964 Helsinki Declaration and its later
amendments or comparable ethical standards. Approval for
this study was given by our institutional research committee.
For this type of study (i.e., retrospective), formal consent is
not required from each patient.
A trauma team was activated after receiving a call from
EMS staff rescuing RTA victims at the scene. Trauma patients
were treated according to recent guidelines [19]. After their
arrival at our ED, the temporal sequence of resuscitative events
was recorded on an integrated clinical database CAP-2000
(Nihon Kohden, Tokyo, Japan) by emergency medicine resi-
dents. Automobile drivers brought to our ED by EMS after a
RTA from January 2013 to December 2017 were identified
from the database. Front/rear seat passengers were excluded
from analysis. Detailed information on the RTA, including
seatbelt usage, airbag deployment, collision patterns, vehicle
types, and estimated precrash speed of the damaged vehicle,
was provided by EMS staff in a rescue report. Collision patterns
were classified into three categories (frontal, lateral, and roll-
over), while rear-end collisions were excluded from analysis.
Vehicle types were classified into three categories (Compact,
with an engine capacity of <660 cc; Sedan, with an engine
capacity of ≥ 660 cc; and truck). Wagons, vans, and sports
utility vehicles were included under the Sedan category. The
distinction between Compact and Sedan type according to en-
gine capacity was based on the Road Traffic Act of Japan [20].
EMS staff had been trained to estimate the severity of ve-
hicular damage, subsequently classifying the damage into
three categories (mild, moderate, and severe) at the crash
scene [5, 18, 22, 25, 26]. In 2010, we started a prospective
WBCT protocol for occupants involved in RTAs on the con-
dition that (1) their vehicle was determined by EMS staff to
Acta Neurochir
have sustained severe damage and (2) the estimated precrash
speed had exceeded 40 km/h. Emergency medicine residents
explained the benefits and risks of WBCT, including the risk
of radiation exposure, to the drivers or their surrogates [8,
16]. WBCT was not performed in drivers who refused to
give consent and in those who were pregnant. A 64-
detector row helical CT scanner (SOMATOM Definition
AS, Siemens, Erlangen, Germany) was used throughout
the study period.
Clinical/radiological evaluation
The level of consciousness of drivers who sustained severe
RTAs was evaluated using the GCS [24]. The severity of bodily
injuries was classified and evaluated using the AIS [4]. Drivers
who sustained TBI were followed regularly for at least 30 days
after the RTA. Follow-up brain CTs were also obtained regular-
ly. The outcomes 30 days after the RTA were evaluated using
the GOS by an investigator (JI) who had not participated in
acute patient care [3]. CT diagnosis of TBI and other bodily
injuries had been established by a board-certified radiologist
on-call. TBI was defined as skull fractures and/or any type of
intracranial bleedings. Mild and severe TBIs were defined as
injuries with a GCS score of 13–15 and ≤ 8, respectively [13].
Statistical analysis
Fisher’s exact test was used to compare differences in categor-
ical variables, while Student’s t test was used to compare dif-
ferences in numerical variables. Numerical data are expressed
as mean ± SD, and p < 0.05 was considered statistically sig-
nificant. Multivariate regression analysis was performed using
SPSS for Windows Ver. 18.0 (SPSS Inc., Chicago, IL, USA)
to identify variables associated with TBI in restrained drivers.
Variables included age, sex, presence of AMS defined as a
GCS score ≤ 14, presence of LOC, airbag deployment, frontal
collision, use of Compact vehicle, and concomitant severe
(AIS ≥ 4) bodily injuries [4].
Results
Restrained vs. unrestrained drivers
During the 5-year study period, a total of 464 drivers whose
vehicle had been severely damaged in RTAs were brought to
our institution and underwent WBCT for evaluation of inju-
ries. They were dichotomized into restrained (n = 409, 88.1%)
and unrestrained drivers (n = 55, 11.9%) based on the use of
seatbelt at the time of RTA (Table 1). Among the 409 re-
strained drivers, 31 (24 men/7 women, mean age of 49.8 ±
20.1 years) were found to have sustained TBI. The frequency
of TBI was significantly lower in the restrained drivers than
Acta Neurochir
Table 1 Comparison between
