Needle Thoracostomy in the Treatment of a Tension
Pneumothorax in Trauma Patients: What Size Needle?
Imme Zengerink, MD, Peter R. Brink, MD, PhD, Kevin B. Laupland, MD, MSc, FRCPC,
Earl L. Raber, MD, FRCPC, Dave Zygun, MD, MSc, FRCPC, and John B. Kortbeek, MD, FRCSC, FACS
Background: A tension pneumotho-
rax requires immediate decompression us-
ing a needle thoracostomy. According to
advanced trauma life support guidelines
this procedure is performed in the sec-
ond intercostal space (ICS) in the mid-
clavicular line (MCL), using a 4.5-cm
(2-inch) catheter (5-cm needle). Previous
studies have shown a failure rate of up
to 40% using this technique. Case re-
ports have suggested that this high fail-
ure rate could be because of insufficient
length of the needle.
Objectives: To analyze the average
chest wall thickness (CWT) at the second
ICS in the MCL in a trauma population
and to evaluate the length of the needle used
pleural cavity during inspiration but does not permit it to
escape during expiration.1,2A technique for emergent chest
decompression is needle thoracostomy (NT). NT can be a
life-saving technique that releases the pressure inside the
thoracic space allowing time to properly insert a thoracos-
tomy tube. NT is frequently used to diagnose a TPT by the
rush of air after insertion and converts the existing TPT to a
normal pneumothorax.3In this procedure, a large caliber
needle is inserted in the second intercostal space (ICS) in the
midclavicular line (MCL) of the affected hemithorax, as rec-
ommended by the advanced trauma life support guidelines.4
The needle portion is removed, and the plastic sheath is left
in place.5The largest caliber needle typically used for this
in needle thoracostomy for emergency de-
compression of tension pneumothoraces.
Methods: Retrospective review of
major trauma admissions (Injury Severity
Score >12) at the Foothills Medical Cen-
tre in Calgary, Canada, who underwent a
computed tomography chest scan admit-
ted in the period from October 2001 until
March 2004. Subgroup analysis on men
and women, <40 years of age and >40
years of age was defined a priori. CWT
was measured to the nearest 0.01 cm at the
second ICS in the MCL.
Results: The mean CWT in the 604
male patients and 170 female patients
studied averaged 3.50 cm at the left sec-
ond ICS MCL and 3.51 cm on the right.
The mean CWT was significantly higher
for women than men (p < 0.0001). About
9.9% to 19.3% of the men had a CWT >4.5
cm and 24.1% to 35.4% of the women
Conclusions: A catheter length of 4.5
cm may not penetrate the chest wall of a
substantial amount (9.9%–35.4%) of the
population, depending on age and gender.
This study demonstrates the need for a vari-
able needle length for relief of a tension pneu-
mothorax in certain population groups to
improve effectiveness of needle thoracostomy.
Key Words: Needle thoracostomy,
Tension pneumothorax, Chest wall thick-
ness, Computed tomography.
J Trauma. 2008;64:111–114.
tension pneumothorax (TPT) is a life-threatening con-
dition that is caused by a laceration of the lung. A flap
valve effect may develop that allows air to enter the
purpose is the 14-gauge 5-cm needle with a sheath length of
4.5 cm (2 inch).
Previous studies show a failure rate in the prehospital-
environment of as high as 40%.6,7Numerous case reports
have been published during the last years, suggesting that the
failure rate might be caused by an insufficient length of the
cannula used in NT.3,8–12
In a study by Britten et al.,13the chest wall thickness
(CWT) was measured by ultrasound on patients scheduled for
abdominal ultrasound. Only 4% were found to have a CWT
?4.5 cm. Other studies have measured the CWT on com-
puted tomography (CT) scans in small convenience samples
of trauma patients and found that 25% to 33% of all patients
have a CWT of ?5 cm.14,15
The true CWT at the second ICS MCL is poorly defined
and may have important therapeutic implications.
The primary objective is to analyze the average CWT at
the second ICS in the MCL in a trauma population. The
second is to evaluate the length of the needle used in NT for
emergency decompression of tension pneumothoraces.
This study is a retrospective review of chest CT scans
performed on trauma patients at the Foothills Medical Centre
in Calgary, Canada. For this study, approval from the Con-
joint Health Research Ethics Board was granted. The Foot-
Submitted for publication March 24, 2006.
Accepted for publication July 12, 2006.
