ORIGINAL ARTICLE

Early Onset Pneumonia in Severe Chest Trauma: A Risk

Factor Analysis
Pierre Michelet, MD, PhD, David Couret, MD, Fabienne Bregeon, MD, PhD, Gilles Perrin, MD,
Xavier-Benoit DJourno, MD, Veronique Pequignot, MD, Veronique Vig, MD, and Jean-Pierre Auffray, MD

Background: The development of an early-onset pneumonia (EOP), occurring within the first 72 hours after admission, represents a critical event in
severe thoracic trauma population. The aim of this study was to determine
risk factors associated with the occurrence of this complication in this
specific population.
Methods: A retrospective review of a prospective implemented trauma
registry was conducted during a 4-year period in a Level I trauma center.
Over the study period, 223 severely injured patients were admitted with
severe thoracic trauma (Injury Severity Score 16 and Thorax Abbreviated
Injury Score 2). Multiple logistic regression analysis was used to determine
the independent predictors of EOP based on the clinical characteristics and
the initial management both in the field and after admission in the trauma
center.
Results: Independent predictors of EOP were the necessity of intubation and
mechanical ventilation in the field (adjusted odds ratio [OR]: 11.8; 95%
confidence interval [CI]: 4.332.7), a history of aspiration (OR: 28.6; 95%
CI: 4.0 203.5), the presence of pulmonary contusion (OR: 7.0; 95% CI:
2.0 23.9), and the occurrence of a hemothorax (OR: 3.2; 95% CI: 1.4 7.6).
Conclusion: These results emphasize the influence of prehospital and early
factors in the further occurrence of EOP, which allows the development of
early and specific clinical management to prevent it.
Key Words: Multiple trauma, Thoracic injuries, Pneumonia, Hemothorax,
Drainage, Trauma centers.
(J Trauma. 2010;68: 395 400)

The occurrence of pneumonia has been shown to precede and promote posttraumatic multisystemic organ failure
and late mortality.5,7,10 Furthermore, pneumonia appears as
the most common infectious complication in multiple injuries
patients with a incidence reaching 30% in severe population.11 Although the influence of ventilator-associated pneumonia (late pneumonia) in the clinical course has been
recently documented,12 the occurrence of early-onset pneumonia (EOP), occurring within the first 72 hours after admission, remains poorly studied. Discriminating EOP from the
whole of pneumonia or later pneumonia rely on the fact that
identification of EOP predictors could represent the first step
of understanding further preventive management. Indeed, the
interest in EOP is supported by the precocity of lung dysfunction after the trauma,5 the potential influence of the initial
management in the field, during the transfer and in the trauma
center on this early complication. Despite the interest of this
issue, there are only few available data in regard to factors
implicated in the development of EOP specifically in the most
severe patient population.
Therefore, the aim of this study was to perform a
multivariate analysis on risk factors associated with the occurrence of EOP in a severe chest trauma population.

PATIENTS AND METHODS

evere chest trauma remains a leading cause of trauma

death after head injury.1 In this specific population, the
impairment of pulmonary function is frequent and multifactorial involving lung contusion, pleural effusions, ventilation
to perfusion abnormalities2, and inflammatory injury.3 The
implication of chest trauma in mortality4,5 is related to the
persistent respiratory insufficiency, the development of septic
complications such as pneumonia,6 and multisystemic organ
failure.79
Submitted for publication September 5, 2008.
Accepted for publication March 18, 2009.
Copyright 2010 by Lippincott Williams & Wilkins
From the Reanimation des Urgences (P.M., D.C., G.P., V.P., V.V., J.P.A.), and
Service de Chirurgie Thoracique (X.B.D.J.), Hopital Sainte Marguerite, Universite de la Mediterranee; and UMR MD2 (P.M., F.B., J.P.A.), Institut Jean
Roche, Universite de la Mediterranee, Marseille, France.
Address for reprints: Pierre Michelet, MD, PhD, Reanimation des Urgences, Pole
Reanimation Urgence SAMU Hyperbarie, Hopital Sainte-Marguerite, 270
Boulevard Sainte Marguerite, 13274 Marseille Cedex 9, France; email:
pierre.michelet@ap-hm.fr.
DOI: 10.1097/TA.0b013e3181a601cb

