Eur J Trauma Emerg Surg
DOI 10.1007/s00068-017-0873-8
ORIGINAL ARTICLE
Trauma patients centralization for the mechanism of trauma: old
questions without answers
S. Magnone1 · A. Ghirardi2 · M. Ceresoli1 · L. Ansaloni1
Received: 11 July 2017 / Accepted: 4 November 2017
© Springer-Verlag GmbH Germany, part of Springer Nature 2017
Abstract Conclusions Significant association with major trauma
Introduction Centralization of trauma patients has become was demonstrated in the multivariate analysis of different
the standard of care. Unfortunately, overtriage can overcome mechanisms of trauma in patients triaged only for dynam-
the capability of Trauma Centres. This study aims to ana- ics. A revision of our field triage protocol with a prospective
lyse the association of different mechanisms of injury with validation is needed to improve overtriage that is above the
severe or major trauma defined as Injury Severity Score suggested limits.
(ISS) greater than 15 and an estimation of overtriage upon
our Trauma Centre. Keywords Trauma centres · Injury Severity Score ·
Methods A retrospective review of our prospective data- Triage · Multiple trauma · Mechanism of injury
base was undertaken from March 2014 to August 2016. Uni-
variate and multivariable logistic regression models were
used to estimate the association between covariates (gender, Introduction
age, and mechanisms of injury) and the risk of major trauma.
Results The trauma team (TT) treated 1575 patients: Traumatic injuries usually ask for demanding organisa-
among the 1359 (86%) were triaged only because of dynam- tion both for time-related needs and a comprehensive and
ics or mechanism of trauma. Overtriage according to an immediate multidisciplinary approach to reach acceptable
ISS < 15, was 74.6% on all trauma team activation (TTA) outcomes [1, 2]. Moreover, regionalisation of trauma care
and 83.2% among the TTA prompted by the mechanism of has proved better results regarding mortality [3]. In Lom-
injury. Patients aged 56–70 years had an 87% higher risk of bardia Region, Italy, a trauma system was established in
having a major trauma than younger patients (OR 1.87, 95% October 2012 [4], aiming at the creation of six Level I
CI 1.29–2.71) while for patients aged more than 71 years Trauma centres, connected with several other hospitals as
OR was 3.45, 95% CI 2.31–5.15. Car head-on collision (OR lower level centres. No standard monitoring or regulations
2.50, 95% CI 1.27–4.92), intentional falls (OR 5.61, 95% CI for the managing of data exist in Italy, as well as in Lom-
2.43–12.97), motorbike crash (OR 1.67, 95% CI 1.06–2.65) bardia in the north part of the State. Lombardia is a 10 mil-
and pedestrian impact (OR 2.68, 95% CI 1.51–4.74) were lion inhabitants region and has four Emergency Medical
significantly associated with a higher risk of major trauma System (EMS) control rooms. Bergamo province is in the
in a multivariate analysis. middle of the region and accounts for a 1.1 million popu-
lation served by the Papa Giovanni XXIII Hospital, the
only Level I-II Trauma Centre with neurosurgery, which
* S. Magnone is the referral for unstable patients also for the Sondrio
smagnone@asst-pg23.it
province (further 181,000 inhabitants), and it holds one of
1
General Surgery Unit, Ospedale Papa Giovani XXIII, Piazza the EMS control rooms. At the beginning of our experi-
OMS 1, 24127 Bergamo, Italy ence overtriage was the main concern of the field triage
2
FROM Research Foundation, Papa Giovanni XXIII Hospital, protocol because it could wear out the Trauma Team and
Bergamo, Italy the Emergency Department. Moreover, trauma care has
13
Vol.:(0123456789)
, S. Magnone et al.
a well-renowned impact on cost and the National Health
System [5, 6]. Because the vast majority of Trauma Team
Activation (TTA) comes from the mechanism of injury
(or dynamics), rather than measured vital signs or from
specific anatomic lesions [12], this could be considered the
first cause of overtriage [7]. On the other hand, undertriage
is far more dangerous for the safety of patients to be sent
to Trauma Centres because undertriage may result in pre-
ventable mortality or morbidity from delays in definitive
care. Overtriage is a matter of human and technological
resources, and although it has minimal adverse medical
consequences for the patient, it results in excessive costs
and burden for higher-level trauma centres in the routine
care of injured patients [8]. We published a preliminary
report on this subject [9], and this is the completion of
the analysis of all available data. Aims of this study are,
therefore: (1) to estimate the rate of overtriage and under-
triage upon our Trauma Centre and (2) to investigate the
association between different mechanisms of injury and
the risk of major trauma in a subgroup of TTA driven by Fig. 1 Field triage protocol for trauma team activation
the trauma dynamic.
Materials and methods
This is a retrospective study based on a prospective registry Statistical analysis
of patients presenting to Papa Giovanni XXIII Hospital in
Bergamo. One of the authors (SM) supervised the collection The rate of overtriage and undertriage was calculated for
of all data. The registry was established in March 2014 and TTA that was prompted only by the mechanism of injury,
includes all patients managed by the Trauma Team (TT) using the above definition and expressed as a propor-
which is prompted at a unique level of activation by the tion. This is because; an assumption was made that only
Emergency Medical System or in the Emergency Depart- patients without altered vital signs or anatomic lesions,
ment (ED) Triage Room according to a shared field tri- as in Fig. 1, could be transported to non-Level I Trauma
age protocol (Fig. 1). According to this protocol, TTA is Centre such as Papa Giovanni XXIII.
driven by the following criteria: (1) alteration of vital signs For the subsequent analyses, only TTA for the mecha-
of patients, (2) presence of clinically evident lesions or (3) nism of injury was considered, and the characteristics of
mechanism of injury. patients were reported.
No data are available for patients that did not prompt Continuous variables were expressed as median and
TTA. Anatomical injury severity was expressed by the Interquartile Range (IQR), while categorical variables
Injury Severity Score (ISS), and major trauma is commonly were expressed as proportions. Mann–Whitney U test and
identified by an ISS greater than 15. the Chi-square test (or Fisher’s exact test, when appropri-
Different definitions of overtriage were compared to eval- ate) were used to compare continuous and categorical vari-
uate the degree of eventually excessive TTA for dynamics in ables, respectively. Univariate and multivariable logistic
our Centre: (1) patients with an ISS < 15; (2) patients with regression models were used to estimate the unadjusted/
a length of stay < 48 h without any surgical intervention nor adjusted Odds Ratios (ORs) and the corresponding 95%
ICU admission, [10–12] or (3) patients discharged from confidence intervals (CIs) for the association between
the ED. Undertriage was defined as (1) the percentage of covariates (gender, age and mechanisms of injury) and
patients transferred from other hospitals or ED within 24 h the risk of major trauma.
from the event of trauma and who had an admission longer For all tested hypotheses Two-tailed p values less than
than 2 days or a surgical treatment or (2) the percentage of 0.05 were significant. Analyses were performed using
patients with ISS > 15 transferred from other hospitals or ED STATA software, release 13 (Stata Corp LP, College Sta-
within 24 h from the index event [12]. tion TX, USA).
13