Resuscitation 81 (2010) 440445

Contents lists available at ScienceDirect

Resuscitation
journal homepage: www.elsevier.com/locate/resuscitation

Clinical paper

The importance of pre-trauma centre treatment of life-threatening events on the

mortality of patients transferred with severe trauma
Ernestina Gomes a, , Rui Arajo a , Antnio Carneiro a , Cludia Dias b ,
Altamiro Costa-Pereira c , Fiona E. Lecky d
a

Unidade de Cuidados Intensivos Polivalente, Centro Hospitalar do Porto, Hospital de Santo Antnio, 4099-001 Porto, Portugal
Servico de Bioestatstica e Informtica Mdica, Faculdade de Medicina da Universidade do Porto, Alameda Professor Hernni Monteiro, 4200-319 Porto, Portugal
CINTESIS (Centro de Investigaco em Tecnologias da Sade e Sistemas de Informaco em Sade), Servico de Bioestatstica e Informtica Mdica,
Faculdade de Medicina da Universidade do Porto, Alameda Professor Hernni Monteiro, 4200-319 Porto, Portugal
d
Trauma Audit and Research Network, University of Manchester, Manchester Academic Health Sciences Centre, Hope Hospital, Salford M6 8HD, United Kingdom
b
c

a r t i c l e

i n f o

Article history:
Received 24 May 2009
Received in revised form
15 December 2009
Accepted 20 December 2009

Keywords:
Trauma
Mortality
Trauma centre
Pre-hospital
Trauma system
Life-threatening event

a b s t r a c t
Aim: The benet of a well organised trauma system is acknowledged but doubts remain concerning the
optimal pre-hospital trauma care model. We hypothesise that the treatment of life-threatening events
before arrival at trauma centre either pre-hospital or rst hospital may be more relevant to decreasing
mortality than shortening the time to trauma centre.
Methods: A cohort of 727 trauma patients with life-threatening events identied as airway, breathing, circulation or neurological disability requiring transfer to a trauma centre were studied. Data on
patients characteristics, trauma features, and mortality were taken from a trauma registry. Patients were
divided into 3 groups depending on the place of treatment of life-threatening events: pre-hospital, rst
hospital or trauma centre. Survival KaplanMeier curves and logistic regression were used to assess the
effect of place of treatment of life-threatening events on mortality.
Results: Patients from the pre-hospital and rst hospital groups had 20% and 27% mortality respectively,
compared to 38% among those whose life-threatening events were corrected only at the trauma centre.
Logistic regression showed that patients whose life-threatening events were corrected only at the trauma
centre had an odds of death 3.3 times greater than those from the pre-hospital group, adjusted for patient
and trauma characteristics and time to trauma centre.
Conclusion: In trauma patients requiring transfer to a trauma centre, pre-hospital interventions to treat
life-threatening events may signicantly decrease mortality when compared to similar interventions
performed later at the trauma centre.
2010 Elsevier Ireland Ltd. All rights reserved.

1. Introduction
The benet of a well organised trauma system with specialized trauma centres is fully acknowledged,14 but doubts
still remain concerning what could be considered as the optimal pre-hospital trauma care. Several authors have studied
the effect of pre-hospital advanced life support (ALS) in comparison with pre-hospital basic life support (BLS) approaches
on trauma mortality and other outcomes.59 However, there
are no large or controlled studies on that subject and the
controversy is far from being resolved. Even studies that

A Spanish translated version of the summary of this article appears as Appendix

in the nal online version at doi:10.1016/j.resuscitation.2009.12.014.
Corresponding author. Tel.: +351 917616900; fax: +351 222009483.
E-mail address: [email protected] (E. Gomes).
0300-9572/$ see front matter 2010 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.resuscitation.2009.12.014

address the effect of specialized trauma teams and helicopters on trauma patients outcome showed conicting
results.1012
Moreover, studies that compare pre-hospital ALS and BLS
approaches often have different denitions for ALS and BLS, study
only partial ALS attitudes or techniques instead of the whole concept and above all do not study and compare ALS and BLS in a
population of trauma patients with life-threatening events needing
treatment.13
Traumatic life-threatening events were dened in the medical
literature as events that endanger life and that should be corrected
during the primary survey of treatment of trauma patients. By
denition life-threatening events should be treated urgently. The
methodology to identify and treat life-threatening events is also
known as ABCD methodology and is taught in different trauma
courses around the world namely European Trauma Course14 and
ATLS.15