Study Population
A retrospective review of the prospective implemented
trauma registry and medical records was performed for all
patients admitted to a Level I trauma center for thoracic
trauma during a 4-year period ending on 2007. The institutional review board approved the study that was performed in
accordance with the 1964 declaration of Helsinki. Informed
consent was obtained from the next of kin or the patients or
both. Inclusion criteria included the following: severe traumatized patients including thoracic trauma (Injury Severity
Score [ISS] 17 and higher and Thorax Abbreviated Injury
Score [AIS] 3),13 age 18 years. Exclusion criteria included burn injury, patients presenting a cardiac arrest on the
field, and death within 6 hours of admission.
Data recollection included the following: demographics, vital signs both in the field and on arrival in the trauma
center, mechanism of injury, duration of transfer to the Level
I center, occurrence of EOP (occurring in the first 72 hours),
intensive care unit (ICU) length of stay, and 30-day mortality.
The Glasgow Coma Scale (GCS), the revised trauma score,14

The Journal of TRAUMA Injury, Infection, and Critical Care Volume 68, Number 2, February 2010

395

Michelet et al.

The Journal of TRAUMA Injury, Infection, and Critical Care Volume 68, Number 2, February 2010

the AIS for head and thoracic trauma, and the ISS13 were
determined for all patients. Most of the relevant factors for
pneumonia have been selected and recorded including an
aspiration history, an antibiotic therapy, blood loss, rib fractures, hemothorax, pneumothorax, and need for urgent tracheal intubation.
The presence of a hemothorax was recorded as such if
the pleural effusion exceeded 300 mL at the first ultrasonography measurement.15 Pulmonary contusion was recorded as
such if the extent of contusion exceeded 20% of the pulmonary parenchyma on the chest computed tomographic scan.16

Airway Management
All the intubations were justified by the prehospital
medical teams including an emergency physician. To clarify
the respective influence of neurologic and respiratory impairment in the intubation indications, we have requested the
team in charge of the patient to specify the cause for intubation in the field as follows:
A neurologic cause was reported when the neurologic status
imposed the intubation: GCS score 8 or a rapid impairment in neurologic status.
A respiratory cause was reported when the respiratory function was impaired with the presence of a oxygen saturation
90% despite high oxygen delivery associated with a
respiratory rate 30/min or 10/min.
As a neurologic and a respiratory impairment were
often associated, the main cause for intubation was determined by the physician in charge of the patient. Furthermore,
noninvasive ventilation could be indicated in case of isolated
thoracic trauma without clinical signs of neurologic or hemodynamic failure at the discretion of the physician in charge of
the patient.
The use of a Levin tube was not included in the management protocol neither in the prehospital nor in the ICU.
An aspiration was defined by the macroscopic clearness
of gastric content in the upper airway tract during the initial
airway management. To prevent false-positive cases, each
patient with a doubtful presence of aspiration was finally not
recorded as such. Finally, the sole presence of an aspiration
was not retained as an indication for intubation.

Early-Onset Pneumonia
EOP was defined using the following criteria, with at
least two of the following findings being required for diagnosis: (1) temperature 38C or 36C, white blood cells
count of 12 109 or 4 109, and the presence of
purulent tracheobronchial secretions; (2) new or changing
infiltrate on chest radiograph within the first 72 hours after
admission; (3) confirmation by fiberoptic bronchoscopy with
bronchoalveolar lavage (BAL) and quantitative culture (105
colony forming units/mL)17 for patients submitted to mechanical ventilation.

Prophylactic Antibiotics
During the study period, data indicating a potential
interest in prophylactic antibiotics use were not available and
therefore no systematic antibiotherapy was done.18 Because
396

extended pulmonary contusion may cause elevated body

temperature and elevated leukocytosis, no antimicrobial therapy related to a pulmonary cause was prescribed before BAL
was performed. If other injuries justified an early antimicrobial therapy, this data were collected as such. Only the first
episode of pneumonia in each patient was considered.

