Survey of traumatic intracranial hemorrhage in Taiwan.

Abstract

Injury ranked third among the top 10 leading causes of death in Taiwan from 1964 to 1996 and is still among the top 10 presently. Among transportation accidents, motor vehicle-related injury had the highest incidence rate, often resulting in traumatic head injury.
This survey was conducted from July 1, 1994, to June 30, 2002, and was collected from 55 major hospitals in Taiwan. A total of 90250 patients with TBI were enrolled, and 27585 cases were identified to have TIH. In this study, SPSS 10.0 (SPSS, Chicago, Ill) was used to process the data. Regarding the rating model itself, reliability and correlation tests were conducted to calculate the coefficiency, and factor analysis was carried out to verify its validity.
The incidence rate of male-to-female ratio was 2.65. Traffic injuries (67.6%) were the leading causes of TIH. Among the traffic injuries, motorcycle-related traumatic injuries had the highest incidence rate (69.6%). In the logistic regression analysis, older patients had the highest risk of developing TIH. Patients without a motorcycle helmet had a higher risk (odds ratio, 1.40) of developing TIH than those with a helmet. As regards the types of injuries, pedestrian injury (odds ratio, 1.61) had the highest risk of developing TIH.
Although traffic injuries, especially those caused by motorcycles, are the major cause of intracranial hemorrhage, we have to pay due attention to falling, pedestrian, and bicycle injuries, as these are all major causes of intracranial hemorrhage.

Full text

Page 1

Trauma
Survey of traumatic intracranial hemorrhage in Taiwan
Jia-Wei Lin, MDa,1, Shih-Han Tsai, MD, PhDa,b,1, Wan-Chen Tsai, MSd,
Wen-Ta Chiu, MD, PhDa,e,4, Shu-Fen Chu, RNe, Chien-Min Lin, MDa,
Che-Ming Yang, MD, JDc, Ching-Chang Hung, MD, PhDa
Departments ofaNeurosurgery,bEmergency Medicine,cNuclear Medicine, Taipei Medical University–Wanfang Hospital, Taipei 116, Taiwan
dAdministration Office, Taipei Medical University–Municipal Wanfang Hospital, Taipei 116, Taiwan
eInstitute of Injury Prevention and Control, Taipei Medical University, Taipei 110, Taiwan
Received 25 April 2006; accepted 29 April 2006
AbstractBackground: Injury ranked third among the top 10 leading causes of death in Taiwan from 1964 to
1996 and is still among the top 10 presently. Among transportation accidents, motor vehicle–related
injury had the highest incidence rate, often resulting in traumatic head injury.
Methods: This survey was conducted from July 1, 1994, to June 30, 2002, and was collected from
55 major hospitals in Taiwan. A total of 90250 patients with TBI were enrolled, and 27585 cases
were identified to have TIH. In this study, SPSS 10.0 (SPSS, Chicago, Ill) was used to process the
data. Regarding the rating model itself, reliability and correlation tests were conducted to calculate
the coefficiency, and factor analysis was carried out to verify its validity.
Results: The incidence rate of male-to-female ratio was 2.65. Traffic injuries (67.6%) were the
leading causes of TIH. Among the traffic injuries, motorcycle-related traumatic injuries had the
highest incidence rate (69.6%). In the logistic regression analysis, older patients had the highest risk
of developing TIH. Patients without a motorcycle helmet had a higher risk (odds ratio, 1.40) of
developing TIH than those with a helmet. As regards the types of injuries, pedestrian injury (odds
ratio, 1.61) had the highest risk of developing TIH.
Conclusions: Although traffic injuries, especially those caused by motorcycles, are the major cause
of intracranial hemorrhage, we have to pay due attention to falling, pedestrian, and bicycle injuries,
as these are all major causes of intracranial hemorrhage.
D 2006 Elsevier Inc. All rights reserved.
Keywords:
Head injury; Traumatic intracranial hemorrhage; Traumatic brain injury
1. Introduction
Injury ranked third among the top 10 leading causes of
death in Taiwan from 1964 to 1996. Since July 1997, the
law regarding compulsory helmet wearing by all motorcycle
riders has been in force, and this helmet law reduced the
head trauma cases by one third and the number of cranial
operations by 42%, which results in a drop in the ranking of
injury to fifth place. However, the mortality of injury
remains high in Taiwan especially among individuals aged
younger than 34 years [1,2].
Among the transportation injuries in Taiwan, motor
vehicle–related injury has the highest incidence rate and
often results in traumatic head injury. Intracranial hemor-
rhage complicates the TBI occasionally, and this miserable
situation occurs mainly in young males. Because of the
high productivity and work ability of the young people,
especially young males, such injuries will affect their
welfare and that of their family, and even the whole society
as well. Family income may be affected if a young man is
disabled by accidental head injury. From the viewpoint of
social economy, the huge medical expenses from long-term
0090-3019/$ – see front matter D 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.surneu.2006.04.005
Abbreviations: GCS, Glasgow Coma Scale; GOS, Glasgow Outcome
Scale; ICD-9, international classification of diseases (ninth edition); TBI,
traumatic brain injury; TIH, traumatic intracranial hemorrhage.
4 Corresponding author. Taipei Municipal Wanfang Hospital, Taipei
116, Taiwan, Tel.: +886 2 29307930x8819; fax: +886 2 29335221.
E-mail address: wtchiu@tmu.edu.tw (W.-T. Chiu).
1These authors contributed equally to this article.
Surgical Neurology 66 (2006) S2:20–S2:25
www.surgicalneurology-online.com