E. Gomes et al. / Resuscitation 81 (2010) 440445

Our aim was to estimate the effect on mortality of treating the

life-threatening events prior to arrival at the trauma centre (either
in the pre-hospital scene or in the rst hospital) in comparison
with treatment of life-threatening events only at the trauma centre.
We considered two hypotheses: (1) The early treatment of lifethreatening events in trauma patients decreases the risk of death.
(2) Time from accident to denitive treatment at trauma centre is
an independent factor contributing to mortality.
2. Methods
2.1. Setting
We analysed data from a trauma registry that includes all
patients admitted to a neurosurgical centre in the North of Portugal (Hospital de Santo AntnioHGSA). HGSA is a 700 bedded
University acute central hospital with neurosurgical cover and
intensive care facilities. No other hospital within this trauma system has emergency neurosurgical coverage. HGSA serves Northeast
Portugal (a mixed urban/rural area) with a catchment area of
approximately 2,500,000 inhabitants. The Emergency Medical System (EMS) within this region consists of a dispatch centre that
answers all emergency calls (112) and sends the most appropriate resource available. One helicopter and 10 fast vehicles with
trained medical doctors are available for this response.16 The decision whether or not to dispatch a physician to the eld is reached on
the basis of the information received on the phone and the available
resources. As the resources are not uniform for the same severity of
trauma, different approaches are possible from an ALS vehicle with
doctor and nurse to an BLS Ambulance staffed only by re-ghters
with basic skills. If an ALS resource is not available BLS ambulances
attend the scene and transport the patient to the nearest hospital
with an emergency department (ED). Such ALS teams with medical
involvement are only available in urban areas. As a consequence the
level of trauma care in the designated area is not uniform. In Portugal the trauma centre levels, according to the American College of
Surgeons,17 are not used for accreditation; most rst hospitals with
ED are comparable to level III trauma centres and if a neurosurgical
and/or intensive care is needed the patient is transferred to HGSA,
the level I trauma centre for this region.
2.2. Study design, patients and denitions
A trauma registry at HGSA has been prospectively maintained
using Trauma Injury Severity Score methodology (TRISS) for severity coding.18 For the purpose of this study we analysed data
between August 2001 and December 2006. In addition the following information was collected after retrospective review of the
patients clinical notes: type of life-threatening event; type of lifethreatening event treatment and location where the treatment was
performed (pre-hospital, rst hospital or trauma centre).
2.2.1. Patients
We included in this study all trauma patients admitted to the
emergency room of HGSA that were transferred from another hospital in the trauma system, corresponding to 72% of the severe
trauma patients admitted to this trauma centre.
2.2.2. Life-threatening events
Life-threatening events were dened as problems in the airway
(A), breathing (B), circulation (C) or neurological disability (D) that
may endanger life if not immediately identied and controlled as
taught in trauma courses. The airway breathing circulation disability approach (ABCD approach) is a method of identifying and
treating life-threatening events in trauma patients. It is taught
in different trauma courses and emergency medicine courses for