Statistical Analysis
Statistical analyses were conducted using statistical
software (SPSS 15.0.1; Chicago, IL). Statistical significance
was established at p 0.05. For continuous variables, comparisons were performed using the Students t test or the
Wilcoxons test when the distribution was nongaussian. The
2 test was used for comparison of discrete variables (or
Fisher exact test, as appropriate). Identification of the study
variables (continuous and categorical variables) contributing
to the development of EOP (dependent variable) was performed using multivariate analysis. Multivariate analyses
were performed using a logistic regression model to determine the odds ratio along with 95% confidence interval.
Variables significantly associated in univariate analysis with
a p value 0.20 were entered into the model and retained
through forward selection (p 0.05). Because some variables might be surrogate markers of others, a careful appraisal of the variables selected was performed to prevent
colinearity. Calibration of the logistic model was assessed
using the Hosmer-Lemeshow goodness-of-fit test to evaluate
the discrepancy between observed and expected values. In
addition, receiver operating characteristic curve was used to
predict the occurrence of EOP. The area under the receiver
operating characteristic curve represents the probability that a
randomly chosen patient with EOP is correctly ranked for a
given risk factor with greater suspicion than a randomly
chosen patient without EOP.

RESULTS
Study Population
During the study period, 223 severely injured patients
were admitted to the trauma center with severe thoracic
trauma. The patient population consisted of 76% men and
24% women with an average age of 37 17 years and an
average body mass index of 24 3 kg/m2. Blunt thoracic
injury was the predominant type with 86% of the study
population. The remaining 32 patients sustained penetrating
wounds consisting of 17 stab wounds (53%) and 15 gunshot
wounds (47%). The mean ISS for the study population was
27 19 with a mean revised trauma score of 6.5 1.9 and
a GCS score of 11 4 at the scene. Main characteristics of
both populations (with or without EOP) are reported in the
Table 1. Among all patients, 64 have received antimicrobial
therapy before performing BAL because of extrapulmonary
infection.

Early-Onset Pneumonia
Sixty-five patients (29%) developed early pneumonia
in the first 72 hours after admission. Patients presenting
pneumonia were characterized by the seriousness of thoracic
trauma with frequent rib fractures, pulmonary contusions,
2010 Lippincott Williams & Wilkins

The Journal of TRAUMA Injury, Infection, and Critical Care Volume 68, Number 2, February 2010 Early Onset Pneumonia in Severe Chest Trauma

TABLE 1. Characteristics of Patients With or Without Early

Onset Pneumonia
Characteristics

EOP
(n 65)

No EOP
(n 158)

General characteristics
years
38 16
37 17
Body mass index, kg/m2
24.5 2.3 23.7 4.6
Sex, male gender, n (%)
53 (81)
118 (75)
Cardiovascular or respiratory
16 (25)
44 (28)
comorbidities
Associated trauma, n (%)
56 (86)
131 (83)
Head trauma, n (%)
44 (68)
97 (61)
Vertebral trauma, n (%)
24 (37)
32 (20)
Transfer duration, h*
3.6 2.4
3.9 2.7
Scores
Glasgow Coma Scale at the
9.8 4,8 11.9 4.3
scene
Glasgow Coma Scale on
7.1 5.7
9.8 5.9
admission
AIS head trauma
3.2 1.4
2.7 1.3
AIS thoracic trauma
4.0 1.1
2.9 1.2
Lung injury score on admission
1.4 0.3
1.3 0.3
ISS
30 20
26 20
RTS
62
72
TRISS, %
31 37
19 31
Initial evolution
Intubation in the field
44
74
For neurological distress, n (%)
21 (47)
43 (58)
For respiratory distress, n (%)
23 (53)
31 (42)
Initial respiratory distress, n (%)
47 (72)
61 (39)
Aspiration, n (%)
12 (18)
5 (3)
Initial hemodynamic distress, n (%) 19 (29)
35 (22)
Hemothorax, n (%)
49 (75)
75 (47)
Pneumothorax, n (%)
46 (71)
93 (59)
Associated hemopneumothorax
34 (52)
56 (35)
Pulmonary contusion
59 (91)
109 (69)
Rib fractures
45 (69)
81 (51)
Rib fractures 5
23 (35)
43 (27)
Transfusion of red blood cells,
26 (40)
68 (43)
n (%)
Mechanical ventilation, n (%)
49 (75)
78 (49)
265 156 363 211
PaO 2/FIO2 ratio on admission
Antimicrobial therapy within
21 (32)
43 (28)
the first 12 h
Outcome
Mechanical ventilation
24.5 16.9 7.7 9.7
duration, d
ICU length of stay, d
30.2 25.6 8.3 11.9
Global mortality
11 (17%)
33 (20%)
Late mortality (120 h)
6
6

