Damage Control for Thoracic Trauma

MICHAEL J. MACKOWSKI, M.D., REBECCA E. BARNETT, M.B.B.S., B.SC.,

BRIAN G. HARBRECHT, M.D., KEITH R. MILLER, M.D., GLEN A. FRANKLIN, M.D.,
JASON W. SMITH, M.D., J. D. RICHARDSON, M.D., MATTHEW V. BENNS, M.D.

From the Department of Surgery University of Louisville School of Medicine, Louisville, Kentucky
Damage control surgery involves an abbreviated operation followed by resuscitation with plan
ned re-exploration. Damage control techniques can be used in thoracic trauma but has been in
frequently reported. Our goal is to describe our experience with the use of damage control
techniques in treating thoracic trauma. A retrospective analysis of all patients undergoing damage
control thoracic surgery related to trauma from January 1,2010, to January 1, 2013, at University of
Louisville Hospital, a Level I trauma center. Variables studied included injury characteristics,
Injury Severity Score, surgery performed, duration of packing, length of stay (LOS), ventilator
days, transfusion requirements, complications, and mortality. Twenty-five patients underwent
damage control surgery in the chest with packing, temporary closure, and planned re-exploration
after stabilization. Seventeen patients underwent anterolateral thoracotomy, and eight patients
underwent sternotomy. The mean LOS and duration of temporary packing was 20.6 and 1.4 days in
the thoracotomy group, respectively, and 19.5 and 1 day in the sternotomy group, respectively. The
overall mortality rate was 40 per cent, 35 per cent in the thoracotomy group and 50 per cent in the
sternotomy group. Like in severe abdominal trauma, damage control techniques can be used in the
management of severe thoracic injuries with acceptable results.

atients sustaining thoracic trauma represent

a special injured population and providing care
for these patients can be very challenging in the face
of coagulopathy and hemodynamic instability. Dam
age control surgery involves an abbreviated operation
with planned re-exploration and has become a main
stay in the treatment of severe abdominal trauma with
coagulopathy.1 The use of damage control techniques
has been demonstrated to improve survival rates for the
severely injured trauma patient. A common element to
the application of damage control surgery is the use of
surgical packing to control nonsurgical hemorrhage.
The timely application of these techniques in treating
the critically injured patient has shown to be related to
patient survival.2 Patient characteristics that would
predict the need to use damage control techniques in
clude: hypothermia (34C or less), Acidosis (pH 7.2 or
less), serum bicarbonate 15mEq/L or less, transfusion
4000 mL or greater blood or 5000 mL or greater blood
and blood products, intraoperative volume replace
ment 12,000 mL or greater, and clinical evidence of

Presented at the Annual Scientific Meeting and Postgraduate

Course Program, Southeastern Surgical Congress, Savannah, GA,
February 22-25, 2014.
Address correspondence and reprint requests to Michael
J. Mackowski, M.D., Assistant Department of Surgery, University
of Louisville, 550 South Jackson Street, Louisville, KY 40202.
E-mail: mjmack02@louisville.edu.

coagulopathy.3 Similarly, a majority of the factors that

Asensio and colleagues4 have described as indications
for damage control surgery are also associated with
increased mortality in the trauma patient population.
Improvements in postoperative resuscitation with goaldirected therapy toward correction of coagulopathy
and restoration of normal physiology have contributed
to the improved survival of the patient requiring
damage control.5
Damage control principles have been applied to
other surgical subspecialties and are also used when
caring for patients with intra-abdominal hypertension.
Bleeding after cardiac surgery can be a problem for
even the most skilled surgeon. Packing the chest with
gauze or laparotomy sponges followed by temporary
closure and resuscitation with reversal of coagulopathy
is acceptable in the setting of elective cardiac pro
cedures. Surgical packing of the mediastinum has been
described by Bouboulis6 in the setting of coagulopathic hemorrhage after elective cardiac surgery with
acceptable results. Application of this technique for
hemorrhage control lends itself well in the setting of
surgery for thoracic trauma. With the relative paucity
of literature available regarding the application of
damage control for thoracic trauma, the goal of this
retrospective case series is to describe our experience
with the use of damage control techniques for patients
requiring thoracic surgery for trauma.