441

doctors, nurses and paramedics around the world.14,15,19 At the

end of the primary survey patients should have a patent airway,
have oxygen on and have the cervical spine immobilized. Adequate ventilation must be conrmed afterwards. This includes an
examination of the chest to clinically exclude or treat immediately
life-threatening thoracic conditions. Tension pneumothorax, open
chest wound, massive haemothorax, ail chest or cardiac tamponade are the thoracic life-threatening events. Regarding circulation
any overt bleeding must be controlled by direct pressure. If a lifethreatening event is identied in circulation (clinical evidence of
shock using skin colour, clamminess, capillary rell time, heart
rate, blood pressure, pulse pressure, conscious level and respiratory rate) intravenous access must be established and uids started.
Examinations should identify the origin of the problem but (apart
from the external haemorrhage that should be controlled using
direct pressure) usually that is only possible at hospital. Dysfunction of the central nervous system can be a life-threatening event.
In our trauma system patients with D problems must be transferred to the trauma centre, and if the Glasgow coma scale of the
patient is bellow 8, control of the airway is mandatory. When several life-threatening events were identied in the same patient, we
considered the rst one in alphabetic order to be treated, as dened
by medical educational principles.
To avoid retrospective case denitions of head and severe
injury as opposed to what the pre-hospital personnel identies
at the scene (paramedics or physicians at scene will not always
know whether or not the unconscious trauma patient has a head
trauma or degree of severity) we studied only patients with a
life-threatening event identied (regardless of subsequent injury
diagnoses).
Patients were divided into 3 groups related to the place of
treatment of life-threatening events: pre-hospital, rst hospital or trauma centre. Life-threatening events and treatments are
described in Table 1.
2.2.3. Time to trauma centre
Time to trauma centre was dened as time from the accident
occurrence to the arrival at the level I trauma centre (HGSA).
The primary outcome measure used was trauma centre mortality. The descriptive analysis was made using appropriate summary
tests. Categorical variables were described using absolute and relative (%) frequencies, continuous variables were analysed using
mean as central tendency and standard deviation as a dispersion measurement and non-continuous variables using median as
central tendency and inter-quartile range (IQR) as a dispersion
measurement. When appropriate Chi-Square testing and Fishers
exact testing were applied to test hypothesis related to categorical variables. Survival KaplanMeier curves were done for patients
in the 3 groups (pre-hospital, rst hospital and trauma centre
treatment of life-threatening events). Test of equality of survival
distributions were done using Cox regression. To assess the predictive effect of the place of treatment of life-threatening events
on mortality, several logistic regression models were constructed,
with mortality as dependent variable and population characteristics as independent variables. Models were built using different
independent variables from previous well know prognostic trauma
models.18,20,21 Statistical signicance was set at the 0.05 level.
SPSS 15.0 was used for statistical analysis.
3. Results
3.1. Population characteristics
Between 1st January 2001 and 31st December 2006, 1050
trauma patients were admitted to the emergency room of HGSA

442

Table 1
Patients variables according to mortality and place of treatment of life-threatening events.
Total (n = 727)

pa

Case fatality
Dead (n = 204)

157 (22)
570 (78)
44 (21)

99 (19)
424 (81)
42 (21)

58 (28)
146 (72)
48 (21)

Type of trauma n (%)

Blunt
Penetrating

690 (95)
37 (5)

505 (97)
18 (3)

185 (91)
19 (9)

Mechanism of injury n (%)

Road trafc collision
Fall
Assault, blunt
Other, blunt
Stabbing
Gunshot

422 (58)
217 (30)
5 (1)
49 (7)
2 (0)
32 (4)

331 (63)
141 (27)
4 (1)
32 (6)
2 (0)
13 (3)

91 (44)
76 (37)
1 (1)
17 (8)
0 (0)
19 (9)

AIS, mean (SD)

Head
Face
Thorax
Abdomen
Extremities
ISS, median (IQR)
RTS, median (IQR)
Probability of survival, mean (SD)

4.1 (0.9)
1.9 (0.7)
3.3 (1.0)
2.7 (1.1)
2.2 (0.9)
25 (1832)
5.9 (56.9)
0.69 (0.28)

Outcome n (%)
Died
Discharged

204 (28)
523 (72)

Life-threatening events
A
B
C
D

30 (4)
66 (9)
64 (9)
567 (78)

a
b
c
d

Qui-Square test.
Independent sample test.
MannWhitney.
KuskallWallis; IQR: inter-quartile range.

4.0 (0.9)
1.9 (0.7)
3.3 (1.0)
2.6 (1.1)
2.3 (0.9)
25 (1730)
5.9 (56.9)
0.75 (0.25)

25 (5)
49 (9)
47 (9)
402 (77)

4.7 (0.7)
1.8 (0.6)
3.5 (1.0)
2.8 (1.2)
2.2 (0.8)
25 (2534)
5.0 (45.9)
0.54 (0.29)

Pre-hospital (n = 49)
0.005

Trauma centre (n = 98)

11 (22)
38 (78)
38 (19)

123 (21)
457 (79)
44 (21)

23 (24)
75 (77)
48 (22)

0.001

48 (98)
1 (3)

546 (94)
34 (6)

96 (98)
2 (2)

35 (71)
8 (6)
0 (0)
5 (10)
0 (0)
1 (2)