TABLE 2. Microbiologic Isolates (n 48) in the 65

Early-Onset Pneumonia
Pathogen

p
0.68
0.40
0.35
0.74
0.69
0.46
0.02
0.44
0.002
0.002
0.01
0.001
0.02
0.01
0.009
0.18

0.66
0.62
0.001
0.001
0.34
0.001
0.13
0.03
0.001
0.02
0.29
0.78
0.001
0.001
0.61

0.001
0.001
0.04

n (%) is patients number (percentage of the referent population, EOP, or no EOP).

Associated trauma is the existence of other trauma than thoracic.
*Duration between the first call and trauma centre admission.
Necessity of red blood cells transfusion during the first 6 h after admission.
RTS, revised trauma score; TRISS, Prognostic Severity Index.

Staphylococcus aureus
Streptococcus pneumoniae
Escherichia coli
Haemophilus influenzae
Pseudomonas aeruginosa
Enterobacter spp.
Acinetobacter spp.

Early-Onset Pneumonia (n)

23
8
8
6
1
1
1

(Table 1). The occurrence of an early pneumonia was associated

with a longer ICU stay and longer mechanical ventilation duration. Although the overall mortality was not different between
patients who presented and not presented pneumonia, the late
deaths (120 hours after admission) were more frequent in
the pneumonia group. Table 1 is descriptive and in this way,
the EOP was presented as the independent variable. Conversely, in the multivariate model, the EOP was considered as
the dependent variable.
Although the late deaths were mainly related to head
injuries and subsequent consequences in the no pneumonia
group (four of the six deaths), the causes in the pneumonia
group were related to postinjury multiple organ failure in
three cases, head trauma in two, and pulmonary embolism in
one. Pathogens isolated in EOP are displayed in Table 2.
From all the BAL performed, 47 patients presented one
identification, 17 patients presented no identification, and
only 1 patient has presented two identifications (Staphylococcus aureus and Haemophilus influenzae).

Risk Factors
After univariate analysis, some variables appeared to be
surrogate markers of others. Therefore, after an appraisal of
potential colinearity, several factors were not entered in the
final model. For example, the GCS in the field was closely
correlated with the necessity for tracheal intubation and
mechanical ventilation (GCS score of 7.9 4.3 for patients
intubated vs. GCS score of 14.7 0.8 for patients not
intubated). Multiple logistic regression analysis revealed that
four variables were independently associated with the occurrence of EOP after adjustment for the sex gender, age, and
ISS. These variables were the necessity of intubation with
mechanical ventilation in the field, the presence of a pulmonary contusion, a history of aspiration during the field management, and the presence of a hemothorax (Table 3). The
results of the Hosmer-Lemeshow goodness-of-fit test were a
2 of 4.095 with 8 ddl, p 0.848. This model has shown a
discriminant ability tested by the area under the receiver
operating characteristic curve. The value of the area under the
receiver operating characteristic curve for the model was
0.896, 95% confidence interval: 0.786 0.905.