Page 2

treatment and rehabilitation will not only seriously consume
medical resources but also affect the overall economic
development and increase the burden of the whole country
indirectly. Therefore, we should consider this as a key issue
in determining the policy of national health care [3].
To determine the severity and prognosis of intracranial
hemorrhage caused by head trauma in Taiwan, we carried
out a study on 90250 head trauma cases from 55 hospitals
from July 1, 1994, to June 30, 2002. There were a total of
27585 head trauma–related intracranial hemorrhage cases.
By studying the causes, severity, and prognosis of these
cases, we made an attempt to learn more details about
traumatic intracranial bleeding and to discover the risk
factors for future prevention of the complication.
2. Methods
2.1. Study subjects
Basing on ICD-9, trained personnel reviewed from
July 1, 1994, to June 30, 2002, the head injury admission
notes and diagnostic charts of 55 hospitals in Taiwan, and
transcribed those head trauma medical records into a
structured questionnaire. All collaborating hospitals had
more than 100 beds and also had adequate facilities and
experience for head trauma management. The questionnaire
used for this study had been tested for both reliability and
validity, and the basic information about head trauma such
as the medical history, sex, age, head trauma condition and
severity (GCS) were recorded. The questionnaire also
addressed the date of hospital admission, the presence or
absence of intracranial hemorrhage and the bleeding area,
the injury condition, whether or not a helmet was worn,
treatment status, and outcome (GOS). The study focused
only on hospitalized cases, and outpatient or dead-on-arrival
cases were excluded.
2.2. Head trauma definition
Head trauma was defined as obvious brain injury,
concussion, or skull fracture. The definition of clinical
manifestations included loss of consciousness, posttraumat-
ic amnesia, and neurologic deficits. The presence of any one
of the abovementioned signs or symptoms was diagnosed as
head trauma [4-6].
2.2.1. Glasgow Coma Scale
The GCS is an index for assessing patient’s conscious-
ness. By grading an individual’s eye opening, motor
responses, and verbal responses, the level of consciousness
can be assessed. The highest score for an adult is 15,
whereas the lowest score is 3. Hung et al [7] classified the
GCS into 3 levels: severe injury (3-8 points), moderate
injury (9-12 points), and mild injury (13-15 points).
2.2.2. Glasgow Outcome Scale
The GOS was used to categorize the outcome of patients
with TBI at the time of hospital discharge as follows:
(1) death; (2) persistent vegetative state; (3) severe
disability—conscious but dependent; (4) moderate disabil-
ity—disabled but independent; and (5) good recovery [6,8].
2.2.3. Traumatic intracranial hemorrhage definition
Traumatic intracranial hemorrhage was based on com-
puted tomography scan diagnosis and defined as head
trauma–related injury. Traumatic intracranial hemorrhage
included epidural hematoma, subdura hematoma, intracere-
bral hemorrhage, subarachnoid hemorrhage, and intraven-
tricle hemorrhage. If there were more than 2 kinds of
hemorrhage identified by the computed tomography
scan, we registered this patient to each category of
hemorrhage simultaneously.
Fig. 1. Correlation of age and sex in intracranial hemorrhage.
Fig. 2. Causes of intracranial hemorrhage.
J.-W. Lin et al. / Surgical Neurology 66 (2006) S2:20–S2:25
S2:21