restrained and unrestrained Demographic variables
drivers who were involved in
severe road traffic accidents Age (mean ± SD)
Male: Female
Airbag deployment
Traumatic brain injury
30-day mortality
N/A not available, SD standard deviation
*, **, statistically significant
unrestrained drivers (7.6 vs. 27.2%, p < 0.001) (Table 1).
Similarly, 30-day mortality rate was significantly lower in
the former (1.2 vs. 12.7%, p < 0.001). The airbag deployment
rate did not differ significantly.
Characteristics of restrained drivers who sustained
TBI
Clinical and radiographic characteristics and outcomes of the
31 restrained drivers who sustained TBI are summarized in
Table 2. The distribution of GCS scores was 15 in 9, 13–14 in
9, 9–12 in 4, and ≤ 8 in 9 drivers. Meanwhile, a total of 306
drivers exhibited a GCS score of 15 at ED, and the frequency
of TBI was 2.9% among drivers who were fully conscious at
ED (9/306). The mean hospital stay of the 31 drivers with TBI
was 6.2 ± 5.4 days. Four drivers sustained AIS ≥ 4 bodily
injuries for which surgical or radiological interventions were
necessary. While the great majority of restrained drivers with
TBI were treated conservatively, three required an emergency
craniotomy for a hematoma. At the 30-day follow-up, 20
drivers (64.5%) were determined to have a GOS score of 4–
5 (Table 2). The 30-day mortality rate was 9.7%. CT classi-
fication of TBI according to frequency was as follows: trau-
matic SAH in 9, interhemispheric ASDH in 8, subcortical
petechial hemorrhage in 6, convexity acute epidural hema-
toma in 3, convexity ASDH in 3, and contusion in two
drivers (Table 2).
Predictive factors
The 409 restrained drivers were dichotomized according to
the presence of TBI, and variables were compared between
the 31 drivers with TBI and 378 drivers without TBI (Table 3).
The frequency of AMS was significantly higher in drivers
with TBI (71.0 vs. 18.5%, p < 0.001). Similarly, the frequency
of LOC was significantly higher in the former (63.3 vs.
15.1%, p < 0.001). LOC without superficial scalp/facial injury
was observed significantly more frequently in the latter group
(5.0 vs. 29.8%, p = 0.03). Airbag deployment rate was signif-
icantly lower in drivers with TBI (38.9 vs. 64.4%, p = 0.04).
However, information on airbag deployment was unavailable
Restrained (n = 409) Unrestrained (n = 55) p
45.0 ± 19.1 40.4 ± 18.6 0.09
284: 125 40: 15 0.75
170/273 (N/A in 136) 24/42 (N/A in 13) 0.61
31 (7.6%) 15 (27.2%) < 0.001*
5 (1.2%) 8 (12.7%) < 0.001**
in 138 drivers (33.7%). No significant differences in the fre-
quency of severe bodily injuries, use of anti-thrombotic
agents, collision pattern, and vehicle type had been observed
between the two groups.
Results of multivariate regression analysis are shown in
Table 4. Accordingly, the presence of AMS (OR, 4.933;
95% CI, 1.135–21.431; p = 0.033) and LOC (OR, 6.492;
95% CI, 1.669–25.249; p = 0.007) was associated with TBI
in restrained drivers. By contrast, neither collision pattern nor
vehicle type was associated with TBI (Table 4).
Collison pattern and TBI
TBI was observed in 6% (14/251), 9% (10/107), and 13% (7/
52) of drivers with frontal, lateral, and rollover collisions,
respectively (Fig. 1a). The frequency of TBI after rollover
collisions tended to be higher compared to that after frontal
collisions (p = 0.07). The frequency of severe TBI was 14%
(2/14), 40% (4/10), and 43% (3/7) in drivers who had frontal,
lateral, and rollover collisions, respectively. Although the fre-
quency of TBI was higher in drivers with lateral and rollover
collisions than in those with frontal collisions, the differences
were not statistically significant (Fig. 1b).
Vehicle type and TBI
Detailed information on vehicle type was available in 383
drivers (93.6%). TBI was observed in 10% (15/155), 7%
(14/212), and 4% (1/26) of drivers of a Compact, Sedan, and
truck, respectively (Fig. 1c). No significant difference in the
frequency of TBI had been found between drivers of a
Compact and Sedan (p = 0.24). Similarly, no significant dif-
ference in the frequency of severe TBI (GCS score ≤ 8) had
been observed between drivers of a Compact and Sedan (20
vs. 43%, p = 0.25) (Fig. 2b). Given the small sample size,
statistical analysis was not conducted on truck drivers.
Collision pattern and CT findings
Brain CT images of the 31 drivers with TBI were classified
based on collision patterns. In the frontal collision group, 6 of
 Acta Neurochir