Patients Chest Drainage

hemothorax, and related respiratory insufficiency (Table 1).
Those patients also presented frequent neurologic trauma
with the necessity of intubation and mechanical ventilation
2010 Lippincott Williams & Wilkins

After the chest trauma, 44 patients required chest drainage in the field and 101 patients benefited from the drainage
after their admission in the Level I center. The average of
397

The Journal of TRAUMA Injury, Infection, and Critical Care Volume 68, Number 2, February 2010

Michelet et al.

TABLE 3.

Results of Multivariate Analysis

Variable
Intubation in the field*
Aspiration
Pulmonary contusion
Hemothorax

Odds Ratio (95% CI)

11.8 (4.32.7)
28.6 (4.0203.5)
7.0 (2.03.9)
3.2 (1.40.6)

0.001
0.001
0.002
0.008

*Necessity for intubation and mechanical ventilation in the field.

History of aspiration.

hemothorax in the first 4 hours after drainage for the whole

population was 1223 932 mL.
In the 44 patients who benefited from chest drainage in
the field, an isolated pneumothorax was found in 11 patients
(25%), an isolated hemothorax in 2 patients (4.5%), and an
association of both in 31 patients (70.5%) with an average
hemothorax of 1017 657 mL (from the drainage to 4 hours
after admission). The success rate for urgent chest tube
insertion, defined by the correct location at the computed
tomographic scan and clinical effectiveness, was 84%.
In the 101 patients who required chest drainage after
admission, an isolated pneumothorax was found in 23 patients (23%), an isolated hemothorax in 25 patients (25%),
and an association of both in 53 patients (52%). From those
101 patients, 54 required urgent thoracic drainage on arrival,
indicating the appearance of drainage during the transfer to
the trauma center.

Tracheal Intubation
The main reason for intubation in the field for the whole
population was neurologic (64 patients) followed by respiratory reason (54 patients). Aspiration was suspected (initial
clinical suspicion during the intubation) in 17 patients intubated for neurologic reason. Considering separately patients
with or without EOP, patients from the pneumonia group
were mainly intubated for respiratory causes compared with
neurologic ones (23 vs. 21 patients), whereas patients from
the no pneumonia group were characterized by a lesser
respiratory cause rate (31 patients vs. 43 patients). No patient
required noninvasive ventilation during the transfer. Nine
patients required tracheal intubation after admission in the
trauma center, seven for respiratory aggravation, and two for
neurologic aggravation.

DISCUSSION
This study was performed to determine risk factors for
EOP in a severe thoracic trauma population. The main result
was that development of EOP was associated not only with
factors related to the seriousness of thoracic injury (pulmonary contusions and hemothorax) but also with factors reflecting the management in the field.
In the severe thoracic trauma population, the impairment of lung function is implicated both in the early mortality
with related hypoxemia and participation to blood loss by
hemothorax7 and in the late mortality (i.e., after the first 120
hours) related to the development of septic complications,
respiratory insufficiency, and multisystemic organ failure.79
The choice of EOP (occurring within the first 72 hours after
398

admission) as our main depending factor was based first on

the precocity of lung dysfunction after the trauma and second
on the lack of precise data concerning the most concerned
population: severe thoracic trauma patients.

Implication of EOP on Further Clinical Course

The occurrence of pneumonia could represent the first
step in the development of multiple organ dysfunction, preceding and promoting further worse evolution.5,19 In this
sense, Ciesla et al.5 have reported the development of respiratory dysfunction with an average of 1.6 days postinjury
compared with a later dysfunction of other organs. A precocious lung inflammatory injury has also been reported after
blunt thoracic trauma,3 leading to an additional hit as the
contusion or aspiration. Finally, it has been demonstrated that
pneumonia is an independent predictor of late death in thoracic trauma patients.8 The impact of ventilator-associated
pneumonia in a severe head injury population has been
analyzed recently with an association with greater duration of
mechanical ventilation and ICU stay.12 According to these
findings, our results demonstrated that the occurrence of EOP
was associated with a prolongation in mechanical ventilation
duration, ICU length of stay, and late mortality. The implication of pneumonia in multisystemic organ dysfunction can
be explained by a persistent release of inflammatory mediators from the lung, which impair the recovery of pulmonary
function and represent a second systemic insult after the
trauma.5,20 Moreover, the prolonged period of mechanical
ventilation induced both by the pneumonia and associated
injuries like head trauma represents itself another factor
previously reported.21 We report here a pneumonia incidence
of 29%, which is in accordance with previous results.11