910

No. 9

DAMAGE CONTROL FOR THORACIC TRAUMA

Methods
We reviewed all patients entered into our trauma
database at University of Louisville, a Level I trauma
center in Louisville, Kentucky. We identified patients
who underwent either a thoracotomy or sternotomy for
thoracic trauma and were treated with damage control
principles over a 2-year period (January 2010 to Jan
uary 2013). Charts were retrospectively reviewed for
injury characteristics, Injury Severity Score (ISS),
surgery performed, duration of packing, length of stay
(LOS), ventilator days, transfusion requirements, and
mortality.
Techniques for performing damage control involved
placement of surgical packs to control coagulopathic
hemorrhage. For patients requiring thoracotomy, packs
were placed between the lung parenchyma and parietal
pleura to control hemorrhage from both the paren
chyma itself and from the chest wall. Similarly, for
patients requiring sternotomy, mediastinal packs were
placed until hemorrhage was controlled. No packs
were placed in direct contact with the heart or major
vasculature. A temporary wound closure was per
formed in a manner similar to that of temporary ab
dominal closure with the placement of sterile surgical
towels and chest tubes, used to create negative pres
sure, followed by coverage with Ioban (3M) dressing.
Negative pressure was then applied to the temporary
closure with continuous wall suction while patients
were undergoing resuscitation in the intensive care
unit.

Results
Twenty-five patients underwent damage control
surgery in the chest with packing, temporary closure
with a vacuum dressing (Table 1). Planned re-exploration occurred after correction of coagulopathy and
hemodynamic stabilization. Injuries sustained in
cluded eight blunt injuries and 17 penetrating injuries.
Eighteen patients underwent anterolateral thoracoto
mies and the remainder sternotomies. The mean LOS
in the thoracotomy group was 20.6 days (range, 1 to 50
days) with a mean duration of temporary packing of
1.4 days (range, 1 to 3 days). There was a consistent
duration of temporary packing of 1 day for the entire
sternotomy group with a mean LOS of 19.5 days
(range, 2 to 49 days) (Figs. 1 and 2). The patients in the
thoracotomy required a mean of 13 ventilator days
with a mean of 14.3 in the sternotomy group (Fig. 1).
The mean transfusion requirement was 109 total
products for the thoracotomy group with a mean ISS of
34.2. A total of 55.8 total products were transfused
with a mean ISS of 25.8 in the sternotomy group
(Fig. 3). The rationale for use of damage control

Mackowski et al.

911

techniques were coagulopathic bleeding in the ster

notomy group and either coagulopathic bleeding or
hemodynamic instability in the thoracotomy group.
Our criteria for re-exploration and definitive chest
wall closure consisted of resolution of hemodynamic
instability and correction of coagulopathy. We were
able to close the chest in 17 (68%) of our patients with
our average duration of packing of 1.3 days for the
entire group. Two of these 17 patients died later in their
hospital stay with their deaths attributed to events
unrelated to packing: acute cardiopulmonary collapse
from suspected pulmonary embolism and anoxic brain
injury. All other patients who were not closed died
secondary to cardiac failure from uncontrollable
hemorrhage within the first hospital day.
Complications in the thoracotomy group included
respiratory failure for six patients with three eventually
requiring tracheostomy. Five patients developed ventilatorassociated pneumonia (all of which had a prolonged
ventilator course). One patient developed empyema
requiring chest tube drainage and one patient de
veloped a superficial wound infection that was man
aged with opening of the wound. In the sternotomy
group, two patients required prolonged mechanical
ventilation for respiratory failure and one for acute
respiratory distress syndrome (ARDS) with all three
requiring tracheostomy. All sternotomy patients re
quiring tracheostomy developed ventilator-associated
pneumonia; and one also developed a catheter-associated
urinary tract infection. The overall mortality rate was
40 per cent, 35 per cent in the thoracotomy group and
50 per cent in the sternotomy group. Mortality was 53
per cent in patients with penetrating trauma and 38 per
cent in blunt trauma. The causes of death for patients in
the thoracotomy group were cardiac failure secondary
to uncontrolled hemorrhage, which occurred within 24
hours of the injury and acute cardiopulmonary collapse
from suspected pulmonary embolism on hospital Day
22. In the sternotomy group, the causes of death for
patients were cardiac failure secondary to uncontrolled
hemorrhage, which occurred within 24 hours of the
injury, hypoxic arrest secondary to severe ARDS
occurring on hospital Day 7 and anoxic brain injury
occurring on hospital Day 2. Readmission was not
required for any patient who survived to discharge to
date.
Discussion
The efficacy and complications for damage control
techniques in the abdomen have been well established
in the current body of literature. Less is true for
damage control surgery in the chest. Rotondo7 de
scribed these techniques for both abdominal trauma
and thoracic injuries. The bony envelope of the chest