331 (57)
177 (31)
5 (1)
36 (6)
2 (0)
29 (5)

56 (57)
32 (33)
0 (0)
8 (8)
0 (0)
2 (2)

<0.001b

<0.001b
0.324b
0.058b
0.536b
0.673b
<0.001c
<0.001c
<0.001b

5 (3)
17 (8)
17 (8)
165 (81)

First hospital (n = 580)

0.479

4.2 (0.9)
2.2 (0.8)
3.0 (1.1)
2.4 (0.5)
2.0 (0.7)
22 (1632)
5.9 (46.9)
0.71 (0.31)

4.2 (0.9)
1.8 (0.6)
3.3 (1.0)
2.5 (1.1)
2.2 (0.8)
25 (1829)
5.9 (56.9)
0.70 (0.27)

4.0 (1.1)
2.0 (0.8)
3.7 (0.9)
2.9 (1.7)
2.6 (1.0)

0.871
0.028b

0.118b
0.148b
0.005b
0.155b
<0.001b

29 (2438)
5.9 (57.1)
0.6 (0.32)

0.001d
0.238d
0.011

10 (20)
39 (80)

157 (27)
423 (73)

37 (38)
61 (62)

0.044

0 (0)
2 (6)
3 (8)
31 (86)

12 (2)
37 (7)
13 (3)
426 (88)

4 (6)
12 (16)
22 (30)
35 (48)

E. Gomes et al. / Resuscitation 81 (2010) 440445

Alive (n = 523)
Gender, n (%)
Female
Male
Age, mean (SD)

pa

Life-threatening events treatment

E. Gomes et al. / Resuscitation 81 (2010) 440445

and enrolled in the prospective registry. We included in this study

761 trauma patients transferred from another hospital. Twentyve patients who had no life-threatening events identied were
excluded from subsequent analyses. We also excluded 8 patients
with missing data related to the presence or the treatment of lifethreatening events.
Patients included had a mean age of 44 years (SD = 21). The
majority were male (78%; n = 570) and had sustained injury through
blunt trauma (95%; n = 690) with traumatic brain injury (TBI) in
87% of cases. The context of injury was road trafc collision (RTC)
in 58% of cases (n = 422). All patients had severe trauma with a
mean Abbreviated Injury Score (AIS) in the head of 4.1, a median
Injury Severity Score (ISS) of 25 (IQR 1832) and a median Revised
Trauma Score (RTS) of 5.97 (IQR 56.9). The probability of survival
(TRISS) for these patients was 69% and the real survival rate was
72% (Table 1).
3.2. Life-threatening events
In accordance with the ABCD assessment and treatment
methodology, life-threatening events were identied. Not surprisingly most of the patients in this population had D problems
(n = 567; 78%). We assumed that treatment for D problems was
maintaining an adequate airway (including tracheal intubation if
the GCS was less than 8 points) and aiming for good oxygenation
and perfusion followed with early reference to a neurosurgical centre. For example: if a patient arrived at the rst hospital with a
possible head injury identied, a GCS less than 8 identied and the
protection of the airway was performed in rst hospital prior to
transfer, then that patient was assumed to be part of the rst hospital group (life-threatening event corrected in rst hospital). If the
identication and treatment of D problem was performed in prehospital before arrival at the rst hospital then that patient was part
of the pre-hospital group (life-threatening event corrected in the
pre-hospital scene). If the same patient arrived at the trauma centre
without intubation (and the need was present before arrival) then
that patient would be part of the trauma centre group. The second
most common life-threatening event involved breathing (n = 66;
9%) and it was corrected using a chest drain in 33 cases and mechanical ventilation in 31 patients. The same methodology described
above was used to decide in which group each patient would be.
Circulation life-threatening events were identied in 64 patients
(9%) and uids (6), vasopressors (6) and surgery (52) were the treatments performed. Airway life-threatening events were identied
in 30 patients (4%) and were treated using oxygen (15) intubation
(5), a cervical collar (7) and oxygen with basic airway manoeuvres
(3). Cervical spinal cord injuries were included in the airway lifethreatening events as airway assessment and treatment includes
cervical spine control. It was assumed that all patients should have
had a cervical collar, so if they had a spinal cord injury and no collar
that was classied as a life-threatening event not corrected.
Whenever it was not possible to understand retrospectively if
the life-threatening event was present before arrival at the rst hospital or trauma centre or if it developed only during transportation
to the trauma centre, that patient was excluded because of missing
data. Table 2 lists life-threatening events and treatment and Table 1
shows mortality for different life-threatening events.