Page 3

2.3. Data processing and analysis
Researchers collected the questionnaires from each
hospital regularly. After decoding the collected question-
naires, all data were saved in an Excel file followed by
processing with SPSS 10.0 for statistical analysis (SPSS,
Chicago, IL, USA). Frequency distribution and percentage
were used as basic descriptive statistical data, followed by
the v2test. Logical regression analysis was then performed
to those variables with a significant difference.
3. Results
There were 90250 head trauma cases from 55 hospitals
in Taiwan. A total of 27585 patients were diagnosed to have
intracranial hemorrhage, including 19865 males (72.6%)
and 7514 females (27.4%). The male-to-female ratio was
2.7(Fig. 1). In the mortality rate of intracranial hemorrhage,
male deaths occurred 3 times as often as female deaths
(2369 [75.0%] vs 789 [25.0%]). Among all the age groups,
those from 20 to 29 years carried the highest rate (4852
[17.7%]) (Fig. 1).
Among various causes of intracranial hemorrhage, the
most common was traffic injuries (18573 [67.6%]),
followed by falls (6873 [25.0%]) and assaults (945
[3.4%]). Further analysis of the types of vehicles causing
injuries shows that motorcycle injuries were the first with
12052 cases (69.6%); pedestrian accidents were the second
with 2066 cases (11.9%). Car injuries were the cause in
1623 cases (9.3%), bicycles in 1030 cases (5.9%), and the
other causes, such as train injuries, were rare (Fig. 2).
For the major location of the hemorrhage, we found that
subarachnoid hemorrhage occurred in 13216 (47.9%),
subdural in 11201 (40.6%), epidural in 4822 (17.5%),
intracerebral in 3854 (14.0%), and intraventricular in 1186
(4.3%) (Fig. 3).
By using the GCS to assess the severity of head trauma
with intracranial hemorrhage, we found 13242 (52.5%)
mild cases, 5154 (20.4%) moderate cases, and 6849 (27.1%)
severe cases. Based on the results of the GOS, we observed
3180 (13.5%) individuals who died from intracranial
hemorrhage, 610 (2.6%) who became persistently vegeta-
tive, 2435 (10.3%) who remained severely disabled, 2767
(11.7%) who had moderate disability, and 14581 (61.9%)
who had a good recovery (Fig. 4).
v2Analysis was used to determine the correlation among
variables of sex, age, cause of injury, and hemorrhage
condition within the group of TIH. The analysis showed
that males tended to have a higher chance of intracra-
nial hemorrhage than females (33.3% vs 25.8%), and that
the difference was statistically significant (P b .001).
Fig. 4. Correlation of intracranial hematoma and severity (GCS) and
prognosis (GOS).
Table 1
Correlation between variables of sex, age, cause of injury, and hemorrhage
condition among intracranial trauma cases
Intracranial
hemorrhage
Yes, n (%) No, n (%)Total
P
Sex
Male
Female
Age
0-9 y
10-19 y
20-29 y
30-39 y
40-49 y
50-59 y
60-69 y
z70 y
Causes
Car injuries
Falls
Others
Types of injuries
Motorcycles
Bicycles
Pedestrian
Automobiles
Others
Types of falls
High-height
(N1 m)
Low-height
(V1 m)
Tripped/tumbled
Wearing helmets
Yes
No
b.001
19865 (33.3%)
7514 (25.8%)
39745 (66.7%)
21606 (74.2%)
59610
29120
b.001
1074 (24.8%)
3407 (27.3%)
4852 (24.1%)
3472 (25.7%)
3429 (28.9%)
3099 (35.3%)
3457 (42.8%)
4686 (47.3%)
3260 (75.2%)
9078 (72.7%)
15313 (75.9%)
10038 (74.3%)
8417 (71.1%)
5676 (64.7%)
4613 (57.2%)
5218 (52.7%)
4334
12485
20165
13510
11846
8775
8070
9904
b.001
18573 (31.6%)
6873 (36.2%)
2035 (18.1%)
40256 (68.4%)
12089 (63.8%)
9202 (81.9%)
58829
18962
11237
b.001
12052 (31.2%)
1030 (37.4%)
2066 (39.0%)
1623 (27.1%)
546 (32.3%)
26556 (68.8%)
1724 (62.6%)
3236 (61.0%)
4371 (72.9%)
1145 (67.7%)
38608
2754
5302
5994
1691
b.001
1944 (41.6%)2724 (58.4%) 4668
1517 (33.7%) 2983 (66.3%)4500
2526 (33.8%)4945 (66.2%)7471
b.001
1906 (25.2%)
7646 (32.9%)
5670 (74.8%)
15583 (67.1%)
7576
23229
Fig. 3. Sites for intracranial hemorrhage.
J.-W. Lin et al. / Surgical Neurology 66 (2006) S2:20–S2:25
S2:22