912

THE AMERICAN SURGEON

Table 1.

September 2014

Vol. 80

Patient Characteristics fo r Both Thoracotomy and Sternotomy Groups

Thoracotomy

Gender
Male
Female
Average age (years) [rangel
Mechanism
Penetrating
Blunt
Average ISS [range]

Sternotomy

16
2
42 [23-53]

7
0
39 [22-61]

10
8
34.2 [9-59]

7
0
25.8 [18-36]

Overall
23 (92%)
2 (8%)
40 [22-611
17 (68%)
8 (32%)
30.4 [9-59]

ISS, Injury Severity Score.

30

CO

Thoracotomy

Sternotomy

V e n t Days
F ig. 1.

Thoracotomy

Sternotomy

ICU LOS

Thoracotomy

Sternotomy

H o s p ita l LOS

Duration of mechanical ventilation, intensive care unit length of stay, hospital length of stay. Mean [standard error of mean].

poses some challenges to the placement of packs for

control of nonsurgical hemorrhage. Adequate exposure
is a key component when performing these techniques
in the chest. Likewise, an abbreviated operation with
the focus on hemorrhage control is a critical step that
must be taken early in the operative course. Estab
lished techniques for dealing with lung injury are
nonanatomic wedge resection, lobectomy, or even total
pneumonectomy.7 Packing the chest, like the abdo
men, places the patient at risk for complications. Aside
from the risk of infectious complications, the patient
may be placed at risk for inadequate ventilation or
oxygenation as a result of constraints on lung volume;
and overpacking the mediastinum may cause cardiac
tamponade. The trauma surgeon should be cognizant
that the lungs are a lower pressure hemodynamic
system, and so far fewer packs are necessary to achieve
control of coagulopathic hemorrhage from lung pa
renchyma in comparison to other solid organs.
There have been several case series in the literature
describing the application of damage control tech
niques for thoracic trauma. The reported mortality in
these series ranges from 23 to 69 per cent, depending
on the mechanism of injury.8-12 Our reported mortality
falls within this range. O Connor,12 in the largest series
to date, reported the lowest mortality rate of 23 per
cent with an in-hospital mortality rate of 16 per cent

excluding patients who died on extracorporeal membrane

oxygenation. Compared with patients in our pop
ulation, there was a similar mean ISS; however, our
patients were older and more severely injured because
two of our thoracotomy group patients presented with
an ISS of at least 50.12 Stab wounds were absent from
our patient group compared with O Connors, and the
presence of these lower energy injuries may explain
the improved mortality in their results.13 Our patients
also differed from this population in that no clamshell
w
CO

Fig. 2.
mean].

Duration of thoracic packing. Mean [standard error of

No. 9

DAMAGE CONTROL FOR THORACIC TRAUMA

Mackowski et at.