Table 2 Demographics, CT findings, and outcomes of 31 drivers with traumatic brain injury who were restrained at the time of road traffic accident

No. Age, sex GCS CT findings AIS ≥ 4 bodily injuries Tx for TBI 30-day GOS

1 60 F 15 IH-ASDH None Conservative 5

2 58 M 15 IH-ASDH None Conservative 5
3 36 M 15 IH-ASDH None Conservative 5
4 72 M 15 IH-ASDH None Conservative 5
5 64 M 15 T-SAH None Conservative 5
6 25 M 15 Petechial hemorrhage None Conservative 5
7 33 F 15 IH-ASDH Colon requiring laparotomy Conservative 4
8 79 M 15 T-SAH Colon requiring laparotomy Conservative 4
9 61 M 15 Petechial hemorrhage Colon requiring laparotomy Conservative 4
10 69 M 14 IH-ASDH None Conservative 5
11 45 M 14 IH-ASDH None Conservative 5
12 48 M 14 T-SAH None Conservative 5
13 69 M 14 T-SAH None Conservative 5
14 58 M 14 T-SAH None Conservative 4
15 62 F 14 Petechial hemorrhage None Conservative 5
16 37 M 14 Convexity AEDH None Conservative 5
17 38 M 13 T-SAH None Conservative 5
18 19 M 13 Petechial hemorrhage None Conservative 5
19 26 F 12 T-SAH None Conservative 4
20 55 M 12 Convexity AEDH None Conservative 4
21 82 M 10 T-SAH None Conservative 3
22 22 M 9 IH-ASDH None Conservative 3
23 21 F 8 T-SAH None Conservative 3
24 19 M 7 Petechial hemorrhage None Conservative 3
25 64 M 6 Petechial hemorrhage None Conservative 3
26 75 M 6 Convexity ASDH Spleen requiring IVR Craniotomy 2
27 19 F 5 Convexity ASDH None Craniotomy 2
28 47 F 3 Convexity AEDH None Craniotomy 3
29 81 M 3 Convexity ASDH Spinal cord injury Conservative 1
30 51 M 3 Contusion None Conservative 1
31 50 M 3 Contusion None Conservative 1

AEDH acute epidural hematoma, AIS Abbreviated Injury Scale, ASDH acute subdural hematoma, CT computed tomography, ED emergency department,
GCS Glasgow Coma Scale, GOS Glasgow Outcome Scale, IH interhemispheric, IVR interventional radiology, TBI, traumatic brain injury, T-SAH,
traumatic subarachnoid hemorrhage, Tx treatment

the 14 drivers (43%) sustained interhemispheric ASDHs (Fig.
2); all of which were confirmed to have disappeared on
follow-up CTs. In the lateral collision group, 5 of the 10
drivers (50%) sustained subcortical petechial hemorrhages
(Fig. 3); all of which were confirmed to have disappeared on
follow-up CTs. In the rollover collision group, no distinct
injury pattern on CT seemed to exist.
sustained subcortical hemorrhage. None of the four drivers
sustained extra- or intracranial injuries attributable to RTAs.
The chronological relationship between ICH onset and RTA in
all four drivers was confirmed by EMS staff at the scene.
Discussion