443

Table 2
Life-threatening events and treatments.
Life-threatening events
Airway (A) problem
Airway obstruction
Cervical spine injury
Treatment
Oxygen and Basic airway manoeuvres
Tracheal intubation
Cervical collar
Breathing (B) problem
Pulmonary contusion
Massive Haemothorax
Tension Pneumothorax
Flail chest
Treatment
Oxygen
Ventilation
Chest drain
Circulation (C) problem
Clinical signs of shock
Treatment
Surgery
Fluids
Fluids + Vasoconstrictors
Disability (D) problem
GCS < 9
GCS > 8
Treatment
Tracheal intubation
Referral to Neurosurgery

30 (4)
23 (77)
7 (23)
18 (60)
5 (16)
7 (24)
66 (9)
24 (37)
17 (25)
16 (25)
9 (13)
66 (100)
31 (47)
33 (49)
64 (9)
64 (100)
52 (81)
6 (8)
6 (8)
567 (78)
377 (66%)
190 (34%)
377 (66%)
567 (100)

similar except for age and ISS (patients who had life-threatening
events treated in the pre-hospital phase were younger and less
severely injured). Mortality rates differed across these 3 groups,
20% for patients with life-threatening events treated in the prehospital environment, 27% for those with events treated in rst
hospital and 38% for patients with life-threatening events solved
only in the trauma centre (p = 0.044).
Fig. 1 shows the survival curves for the 3 groups of patients.
Patients were more likely to survive when their life-threatening
events were corrected in the pre-hospital phase, corresponding to a
mortality reduction of almost 50% when compared to those patients
whose life-threatening events were corrected later in trauma
centre.

3.3. Place of treatment of life-threatening events and outcome

Most of the patients had their life-threatening events treated
in pre-hospital or rst hospital (pre-trauma centre) (n = 629; 87%).
Life-threatening events were corrected in the pre-hospital environment in 49 patients (7%), in the rst hospital in 580 (80%) and solely
in trauma centre in 98 (13%) patients. Demographic and injury
severity characteristics (Table 1) of these 3 groups of patients were

Fig. 1. Survival curves for patients with life-threatening events treated in prehospital, rst hospital or trauma centre. Length of stay is depicted in days.

444

E. Gomes et al. / Resuscitation 81 (2010) 440445

Table 3
Regression analysis for mortality.
OR

CI 95%

Gender
Female
Male

1.000
0.549

0.361.837

0.005

Age

1.015

1.0061.024

0.001

ISS

1.044

1.0251.063

<0.001

Type of trauma
Penetrating
Blunt

1.000
0.193

0.0870.429

<0.001

Place of treatment of life-threatening events

Pre-Hospital
1.000
First Hospital
1.457
Trauma Centre
3.256

0.6563.233
1.2768.309

0.355
0.014

RTS

0.617

0.5350.713

<0.001

Time to trauma centre

1.000

0.9991.001

0.568

C Statistic

0.757

Variables in the model: gender, age, ISS, RTS, type of trauma, place of treatment of
life-threatening events and time to trauma centre. Enter method was used.