Page 4

Correlation between age stratification and intracranial
hemorrhage showed a significantly higher incidence in older
patients (N50 years old) with head trauma (P b .001).
Among the causes of head trauma, we found that injury
caused by falling had a higher incidence of intracranial
hemorrhage (36.2%) than traffic injuries (31.6%), and this
was also statistically significant (P b .001). Comparison of
traffic-accident types showed that pedestrian injuries had the
highest rate (39.0%) of intracranial hemorrhages; bicycle
injuries came next (37.4%), followed by motorcycle (31.2%)
and car accidents (27.1%) (P b.001). As to falling injuries,a
fall from a high place (z1 m) had a higher incidence for
intracranial hemorrhage (41.6%) than a fall from a lower
place (b1 m) (33.7%) and a fall on slippery floors (33.8%).
Among motorcycle incidents, helmet wearing was the key
factor affecting intracranial hemorrhage. People who did not
wear a helmet had a higher rate of intracranial hemorrhage
than those who did (32.9% vs 25.2%); the difference was
statistically significant (P b .001) (Table 1).
The correlation between severity (GCS) and intracranial
hemorrhage makes it obvious that severely injured patients
had a far greater chance than patients with minor injuries to
develop intracranial hemorrhage (27.1% vs 2.4%). Those
who did not have intracranial hemorrhage mainly sustained
mild injury (92.4%); yet among those who had intracranial
hemorrhage, 52.5% had mild injury. This result showed
a statistically significant difference (P b .001). When
studying the correlation between prognosis (GOS) and
intracranial hemorrhage, we found that deaths among cases
with intracranial hemorrhage were much more frequent than
among cases without intracranial hemorrhage (13.5% and
0.6%, respectively). In fact, 91.7% of all cases without
intracranial hemorrhage had a good prognosis, yet only
61.9% of cases with intracranial hemorrhage had a good
prognosis. The data also showed statistical significance
(P b .001) (Table 2).
Finally, we added the variables such as injury causes,
vehicle-accident types, falling, and wearing of a helmet or
not at the time of the injury, and reanalyzed those variables
with age and sex to obtain the most significant risk factors
for intracranial hemorrhage (Table 3). We found that injury
caused by falling carried a risk 2.34 times (95% confidence
interval [CI], 2.21-2.48) higher than that of other variables,
and the risk was 2.18 times higher for traffic injuries
(95% CI, 2.07-2.29). As to the types of traffic injuries,