Spontaneous intracerebral hemorrhage as a cause
of RTA
Among the 407 restrained drivers involved in RTAs, 4 (1.0%)
were found to have sustained spontaneous ICH. Among them,
two sustained thalamic hemorrhage, whereas the other two
A seatbelt is the most important device protecting automobile
occupants after RTAs [6, 17], and its efficacy in reducing the
frequency of TBI and mortality rate was clearly shown in this
study (Table 1): it is not surprising that both the mortality rate
and frequency of TBI were significantly lower in the re-
strained drivers than unrestrained drivers. Meanwhile, it has
Acta Neurochir

Table 3 Comparison of restrained drivers with and without traumatic brain injury

Demographic variables TBI + (n = 31) TBI − (n = 378) p

Age (mean ± SD) 49.8 ± 19.5 44.6 ± 19.1 0.14

Male: Female 24: 7 260: 118 0.32
Altered mental status (GCS < 15) 22 (71.0%) 70 (18.5%) < 0.001*
Loss of consciousness 20 (63.6%) 57 (15.1%) < 0.001**
LOC without scalp/facial injury 1/20 (5.0%) 17/57 (29.8%) 0.03***
AIS ≥ 4 bodily injuries 4 (12.9%) 26 (6.9%) 0.27
Nonlife-threatening fractures 12 (38.7%) 140 (37.0%) 0.85
Use of anti-thrombotic drugs 6 (19.4%) 44 (11.4%) 0.25
Collision type (frontal:lateral:rollover) 14:10:7 237:97:45
% lateral/rollover collision 17 (54.8%) 142 (37.6%) 0.08
Vehicle type (Compact:Sedan:Truck) 15:14:1 (N/A in 1) 140:198:25 (N/A in 15)
% compact vehicle 15/30 (50.0%) 140/363 (38.6%) 0.25
Airbag deployment 7/18 (38.9%) (N/A in 13) 163/255 (64.4%) (N/A in 123) 0.04****

AIS Abbreviated Injury Scale, GCS Glasgow Coma Scale, IVR interventional radiology, LOC loss of consciousness, SD standard deviation, TBI
traumatic brain injury
*, **, ***, ****, statistically significant

rarely been reported how often and in what circumstances
restrained drivers suffer from TBI after RTAs. What was
unique in this study is that it evaluated, using a prospective
WBCT protocol, the frequency and clinical/radiological char-
acteristics of TBI in restrained drivers who sustained severe
vehicular damage. While WBCTs have been employed in
trauma patients presenting with AMS and/or compromised
vital signs in many studies [9, 14, 20], we focused on identi-
fying mild TBI among seemingly intact drivers.
The key findings of this study, i.e., that (1) TBI was ob-
served in 7.6% of restrained drivers with severe vehicular
damage; and (2) the frequency of TBI in alert and oriented
drivers (i.e., GCS-15) was 2.9%; may provide useful informa-
tion to neurosurgeons and emergency medicine physicians
who take care of RTA victims. The majority of the TBIs in
restrained drivers was classifiable as mild and did not influ-
ence the patient outcomes. All drivers with a GCS score of
13–15 at ED had a GOS score of 4–5 at 30 days (Table 2),
indicating that the GCS score was a reliable prognosticator.
Among variables associated with TBI, it is understandable
that both AMS and LOC are predictive of TBI (Tables 3 and
4). However, AMS is often caused by cerebral concussion,
alcohol intoxication, or hypotension associated with hemor-
rhagic shock resulting in false-positive results. Similarly, the
causes of LOC may be diverse, particularly in drivers without
TBI: seventeen of the 51 drivers without TBI who showed
transient LOC had no superficial injuries to the scalp or face
(Table 3). Majority of the 17 drivers have had preexisting
medical conditions such as diabetes mellitus, and it is possible
that LOC was not the consequence, but was the cause of the
RTAs, at least in some of those drivers.
Analysis on the relationship between CT findings and col-
lision patterns (Figs. 2 and 3) may provide a unique insight
into the pathomechanism of TBI in restrained drivers.