Time from accident to trauma centre (including pre-hospital

and rst hospital times) was also compared. Patients with lifethreatening events solved in pre-hospital, rst hospital and trauma
centre phases had respectively a median time from accident to
trauma centre of 5 h and 31 min (IQR 3:459:24), 4 h and 35 min
(IQR 2:507:02), and 3 h and 30 min (IQR 2:516:56).
As the univariate analyses showed that patients whose lifethreatening events where treated in the pre-hospital environment
had lower mortality but at the same time were younger and less
severely injured, a multivariate logistic regression was performed
to adjust the odds of death to patient characteristics and trauma
severity as well as time from accident to trauma centre. Logistic
regression showed that increases in mortality were associated with
female gender and older age, penetrating type of trauma, higher
anatomic severity (ISS), higher physiological severity (RTS) and
having the life-threatening events corrected only at the trauma
centre (Table 3). Different models using TRISS and other severity
model variables and categories were constructed but did not show
a better performance in this population and are not shown.
4. Discussion
Our main nding was that the treatment of life-threatening
events in the pre-hospital scene (compared to the treatment of
life-threatening events only at rst hospital or the trauma centre)
decreased the mortality of severe trauma patients transferred to a
trauma centre. A reduction of almost 50% in crude mortality was
found for those patients having their life-threatening events treated
in the pre-hospital phase. This observed benet was maintained
after adjustment for confounders in logistic regression analysis. As
a result we believe that in the context of this report pre-hospital
ALS and the ability to treat life-threatening events had a positive
effect on mortality reduction. We accepted our rst hypothesis.
In the literature not many studies support a positive relationship of ALS with outcome but most are able to show an increase of
on-scene time when ALS is performed. Several important reviews
summarise well the results of relevant studies. Liberman in a
meta-analysis reviews and critiques the literature on ALS and BLS
inuence on outcome. He reviews 174 papers, recovers only 15
that permit an outcome analysis, was unable to nd any large randomised prospective study, and concludes that . . .the aggregated
data from the literature have failed to demonstrate a clear benet
from on-site ALS in trauma.10 Bunn et al. in a review supported

by the World Health Organisation identies only one small randomised study that is in favour of the ALS and concludes on the
absence of evidence of the effectiveness of ALS in the pre-hospital
setting.12 However, different denitions for what is called ALS and
BLS, the study of partial ALS attitudes or techniques instead of the
all concept analyses, small study samples, studies not controlled
nor randomised and a retrospective methodology are probably
the main reasons for the difculty in showing a better relationship of ALS with outcome, along with differences between trauma
systems.13 The future of ALS pre-hospital trauma care depends on
nding the groups of patients who benet denitely from it. We
believe that this study contributes to this debate as it shows that
trauma patients transferred to a trauma centre benet from ALS in
the pre-hospital phase. Although studies comparing ALS with BLS
have been done before as far as we know the comparison was never
performed in a well-dened cohort where interventions were definitely needed to treat life-threatening events, as was the case in
this study. Our study is in accordance with recent papers showing benet of ALS in the pre-hospital environment9,22,23 and others
that show benet from pre-hospital ALS only in specic groups of
patients namely those with TBI.2427
A recent before-after controlled clinical trial study from
Canada28 shows an increase in mortality with ALS implementation. Comparison with our report is difcult as the Canadian study
compares ALS done by trained paramedics in urban scenarios.
In our study ALS means the presence of a doctor that identies
and corrects the life-threatening events that each patient had in
either rural or urban scenarios. Moreover they study only patients
admitted directly from the scene and we are studying transferred
patients.
Another important result is that this study strongly supports
the importance of the ABCD methodologya structured approach
to assess and treat life-threatening events in trauma patients. ABCD
methodology is currently taught in trauma courses like ATLS15
and European Trauma Course.29 This nding is in accordance with
other authors like Deakin who found that preventable pre-hospital
deaths were related to failure to deliver good solutions to A, B and
C life-threatening events at the scene.30
We also report that correcting life-threatening events pretrauma centre (pre-hospital and rst hospital) increases the total
time from the accident to trauma centre. The mean times we report
were very long mainly not only because we are analysing only
transferred trauma patients but also because the trauma system we
are analysing has hospitals that are more than three driving hours
away from the trauma centre. Several studies also found an increase
in total pre-hospital time spent if ALS was dispatched to trauma
patients.3133 Moreover some studies even reported a worse outcome if the time to trauma centre was prolonged.5,6 However in
our trauma system time from accident to trauma centre although
also higher did not make a measurable difference in the mortality
of severe trauma patients making us reject our second hypothesis.
This is in accordance with other European studies from Scotland
and Switzerland showing that long pre-hospital times were not
associated with worse outcome.7,23 We believe that the potential
disadvantage of spending more time pre-trauma centre might have
been diluted by the advantages of delivering good standards of
care in the pre-hospital environment or rst hospital correcting
life-threatening events prior to transfer to the trauma centre.
4.1. Study strengths and weaknesses
Our studys strengths lie in the well described cohort of patients;
the prospective evaluation of a single trauma system with all
patients transferred and afterwards treated in the same trauma
centre; the use of the TRISS methodology for severity analysis and
the use of a regression modelling to adjust mortality for case mix.34