Copyright © 2008 by Lippincott Williams & Wilkins
From the Department of Traumatology (I.Z., P.R.B.), University Hos-
pital Maastricht, The Netherlands; Departments of Surgery and Critical Care
(K.B.L., D.Z., J.B.K.) University of Calgary, Foothills Medical Centre,
Calgary Health Region, Canada; and Section of Body CT and MRI (E.L.R.),
Department of Diagnostic Imaging, Foothills Medical Centre, Canada.
Address for reprints: Imme Zengerink, Medical Student, Peter Debye-
laan 25, Maastricht, Limburg, 6229 HX, The Netherlands.
The Journal of TRAUMA?Injury, Infection, and Critical Care
Volume 64 • Number 1
hills Medical Centre is the sole tertiary trauma center serving
Southern Alberta with a referral population of approximately
Radiographic records of all major trauma patients (Injury
Severity Score [ISS] ?12) and age ?18 who were admitted
in the period from October 24, 2001 until March 31, 2004
were reviewed to identify those who underwent a CT scan of
the chest. All of these CT scans were retrieved and reviewed
by the principal author. CT scans that did not meet quality
criteria of visualization and reproduction or did not image
more than half of the clavicles were excluded. The variables
measured were age, gender, CWT, and whether the patient
was scanned with arms up (routine) or down during the
All of the data were collected by review of the CT scans
on the Foothills Medical Centre PACS (Picture Archive and
Communication System) on dedicated diagnostic worksta-
tions (AGFA, Holland, IMPAX platform V 4.5). Information
was collected on a standard data collection Excel-sheet.
Demographic information was obtained from the Alberta
As per the standard of practice for NT treatment of a
TPT, the CWT was measured in the MCL in the second ICS.
Accurate measurement of each patient’s right and left CWT
was calculated by the following method. Review of the coro-
nal scout image (an initial low dose radiograph-like image
from which the subsequent CT scan’s field of view is
planned), allowed for identification of the middle of the
clavicle. From this point on the coronal scout image an
accurate 3D cross-reference was made to the axial images to
subsequently identify the second and third ribs and thus the
second ICS. A precise measurement of the CWT was per-
formed using the workstation’s caliper measurement tool.
Data were analyzed by using Stata 8.0 (Stata Corp,
College Station, TX). Means, standard deviations, and me-
dian with interquartile range (IQR) were used to describe
normally or skewed variables, respectively. Group means
were compared using the Student’s t test and medians using
the Mann-Whitney U test. Fisher’s exact test was used for
comparison of categorical data. A power analysis was per-
formed a priori based on the outcome of the study of Givens
et al.,14with the result of n ? 288. Subgroup analysis on men
and women, ?40 years of age and ?40 years of age were
defined a priori.
A total of 774 patients, of whom at least one side of their
chest wall was measured, were studied; 604 (78%) were men
and 170 (22%) were women. The median age was 40 (IQR,
25–53) years. The mean ISS was 23, the median was 20.
The mean CWT of female patients was significantly
higher than that of the men as shown in Table 1.
For male patients, the CWT increased with age (left side,
3.15–3.60 cm and right side, 3.22–3.62 cm, p ? 0.0001). For
female patients, the chest wall became thinner with age but
the difference was not significant.
In men ?40 years, approximately 10% exceeded a CWT
of 4.5 cm, jumping to 19% for men ?40 years. For women
?40 years, the number with a CWT greater than 4.5 cm was
a third, falling to a fourth for those ?40 years. Figure 1
illustrates the significant difference between the percentages
of each gender in the age group under 40, but not in the age
group of ?40 years.
The effect of the position of the arms during scanning
was particularly noticeable in women. Of the female patients
with their arms down 72.2% (left) and 57.1% (right) had a
CWT ?4.5 cm versus 28.9% (left) and 23.1% (right) with
their arms up (both p ? 0.01). For male patients with their
arms down, these results were 18.9% (left) and 33.3% (right)
versus 14.2% (left) and 13.2% (right) when they had their
arms up in the scanner (p ? 0.01 right, p ? 0.41 left).
Finally, the distance from the anterior wall to the peri-
cardium was measured in each patient. In 19 of 774 patients
(2.45%) a 4.5-cm needle could potentially penetrate the peri-
cardium when passed through the second ICS MCL. There
was no significant difference for age and gender but there was
a nonsignificant tendency toward younger women.