Pneumonia in Severe Thoracic Trauma Patients

Although several studies have tackled the issue of
ventilator-associated pneumonia in severe head injury population12 or larger but inhomogeneous trauma patients,8,11 a
specific analysis of EOP have not been performed since the
study of Antonelli et al.22 Although interesting, the study of
Antonelli et al.22 included all trauma patients without a
sufficient number of patients to perform a specific analyze on
such trauma population. Moreover, apart from the study of
Eckert et al.,11 most previous studies have not included clear
data concerning the management and the trauma history from
the field even though Croce et al.6 have highlighted the
necessity of an early identification of high-risk group. Finally,
most of the previous studies did not concern thoracic trauma
population.

Determination of Risk Factors

Knowing the clinical relevance of pneumonia (without
regard to the time of development), several investigations
have been attempted to identify risk factors so as to define
risk population. These factors included notably age, severe
head injury or GCS score, hypotension on admission, thorax
AIS, increased ISS, or need for urgent intubation.6,11,22,23
Results of our analysis were partially in agreement with
previous results. Indeed, after performing multivariate analysis, four variables were identified as independent risk factors
2010 Lippincott Williams & Wilkins

The Journal of TRAUMA Injury, Infection, and Critical Care Volume 68, Number 2, February 2010 Early Onset Pneumonia in Severe Chest Trauma

including specific factors related to the management in the field

with the necessity of intubation (i.e., with mechanical ventilation), a history of aspiration in the field. The implication of the
necessity for mechanical ventilation has been already reported
by Eckert et al.11 Conversely, the ISS did not reach sufficient
significativity to be retained in our final model. This absence of
discriminant ability has also been pointed out by Miller et al.16
in a thoracic trauma population. Although Pape et al.24 have
demonstrated that the association of extensive contusions and
hemopneumothorax was associated with chest-related death, the
influence of pulmonary contusions and hemothorax on the development of infectious complications remains a matter of debate.
Considering EOP, Antonelli et al.22 have reported pulmonary
contusions as significant in the univariate but not in the multivariate analysis.22 The presence of contusions in a severe head
trauma population seemed to increase morbidity but neither the
occurrence of pneumonia nor the mortality.25 Conversely, after
a careful determination of contusion volume, Miller et al.16 have
reported an association between the extent of lung contusion and
the development of acute respiratory distress syndrome. The
existence of contusions as a risk factor for development of EOP
in our study confirms the Millers results. Another novelty of our
analysis relied on the influence of hemothorax on complication
like EOP. Obviously, the presence of a hemothorax could be
considered as a cause of blood losses8 and ventilation impairment by direct compression resulting in a factor of global tissular
hypoxia. Nevertheless, this result requests more thought on
future clinical implications.

population was a multiple trauma one implied the presence of

others injuries. This artificial restriction was only based on
the fact that our population was mainly and firstly admitted
for a severe chest trauma. As our population included specifically severe chest trauma patients, the influence of specific
injuries (i.e., hemothorax or contusion) on the occurrence of
EOP could not be generalized to the overall trauma population. In nonintubate patients, contusion, aspiration, and pneumonia can lead to similar pictures. To differentiate these
complications, a restrictive definition was applied for each
with the risk of underestimation. This study was conducted in
France with an initial management performed by prehospital
team including an emergency physician, which differs from
others country organization but allowed the collection of
reliable clinical criteria notably in regard to initial airway
management. The thoracic trauma is often associated with
other injuries and especially brain injuries, which make
complex a clear differentiation between respiratory and neurologic factors. As this aspect reflect the difficulty to determine the mechanisms of EOP occurrence in a multiple trauma
population with several confounding factors, our results provide new insights in the understanding notably in regard to
the prehospital phase. Although the potential implication of EOP
in early mortality cannot be assessed from this study related to
the number of patients included, our preliminary results call for
larger multicenter studies to assess this influence.