E. Gomes et al. / Resuscitation 81 (2010) 440445

There are however a number of potential limitations in this

study that must be taken into consideration. First, hospital mortality (more precisely trauma centre mortality), the dependent
variable used, is a crude measure of effectiveness and may be heavily inuenced by the quality of primary hospital care and possible
differences in patient handling after reaching the trauma centre.
Second we did not measure time from the traumatic event to treatment of life-threatening events and assumed that patients who had
events solved in pre-hospital care had them solved earlier than
patients who had them solved in primary or secondary hospital or
trauma centre. We do not know this with certainty because the registry only records total time from injury till arrival in the trauma
centre. However it is unlikely that patients assessed and treated
in the pre-hospital phase, by an emergency physician, had their
events corrected later than patients who were transported to rst
hospital and then transferred to trauma centre. Third we did not
collect data on urban versus rural geography in our trauma system.
A study from Liberman et al. showed no benet of ALS in urban
areas in Canada.35 Other studies suggest benet from pre-hospital
ALS in rural and remote areas.7,36,37 One possible explanation for
our results would be that patients with life-threatening events corrected pre-trauma centre were predominantly from a rural area.
Fourth we cannot make conclusions regarding the inuence of pretrauma centre correction of life-threatening events on pre-trauma
centre mortality as this is a study based only on a trauma centre
registry and not a population based study.
4.2. Conclusions and future research
Interventions able to treat life-threatening events (especially
if made early by pre-hospital ALS teams) decreased transferred
trauma patient mortality even if this prolonged time to trauma centre. Further research should indicate whether this conclusion can
be generalised to different trauma systems.
Conict of interest statement
All authors deny any conicts of interest.
References
1. MacKenzie EJ, Rivara FP, Jurkovich GJ, et al. A national evaluation of the effect of
trauma-center care on mortality. N Engl J Med 2006;354:36678.
2. Nathens AB, Jurkovich GJ, Maier RV, et al. Relationship between trauma center
volume and outcomes. JAMA 2001;285:116471.
3. Demetriades D, Martin M, Salim A, Rhee P, Brown C, Chan L. The effect of trauma
center designation and trauma volume on outcome in specic severe injuries.
Ann Surg 2005;242:5127 [discussion 5179].
4. Celso B, Tepas J, Langland-Orban B, et al. A systematic review and meta-analysis
comparing outcome of severely injured patients treated in trauma centers
following the establishment of trauma systems. J Trauma 2006;60:3718 [discussion 378].
5. Feero S, Hedges JR, Simmons E, Irwin L. Does out-of-hospital EMS time affect
trauma survival? Am J Emerg Med 1995;13:1335.
6. Baez AA, Lane PL, Sorondo B, Giraldez EM. Predictive effect of out-of-hospital
time in outcomes of severely injured young adult and elderly patients. Prehosp
Disaster Med 2006;21:42730.
7. McGufe AC, Graham CA, Beard D, et al. Scottish urban versus rural trauma
outcome study. J Trauma 2005;59:6328.