Table 4 Multivariable logistic

regression analysis to identify Variables OR 95% CI p
variables associated with
traumatic brain injury in Age 1.014 0.984–1.045 0.376
restrained drivers involved in road Male sex 1.580 0.368–6.780 0.538
traffic accidents GCS score ≤ 14 4.933 1.135–21.431 0.033*
Loss of consciousness 6.492 1.669–25.249 0.007**
Airbag deployment 0.748 0.430–1.301 0.304
Frontal collision 0.683 0.205–2.272 0.534
Compact (< 660 cc) 1.373 0.398–4.741 0.616
Concomitant AIS ≥ 4 bodily injury 0.430 0.074–2.496 0.347

AIS Abbreviated Injury Scale, CI confidence interval, GCS Glasgow Coma Scale, OR odds ratio
*, **, statistically significant
 Acta Neurochir

Fig. 1 TBI was observed in 6, 9,

and 13% of drivers with frontal,
lateral, and rollover collisions,
respectively (a). The frequency of
TBI after rollover collisions
tended to be higher compared to
that after frontal collisions. The
frequency of severe TBI, defined
as a GCS score ≤ 8, was 14, 40,
and 43% in drivers with frontal,
lateral, and rollover collisions,
respectively (b). TBI was
observed in 10, 7, and 4% of
drivers who drove a Compact,
Sedan, and truck, respectively (c).
No significant difference in the
frequency of TBI had been
observed between drivers of a
Compact and Sedan. Similarly, no
significant difference in the
frequency of severe TBI had been
found between drivers of a
Compact and Sedan (d)

Although spontaneous ICH was a rare cause of RTA with a
frequency of 1.0%, distinction between traumatic and sponta-
neous ICH was usually possible after meticulous history tak-
ing and evaluation of CT findings [12]. Among the injury
types based on CT findings, interhemispheric ASDH and sub-
cortical petechial hemorrhage may be characteristic TBIs in
restrained drivers: interhemispheric ASDH, observed fre-
quently in drivers with frontal collision (Fig. 2), may have
been caused by friction between the falx cerebri and adjacent
medial frontal lobes after sagittal plane deceleration. On the
other hand, subcortical petechial hemorrhage indicative of ax-
onal injury was observed mainly in drivers with lateral colli-
sion (Fig. 3). In lateral collision, axial rotational force is
superimposed to sagittal plane deceleration, and the current
finding of a possible increase in axonal injury after lateral
collision is in agreement with that of a previous clinical study
by Yogarantan et al. [28] and experimental study by Wang et
al. [27].
The higher frequency of lateral/rollover collision in drivers
with TBI (Table 3), as well as the higher frequency/severity of
Acta Neurochir

Fig. 2 Brain computed

tomography images of the 14
drivers with TBI after a frontal
collision, among whom 6 (43%)
had sustained an interhemispheric
acute subdural hematoma (IH-
ASDH)

TBI in drivers with lateral/rollover collision (Fig. 1a, b), was
also consistent with previous studies [2, 6]. The well-known
fact that an airbag is less likely to be deployed in lateral/
rollover collisions compared to frontal collisions may explain
the lower airbag deployment rate in drivers with TBI (Table 3)
[5]. On the other hand, the lack of a significant difference in
the frequency and severity of TBI between drivers of a
Compact and Sedan (Fig. 1c, d) was unexpected. While
Compact cars are popular in Japan, the majority of Sedans
have a relatively small engine capacity (660–2000 cc) and
vehicle weight [20]. In other words, the difference in vehicle
weight between Compact and Sedan types may have been

Fig. 3 Brain computed

tomography images of the 10
drivers with TBI after a lateral
collision, among whom 5 (50%)
had sustained a subcortical
petechial hemorrhage