Clinical Implications

In regard to our population, the necessity for intubation

in the field, the presence of an aspiration, a hemothorax, and
a pulmonary contusion appear associated with the development of an EOP. This statement could improve the management of the most severe thoracic trauma patients by the
development of specific protocol.

The results of our study emphasize the influence of early

factors on the susceptibility to EOP. Our study confirms that
pulmonary contusions are clearly implicated in the development
of infectious complications with the necessity for precocious
assessment and therapeutic management as the use of noninvasive positive pressure ventilation.26 The present results call
equally for future research on the influence of hemothorax and
its management in the development of local inflammatory process. Indeed, appropriate management of collected air or fluid or
both in the pleural space is based on a complete drainage
allowing full expansion and occupation by the lung, thus protecting the lung and pleural space from subsequent complications. In this way, Aguilar et al.,27 analyzing factors associated
with the development of post traumatic empyema, have found
retained hemothorax and pulmonary contusion as the main risk
factors. Although some authors have supported that chest drainage in the prehospital phases reduces mortality and represented
a safe and effective tool with low associated morbidity,28 30
others emphasized the risk of intrapleural infections or other
related complications.31,32 Therefore, despite controversies
about the best early strategy for hemothorax treatment,30 the
reflection is mandatory. Finally, our results could participate in
the determination of high-risk population benefiting from specific therapies such as antibiotherapy, ventilatory strategy, or
immunomodulatory therapy.

Study Limitation
Although our title mentioned the analysis of EOP
occurrence in severe chest trauma patients, the fact that our
2010 Lippincott Williams & Wilkins

CONCLUSIONS

REFERENCES
1. LoCicero J III, Mattox KL. Epidemiology of chest trauma. Surg Clin
North Am. 1989;69:1519.
2. Batchinsky AI, Weiss WB, Jordan BS, Dick EJ Jr, Cancelada DA,
Cancio LC. Ventilation-perfusion relationships following experimental
pulmonary contusion. J Appl Physiol. 2007;103:895902.
3. Raghavendran K, Davidson BA, Woytash JA, et al. The evolution of
isolated bilateral lung contusion from blunt chest trauma in rats: cellular
and cytokine responses. Shock. 2005;24:132138.
4. Rodriguez JL, Gibbons KJ, Bitzer LG, Dechert RE, Steinberg SM, Flint
LM. Pneumonia: incidence, risk factors and outcome in injured patients.
J Trauma. 1991;31:907914.
5. Ciesla DJ, Moore EE, Johnson JL, Burch JM, Cothren CC, Sauaia A.
The role of the lung in postinjury multiple organ failure. Surgery.
2005;138:749 758.
6. Croce MA, Fabian TC, Waddle-Smith L, Maxwell RA. Identification of
early predictors for post-traumatic pneumonia. Am Surg. 2001;67:105110.
7. Sauaia A, Moore FA, Moore EE, et al. Epidemiology of trauma deaths:
a reassessment. J Trauma. 1995;38:185193.
8. Bergeron E, Lavoie A, Clas D, et al. Elderly trauma patients with rib
fractures are at greater risk of death and pneumonia. J Trauma. 2003;
54:478 485.
9. Kulshrestha P, Munshi I, Wait R. Profile of chest trauma in a level 1
trauma center. J Trauma. 2004;57:576 581.
10. Tejada Artigas A, Bello Dronda S, Chacon Valles E, et al. Risk factors
for nosocomial pneumonia in critically ill trauma patients. Crit Care
Med. 2001;29:304 309.

399

Michelet et al.