445

8. Lerner EB, Billittier AJ, Dorn JM, Wu YW. Is total out-of-hospital time a signicant
predictor of trauma patient mortality? Acad Emerg Med 2003;10:94954.
9. Osterwalder JJ. Mortality of blunt polytrauma: a comparison between emergency physicians and emergency medical techniciansprospective cohort study
at a level I hospital in eastern Switzerland. J Trauma 2003;55:35561.
10. Liberman M, Mulder D, Sampalis J. Advanced or basic life support for trauma:
meta-analysis and critical review of the literature. J Trauma 2000;49:58499.
11. Lerner EB, Moscati RM. The golden hour: scientic fact or medical urban legend? Acad Emerg Med 2001;8:75860.
12. Bunn F, Kwan I, Roberts I, Wentz R. Effectiveness of pre-hospital trauma care.
Cochrane Injuries Group 2001.
13. Roudsari BS, Nathens AB, Cameron P, et al. International comparison of prehospital trauma care systems. Injury 2007;38:9931000.
14. Lott C, Araujo R, Cassar MR, et al. The European Trauma Course (ETC) and the
team approach: past, present and future. Resuscitation 2009 [Epub ahead of
print].
15. Advanced Trauma Life Support Course for doctors, faculty manual. 7th ed. American College of Surgeons; 2004.
16. Gomes E, Araujo R, Soares-Oliveira M, Pereira N. International EMS systems:
Portugal. Resuscitation 2004;62:25760.
17. Resources for optimal care of the injured patient: an update. Task Force of
the Committee on Trauma, American College of Surgeons. Bull Am Coll Surg
1990;7:209.
18. Boyd CR, Tolson MA, Copes WS. Evaluating trauma care: the TRISS method.
Trauma Score and the Injury Severity Score. J Trauma 1987;27:3708.
19. Carley S, Driscoll P. Trauma education. Resuscitation 2001;48:4756.
20. Bouamra O, Wrotchford A, Hollis S, Vail A, Woodford M, Lecky F. A new approach
to outcome prediction in trauma: a comparison with the TRISS model. J Trauma
2006;61:70110.
21. Perel P, Arango M, Clayton T, et al. Predicting outcome after traumatic brain
injury: practical prognostic models based on large cohort of international
patients. BMJ 2008;336:4259.
22. Lossius HM, Soreide E, Hotvedt R, et al. Prehospital advanced life support provided by specially trained physicians: is there a benet in terms of life years
gained? Acta Anaesthesiol Scand 2002;46:7718.
23. Osterwalder JJ. Can the golden hour of shock safely be extended in blunt
polytrauma patients? Prospective cohort study at a level I hospital in eastern
Switzerland. Prehospital Disaster Med 2002;17:7580.
24. Rudehill A, Bellander BM, Weitzberg E, Bredbacka S, Backheden M, Gordon E.
Outcome of traumatic brain injuries in 1,508 patients: impact of prehospital
care. J Neurotrauma 2002;19:85568.
25. Garner A, Crooks J, Lee A, Bishop R. Efcacy of prehospital critical care teams for
severe blunt head injury in the Australian setting. Injury 2001;32:45560.
26. Klemen P, Grmec S. Effect of pre-hospital advanced life support with rapid
sequence intubation on outcome of severe traumatic brain injury. Acta Anaesthesiol Scand 2006;50:12504.
27. Coats TJ, Kirk CJ, Dawson M. Outcome after severe head injury treated by an
integrated trauma system. J Accid Emerg Med 1999;16:1825.
28. Stiell IG, Nesbitt LP, Pickett W, et al. The OPALS Major Trauma Study: impact of
advanced life-support on survival and morbidity. CMAJ 2008;178:114152.
29. Thies K, Gwinnutt C, Driscoll P, et al. The European Trauma Course-From concept
to course. Resuscitation 2007.
30. Deakin C, Davies G. Dening trauma patient subpopulations for eld stabilization. Eur J Emerg Med 1994;1:313.
31. Sampalis JS, Tamim H, Denis R, et al. Ineffectiveness of on-site intravenous lines:
is prehospital time the culprit? J Trauma 1997;43:60815 [discussion 615617].
32. Reines HD, Bartlett RL, Chudy NE, Kiragu KR, McKnew MA. Is advanced life
support appropriate for victims of motor vehicle accidents: the South Carolina
Highway Trauma Project. J Trauma 1988;28:56370.
33. Birk HO, Henriksen LO. Prehospital interventions: on-scene-time and
ambulance-technicians experience. Prehosp Disaster Med 2002;17:1679.
34. Perel P, Edwards P, Wentz R, Roberts I. Systematic review of prognostic models
in traumatic brain injury. BMC Med Inform Decis Mak 2006;6:38.
35. Liberman M, Mulder D, Lavoie A, Denis R, Sampalis JS. Multicenter Canadian
study of prehospital trauma care. Ann Surg 2003;237:15360.
36. Esposito TJ, Maier RV, Rivara FP, et al. The impact of variation in trauma care
times: urban versus rural. Prehosp Disaster Med 1995;10:1616 [discussion
166167].
37. Gonzalez RP, Cummings G, Mulekar M, Rodning CB. Increased mortality in
rural vehicular trauma: identifying contributing factors through data linkage.
J Trauma 2006;61:4049.