pedestrian and bicycle injuries had the highest chances of
developing intracranial hemorrhage, that is, 1.61 times (95%
CI, 1.48-1.75) and 1.55 times (95% CI, 1.41-1.71),
respectively. For the causes of falling injuries, falls from a
higher place (N1 m) had a greater chance (1.64 times) than
those on slippery floors (95% CI, 1.52-1.78) to be com-
plicated with intracranial hemorrhage. In the motorcycle
injuries, the chance of having intracranial hemorrhage
among motorcyclists not wearing a helmet was 1.4 times
higher than among those wearing a helmet (95% CI,
1.32-1.48) (Table 3).
4. Discussion
As far as our survey has revealed, Taiwan probably has
the highest mortality rate of accidental injury in the world
[9]. A review in 1983 showed that there were about 12.5%
traumatic injuries with head trauma. Further analysis of the
causes of mortality in these traumatic patients shows that
Table 2
Correlation between severity (GCS) and intracranial hemorrhage, and
correlation between prognosis (GOS) and intracranial hemorrhage
Intracranial
hemorrhage
Yes, n (%)No, n (%)Total
P
Severity (GCS)
Severe
Moderate
Minor
Total
Prognosis (GOS)
Death
Persistent
vegetative state
Severe disability
Moderate disability
Good recovery
Total
b.001
6849 (27.1%)
5154 (20.4%)
13242 (52.5%)
25245
1441 (2.4%)
3219 (5.3%)
56265 (92.4%)
60925
8290
8373
69507
86170
b.001
3180 (13.5%)
610 (2.6%)
370 (0.6%)
98 (0.2%)
3550
708
2435 (10.3%)
2762 (11.7%)
14581 (61.9%)
23568
1160 (2.0%)
3160 (5.4%)
53206 (91.7%)
57994
3595
5922
67787
81562
Table 3
The most significant risk factors affecting intracranial hemorrhage
ORa
95% CIORb
95% CI
Sex
Male
Female
Age
z70 y
60-69 y
50-59 y
40-49 y
30-39 y
20-29 y
10-19 y
0-9 y
Causes
Car injuries
Falls
Others
Types of car injuries
Motorcycles
Bicycles
Pedestrian
Others
Automobiles
Types of falls
High-height
Low-height
Falls
Wearing helmets
No
Yes
1.44
1
1.40-1.48
–
1.48
1
1.43-1.53
–
2.73
2.28
1.66
1.24
1.05
0.96
1.14
1
2.52-2.95
2.10-2.47
1.53-1.80
1.14-1.34
0.97-1.14
0.89-1.04
1.05-1.23
–
2.72
2.22
1.64
1.22
1.01
0.93
1.11
1
2.51-2.95
2.05-2.41
1.51-1.79
1.12-1.32
0.94-1.10
0.86-1.00
1.02-1.20
–
2.09
2.57
1
1.98-2.20
2.43-2.72
–
2.18
2.34
1
2.07-2.29
2.21-2.48
–
1.22
1.61
1.72
1.28
1
1.15-1.30
1.46-1.77
1.59-1.86
1.14-1.44
–
1.28
1.55
1.61
1.27
1
1.21-1.37
1.41-1.71
1.48-1.75
1.13-1.43
–
1.40
0.99
1
1.30-1.51
0.92-1.10
–
1.64
1.14
1
1.52-1.78
1.05-1.24
–
1.44
1
1.36-1.52
–
1.40
1
1.32-1.48
–
aCrude odds ratio.
bAdjustment for age and sex.
J.-W. Lin et al. / Surgical Neurology 66 (2006) S2:20–S2:25
S2:23