The Journal of TRAUMA Injury, Infection, and Critical Care Volume 68, Number 2, February 2010

11. Eckert MJ, Davis KA, Reed RL II, et al. Urgent airways after trauma:
who gets pneumonia? J Trauma. 2004;57:750 755.
12. Rincon-Ferrari MD, Flores-Cordero JM, Leal-Noval SR, et al. Impact of
ventilator-associated pneumonia in patients with severe head injury.
J Trauma. 2004;57:1234 1240.
13. Baker PS, ONeill B, Haddon W Jr, Long WB. The injury severity score:
a method for describing with multiple injuries and evaluating emergency
care. J Trauma. 1974;14:187196.
14. Champion HR, Sacco WJ, Copes WS, Gann DS, Gennarelli TA,
Flanagan ME. A revision of trauma score. J Trauma. 1989;29:623
629.
15. Roch A, Bojan M, Michelet P, et al. Usefulness of ultrasonography in
predicting pleurla effusions 500 ml in patients receving mechanical
ventilation. Chest. 2005;127:224 232.
16. Miller PR, Croce MA, Bee TK, et al. ARDS after pulmonary contusion:
accurate measurement of contusion volume identifies high risk patients.
J Trauma. 2001;51:223228.
17. Miller PR, Meredith JW, Chang MC. Optimal threshold for diagnosis of
ventilator-associated pneumonia using bronchoalveolar lavage. J Trauma.
2003;55:263267; discussion 267268.
18. Sanabria A, Valdivieso E, Gomez G, Echeverry G. Prophylactic antibiotics in chest trauma: a meta-analysis of high-quality studies. World
J Surg. 2006;30:18431847.
19. Regel G, Grotz M, Weltner T, Sturm JA, Tscherne H. Pattern of organ
failure following severe trauma. World J Surg. 1996;20:422 429.
20. Perl M, Gebhard F, Braumuller S, et al. The pulmonary and hepatic
immune microenvironment and its contribution to the early systemis
inflammation following blunt chest trauma. Crit Care Med. 2006;34:
11521159.
21. Slutsky AS, Tremblay LN. Multiple system organ failure. Is mechanical ventilation a contributing factor? Am J Respir Crit Care Med.
1998;157:17211725.

400

22. Antonelli M, Moro ML, Capelli O, et al. Risk factors for early onset
pneumonia in trauma patients. Chest. 1994;105:224 228.
23. Rello J, Ollendorf DA, Oster G, et al.; VAP Outcomes Scientific
Advisory Group. Epidemiology and outcomes of ventilator-associated
pneumonia in a large U.S. database. Chest. 2002;122:21152121.
24. Pape HC, Remmers D, Rice J, Ebisch M, Krettek C, Tscherne H.
Appraisal of early evaluation of blunt chest trauma: development of a
standardized scoring system for initial clinical decision making.
J Trauma. 2000;49:496 504.
25. Leone M, Albanese J, Rousseau S, et al. Pulmonary contusion in severe
head trauma patients. Chest. 2003;124:22612266.
26. Antonelli M, Conti G, Moro ML, et al. Predictors of failure of
noninvasive positive pressure ventilation in patients with acute hypoxemic respiratory failure: a multi-center study. Intensive Care
Med. 2001;27:1718 1728.
27. Aguilar MM, Battistella FD, Owings JT, Su T. Posttraumatic empyema.
Risk factor analysis. Arch Surg. 1997;132:647 651; discussion 650 651
28. Coats T, Wilson A, Xeropotamous N. Pre-hospital management of
patients with severe thoracic injury. Injury. 1995;26:581585.
29. Schmidt U, Stalp M, Gerich T, Blauth M, Maull KI, Tscherne H. Chest
tube decompression of blunt chest injuries by physicians in the field:
effectiveness and complications. J Trauma. 1998;44:98 100.
30. Spanjesberg WR, Ringburg AN, Bergs EA, Krijen P, Schipper IB.
Prehospital chest tube thoracostomy: effective treatment or additional
trauma? J Trauma. 2005;59:96 101.
31. Eddy AC, Luna GK, Copass M. Empyema thoracis in patients undergoing emergent closed tube thoracostomy for thoracic trauma. Am J
Surg. 1989;157:494 497.
32. Baldt MM, Bankier AA, Germann PS, Poschl GP, Skrbensky GT,
Herold CJ. Complications after emergency tube thoracostomy: assessment with CT. Radiology. 1995;195:539 543.

2010 Lippincott Williams & Wilkins