relatively small, which could explain the lack of difference
between the two vehicle categories. Injuries among truck
drivers had not been analyzed given the small sample size.
This study has some limitations worth noting. First, the
decision to perform WBCT had been made by an on-call
emergency medicine physician after discussing the degree of
vehicular damage with EMS staff. The recognition of severe
vehicular damage may have differed among the EMS staff,
which in turn might have led to difference in the indication of
WBCT [7, 9]. Second, the sample size was small due to the
single-center study design, which may have contributed to the
lack of significant differences (Fig. 1a, b). Third, the classifi-
cation of vehicles according to engine capacity may not be
universal considering that while Minicars with an engine ca-
pacity of < 660 cc are popular in our country [20]; they are
much less used in European and North American countries.
Therefore, the current findings may not directly be applicable
to such countries. Moreover, the classification of collision
patterns into the three categories may also have been too sim-
plistic from the viewpoint of crash investigators. Rear-end
collisions were excluded because their frequency is relatively
small and the many drivers who sustain rear-end collisions do
not seek medical attention [1]. Finally, information on airbag
deployment was available only in two thirds of the 409 drivers
(Table 1), which may have impaired data reliability.
Conclusions
In this study, 7.6% of restrained drivers whose vehicle had
been severely damaged in a RTA sustained TBI. Majority of
the TBIs were mild without need for neurosurgical interven-
tion. Notably, TBI was present in 2.9% of drivers with a GCS
score of 15. The presence of AMS and LOC was predictive of
TBI. The frequency and severity of TBI tended to be higher in
drivers with rollover collisions than in those with frontal col-
lisions. While association seems to exist between the types of
collision and types of brain injury sustained by restrained
drivers, further studies are warranted to verify our findings.
Funding The corresponding author (JI) received a research grant from
the General Insurance Association of Japan. The sponsor had no role in
the design or conduct of this research.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
Ethical approval 1. All procedures performed in studies involving hu-
man participants were in accordance with the ethical standards of the
institutional and/or national research committee and with the 1964
Helsinki Declaration and its later amendments or comparable ethical
standards.
Acta Neurochir
2. Approval for this study was given by our institutional research
committee (Study ID: 2017-13).
3. For this type of (i.e., retrospective) study, our institutional research
committee decided that formal consent was not required from each
patient.
References
1. Abu-Zidan FM, Eid HO (2015) Factors affecting injury severity of
vehicle occupants following road traffic collisions. Injury 46:136–141
2. Bazarian JJ, Fisher SG, Flesher W, Lillis R, Knox KL, Pearson TA
(2004) Lateral automobile impacts and the risk of traumatic brain
injury. Ann Emerg Med 44:142–152
3. Braakman R (1992) Early prediction of outcome in severe head
injury. Acta Neurochir 116:161–163
4. Carroll CP, Cochran JA, Price JP, Guse CE, Wang MC (2010) The
AIS-2005 revision in severe traumatic brain injury: mission accom-
plished or problems for future research? Ann Adv Automot Med
54:233–238
5. Cleveland N, Colwell C, Douglass E, Hopkins E, Haukoos JS
(2014) Motor vehicle crash severity estimations by physicians and
prehospital personnel. Prehosp Emerg Care 18:402–407
6. Conroy C, Hoyt DB, Eastman AB, Erwin S, Pacyna S, Holbrook
TL, Vaughan T, Sise M, Kennedy F, Velky T (2006) Rollover
crashes: predicting serious injury based on occupant, vehicle, and
crash characteristics. Accid Anal Prev 38:835–842
7. Croft AC, Freeman MD (2005) Correlating crash severity with
injury risk, injury severity, and long-term symptoms in low velocity
motor vehicle collisions. Med Sci Monit 11:RA316–RA321