Page 5

55% of the cases were directly related with head injuries.
Among these patients with head-injury, about 30% had
intracranial hemorrhage. Such a high incidence of TIH
compelled us to face this issue seriously.
In our research on patients with intracranial hemorrhage,
19865 (72.6%) were males—much more numerous than
females (7514 [27.4%]). Furthermore, the mortality rate of
intracranial hemorrhage among male patients was 3 times
higher than that among female patients. This result may be
explained by the fact that males are most often in charge of
high-risk jobs, and this gives them a higher chance of being
exposed to danger. Besides, most males are more active than
females and also tend to enjoy adventures or exciting
activities [10].
An analysis of the head injury–related intracranial
hemorrhages with age stratification shows that the age
group from 20 to 29 years had the highest incidence, which
was about 17.7% among all the patients with TIH. The age
in this range coincides with the adolescence-adult period.
People in this age group are usually very energetic, showing
a lack of an adequate coping system. Sometimes they prefer
to release the pressure by way of excessively speedy riding
or driving of the vehicles. Therefore, it is understandable
why so many injuries occurred and caused the highest
incidence of TIH in this age group [11].
As to the cause of TIH, 67.6% of the cases were primary
due to traffic injuries. Studies from other countries also
show that traffic injuries are the most frequent cause,
followed by falling [4,12-20]. In addition, the type of
vehicles involved in the injuries was found to be primarily
motorcycles (69.6%). The data on head injuries also showed
that falling accidents caused intracranial hemorrhage more
often than car injuries. However, among car injuries,
pedestrian injuries had the highest chance, exceeding that
of bicycle injuries with reference to intracranial hemor-
rhage. As motorcycles are the most popular transportation
vehicles used in Taiwan, their number has increased
dramatically in recent years. Because the incidence of
traffic injuries is very high, and also because of the high
proportion of motorcycles, motorcycle injuries are by far the
most frequent cause of injury. These facts may divert our
attention away from accidents caused by falling, pedestrian,
and bicycles, and this will be another important issue
requiring further evaluation.
As shown in Table 1, among all motorcycle injuries that
caused intracranial hemorrhage, the number of those
patients without helmet protection was 12% higher than
that of those with a helmet. Therefore, in 1997, the
Taiwanese government decreed that every motorcyclist
wear a helmet for cranial protection, and this resulted in a
dramatic decease in the number of head injuries, thus
proving the great value of helmet wearing [21]. How-
ever, there are still some injuries caused by motorcyclists
without a license or those under the legitimate age for
qualification. These illegal motorcyclists are not mature
enough to handle traffic situations adequately and, therefore,
may endanger pedestrians, other motorcyclists, or car drivers
on the road. The government should pay more attention to
this serious problem.
5. Conclusion
It is true that in traffic injuries, motorcycle injuries are the
primary cause of intracranial hemorrhage, but we should not
forget to pay proper attention to falling, pedestrian, and
bicycle injuries, as they are also major causes of intracranial
hemorrhage. Policies and laws related to prevention of
intracranial hemorrhage need to be enforced not only on
falling, pedestrian, and bicycle injuries but also on how to
reduce the severity of injury when injuries have taken place.
We believe that an improvement in the traffic environment,
for example, by installing safety equipment (railings or
armrests), can reduce falling injuries, and rearranging
sidewalks for pedestrians and bicycles and strictly prohibit-
ing car entry can also help to reduce injuries. Furthermore,
we may consider the helmet law for bicyclists. This bicycle
helmet law has been enforced in some countries for more
than 10 years. Australia was the first to enforce this law and
reported a reduction in the incidence of head injuries by
39% [22]. The United States is also promoting helmet
use for biking, which has also considerably reduced the
medical expenses and mortality rate [23].
Many policies, such as compulsory helmet wearing and
inspection of drunken drivers, have been implemented in
recent years and have reduced the mortality rate of
accidental injury. In 2002, 8489 people died from injuries
in Taiwan; thus, injury dropped from third to fifth place
among the 10 most common causes of death (Fig. 5).
With this remarkable achievement, the government is still
focusing its primary attention on motorcycle- and vehicle-
injury prevention. We can be sure that if we make
proper efforts in the prevention of falling, pedestrian, and
bicycle injuries, the mortality rate from injuries will
Fig. 5. Trend of Injury-related deaths in Taiwan, 1994-2002.
J.-W. Lin et al. / Surgical Neurology 66 (2006) S2:20–S2:25
S2:24