913

Conclusion

cn

F ig. 3. Transfusion requirements for total length of stay. Mean

[standard error of mean],

As the body of evidence regarding the safety and

efficacy of the application of damage control techniques
grows, its use is going to be applied to other injuries. It
already has a well-respected role in the setting of ab
dominal and orthopedic trauma. The adaptation of
damage control techniques to the thoracic body com
partment will require ongoing study. The results in this
study, among others, show promise that the use of an
abbreviated initial operation followed by packing,
temporary closure of the chest, and goal-directed re
suscitation before definitive chest wall closure can be
safely used with accepted outcomes.

thoracotomies were used, and the main injury sus

tained was laceration to the pulmonary parenchyma.
Our average duration of packing was approxi
mately 1.5 days less than those reported in other
series.8- 9> 12 These differences may be accounted for
by the use of packing with temporary closure in the
setting of hemodynamic instability as was performed
in our series. Additionally, all of our procedures were
performed in an operating room setting. This may
portend a better outcome for patients because we
may be selecting out those who are less likely to
survive transfer to the operating room. This increase
in survival of patients who have their emergency
thoracic procedure performed in an operating room
setting has also been previously described by KarmyJones.13
The majority of the complications patients had in
our case series was respiratory failure requiring a pro
longed ventilatory course with the development of
ventilator-associated pneumonia. All of the patients in
our series received postoperative antibiotic pro
phylaxis that extended through the duration of their
temporary packing. All of our patients did require
massive transfusion, and this is associated with an
increased risk of bacterial infections.14 Bouboulis
described the average duration of packing in the
setting of sternal wound infection and dehiscence in
that series was 33 hours, and the only patient to
develop a surgical site infection in our series had
packs in place for 48 hours. One of the patients in
our series developed empyema; this has been seen in
other series as well.12 This may be attributed to our
relatively shorter duration of packing compared with
the other series in the literature, and further in
vestigation will be needed to see if duration of
packing becomes an independent predictor of in
fectious complications.

1. Rotondo MF, Schwab CW, McGonigal MD, et al. Damage

control: an approach for improved survival in exsanguinating
penetrating abdominal injury. J Trauma 1993;35:375-82.
2. Garrison JR, Richardson JD, Hilakos AS, et al. Predicting the
need to pack early for severe intra-abdominal hemorrhage.
J Trauma Inj Infect Crit Care 1996;40:923-9.
3. Asensio JA, Petrone P, Roldan G, et al. Has evolution in
awareness o f guidelines for institution o f damage control improved
outcome in the management of the posttraumatic open abdomen?
Arch Surg 2004;139:209-14.
4. Ordonez CA, Badiel M, Sanchez Al, et al. Improving mor
tality predictions in trauma patients undergoing damage control.
Am Surg 2011;77:778-82.
5. Duchesne JC, Barbeau JM, Islam TM, et al. Damage control
resuscitation: from emergency department to the operating room.
Am Surg 2011;77:201-6.
6. Bouboulis N, Rivas LF, Kuo J, et al. Packing the chest:
a useful technique for intractable bleeding after open heart oper
ation. Ann Thorac Surg 1994;57:856-60.
7. Rotondo MF, Bard MR. Damage control surgery for thoracic
injuries. Injury 2004;35:649-54.
8. Vargo DJ, Battistella FD. Abbreviated thoracotomy and
temporary chest closure: an application of damage control after
thoracic trauma. Arch Surg 2001 ;136:214.
9. Moriwaki Y, Toyoda H, Harunari N, et al. Gauze packing as
damage control for uncontrollable haemorrhage in severe thoracic
trauma. Ann R Coll Surg Engl 2013;95:20-5.
10. Caceres M, Buechter KJ, Tillou A, et al. Thoracic packing
for uncontrolled bleeding in penetrating thoracic injuries. South
Med J 2004;97:637-11.
11. Lang JL, Gonzalez RP, Aldy KN, et al. Does temporary chest
wall closure with or without chest packing improve survival for trauma
patients in shock after emergent thoracotomy? J Trauma 2011;70:705-9.
12. O Connor J, DuBose J, Scalea T, Cowley RA. Damage control
thoracic surgery: management and outcomes. Presented at the 73rd
Annual Meeting of AAST and Clinical Congress of Acute Care Surgery.
13. Karmy-Jones R, Nathens A, Jurkovich GJ, et al. Urgent and
emergent thoracotomy for penetrating chest trauma. J Trauma
2004;56:664-9.
14. Sihler KC, Napolitano LM. Complications o f massive
transfusion. Chest 2010;137:209-20.

Thoracotomy

Sternotomy

REFERENCES

Copyright of American Surgeon is the property of Southeastern Surgical Congress and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.