8. Davies RM, Scrimshire AB, Sweetman L, Anderton MJ, Holt EM
(2016) A decision tool for whole-body CT in major trauma that
safely reduces unnecessary scanning and associated radiation risks:
an initial exploratory analysis. Injury 47:43–49
9. Desapriya E, Chipman M, Joshi P, Pike I (2005) The risk of injury
and vehicle damage in vehicle mismatched crashes. Int J Inj Control
Saf Promot 12:191–192
10. Hong ZJ, Chen CJ, Yu JC, Chan DC, Chou YC, Liang CM, Hsu SD
(2016) The evolution of computed tomography from organ-
selective to whole-body scanning in managing unconscious patients
with multiple trauma: a retrospective cohort study. Medicine
(Baltimore) 95:e4653
11. Inamasu J, Guiot BH (2007) Thoracolumbar junction injuries after
motor vehicle collision: are there differences in restrained and
nonrestrained front seat occupants? J Neurosurg Spine 7:311–314
12. Inamasu J, Nakatsukasa M, Tomiyasu K, Mayanagi K, Nishimoto
M, Oshima T, Yoshii M, Miyatake S, Imai A (2018) Stroke while
driving: frequency and association with automobile accidents. Int J
Stroke 13:301–307
13. Joseph B, Pandit V, Aziz H, Kulvatunyou N, Zangbar B, Green DJ,
Haider A, Tang A, O'Keeffe T, Gries L et al (2015) Mild traumatic
brain injury defined by Glasgow Coma Scale: is it really mild?
Brain Inj 29:11–16
14. Kahn J, Grupp U, Kaul D, Böning G, Lindner T, Streitparth F
(2016) Computed tomography in trauma patients using iterative
reconstruction: reducing radiation exposure without loss of image
quality. Acta Radiol 57:362–369
15. Kimura A, Tanaka N (2013) Whole-body computed tomography is
associated with decreased mortality in blunt trauma patients with
moderate-to-severe consciousness disturbance: a multicenter, retro-
spective study. J Trauma Acute Care Surg 75:202–206
16. Linder F, Mani K, Juhlin C, Eklöf H (2016) Routine whole body CT
of high energy trauma patients leads to excessive radiation expo-
sure. Scand J Trauma Resusc Emerg Med 24:7
Acta Neurochir
17. Majdzadeh R, Khalagi K, Naraghi K, Motevalian A, Eshraghian
MR (2008) Determinants of traffic injuries in drivers and motorcy-
clists involved in an accident. Accid Anal Prev 40:17–23
18. Newgard CD, Martens KA, Lyons EM (2002) Crash scene photog-
raphy in motor vehicle crashes without air bag deployment. Acad
Emerg Med 9:924–929
19. Price SJ, Suttner N, Aspoas AR (2003) Have ATLS and national
transfer guidelines improved the quality of resuscitation and trans-
fer of head-injured patients? A prospective survey from a Regional
Neurosurgical Unit. Injury 34:834–838
20. Road Traffic Act of Japan, Article 3 (English translation) http://
www.japaneselawtranslation.go.jp/law/detail/?id=2962&vm=
04&re=02 (Accessed May 3, 2018)
21. Rosi Junior J, Figueiredo EG, Rocha EP, Andrade AF, Rasslan S,
Teixeira MJ (2012) Whole-body computerized tomography and
concomitant spine and head injuries: a study of 355 cases.
Neurosurg Rev 35:437–444
22. Schulman CI, Wilbur V, Leibowitz B, Labiste L, Perdeck E,
Bahouth G, Digges K, Augenstein JS (2007) The SceneScore for
improved pre-hospital triage of motor-vehicle crash victims. Annu
Proc Assoc Adv Automot Med 51:49–60
23. Tagliaferri F, Compagnone C, Yoganandan N, Gennarelli TA
(2009) Traumatic brain injury after frontal crashes: relationship
with body mass index. J Trauma 66:727–729
24. Teasdale G, Jennett B (1976) Assessment and prognosis of coma
after head injury. Acta Neurochir 34:45–55
25. Traffic Accident Data Project Steering Committee (1984) Vehicle
damage scale for traffic accident investigators, Third edn. NSC
Publications, Chicago
26. Vaca FE, Anderson CL, Herrera H, Patel C, Silman EF, Deguzman
R, Lahham S, Kohl V (2009) Crash injury prediction and vehicle
damage reporting by paramedics. West J Emerg Med 10:62–67
27. Wang HC, Duan ZX, Wu FF, Xie L, Zhang H, Ma YB (2010) A new
rat model for diffuse axonal injury using a combination of linear
acceleration and angular acceleration. J Neurotrauma 27:707–719
28. Yoganandan N, Gennarelli TA, Zhang J, Pintar FA, Takhounts E,
Ridella SA (2009) Association of contact loading in diffuse axonal
injuries from motor vehicle crashes. J Trauma 66:309–315