An NT is a potentially life-saving emergency procedure
in patients with a TPT. It is a simple procedure that allows
time for a thoracostomy tube to be inserted using a sterile
technique in an adequate setting. If a TPT is not confirmed by
a rush of air upon needle placement possible explanations are
that a TPT was not present or that the needle did not reach the
pleural space.16CT images were reviewed to define anatomic
depth at the site of NT.
Fig. 1. CWT ?4.5 cm.
Table 1 Mean Chest Wall Thickness (cm) in the
Second Intercostal Space Midclavicular Line
3.84 ? 1.17
3.41 ? 1.04
3.92 ? 1.42
3.37 ? 0.99
The Journal of TRAUMA?Injury, Infection, and Critical Care
To eliminate beam hardening artifact from degrading the
images of the chest on CT, standard protocol is to position
patients in the CT scanner with their arms elevated (up)
\whenever possible. However, in certain circumstances, such
as trauma, this is not always possible. We note in this study
that the CWT increased for men and women when their arms
were left ‘down’. For female patients in the supine position
with their arms positioned ‘up’, breast tissue tends to fall
outwards and to the side, resulting in a thinner CWT. In
female patients whose arms are left ‘down’ during the CT
scan, breast tissue may not be as spread out and this may
result in a significantly higher number of female patients with
a CWT ?4.5 cm. This is a very important practical consid-
eration because while most female patients are scanned in the
arms ‘up’ position, most emergent NTs are performed with
the patient’s arms in what would be the arms ‘down’ position.
This study suggests that a 5-cm needle with a 4.5-cm
sheath may fail to decompress a pneumothorax in around
10% of the male patients under 40 years of age and around
19% of the male patients over 40 years of age. For female
patients, this risk is even higher with approximately one-third
of the women under 40 and one-fourth of the women over 40
years of age having a CWT exceeding 4.5 cm. It also shows
that men increase their CWT over the years; for women, the
effects of age on body habitus may actually reduce CWT at
the second ICS. The number of patients whose pneumotho-
races might not be decompressed by a 4.5-cm needle is
potentially even higher given the effect of arm elevation on
decreasing CWT during imaging.
Givens et al.,14in a similar study in 2004 in Texas,
reported a convenience sample of 111 patients. Twenty-two
percent of all patients had a CWT over 5 cm. The authors did
not differentiate in this percentage between male patients and
female patients. We found in this considerably larger study, a
lower percentage of male patients having a CWT over 4.5 cm
and a higher percentage of female patients having a CWT
over 4.5 cm, even with this lower cut-off point. The IV-
catheter from BD Insyte was used as a reference. The sheath
of the needle measures 2.1 ? 45 mm, the needle itself 50 mm.
An assumption was made that a CWT of exactly 4.50 cm
would not be decompressed by a 4.5-cm catheter in case of
The CWT has been measured in a trauma population to
resemble the population in which a TPT is most likely to
occur. Givens et al.14excluded a woman with a CWT over 10
cm because of subcutaneous air; in this study all people with
subcutaneous emphysema, anterior hematomas, or other chest
wall abnormalities are included. The patients requiring an NT
are more likely to be found in this group.
This study demonstrated that in 2.5% of all trauma pa-
tients, a 4.5-cm needle used for NT could puncture the heart
at the second ICS MCL. This is a potential concern on the left
side; interestingly, in one patient it was the case on the right
side because of a mediastinal shift.
Although CT may be the most suitable, accurate, and
reproducible tool to measure the CWT, there are some limi-
tations with this method, especially in a retrospective review.
The accuracy of where to take the measurement of CWT is
critical, as a small error 1 cm to either side of the MCL could
make a significant difference in the measured CWT because
of the acute degree of slope high in the chest cavity at the
second ICS in some patients. In addition, when patients in a
cervical spine collar are positioned in the CT scanner with
their arms ‘up’, their subcutaneous soft tissues can be distorted
and compressed against the collar. This creates abnormal
skin folds that can spuriously increase the measured CWT.
These anatomic and practical limitations could challenge
the reproducibility of this studies findings. However, we
think that the large sample size makes up for these limita-
tions. Since only one investigator measured all the CT scans
with the cut-off point in mind, this could have lead to a bias.
Practical application of NT would undoubtedly encompass
some variability in site of decompression.
The lack of cadaver verification is a limitation of this
study. CT correlation with cadavers has been limited by
freezing and preservation that changes soft tissue dimensions.17
Use of recent CT technology and comparison to fresh cadaver
studies demonstrates very high correlation.18CT imaging is
currently being used for preoperative sizing of implantable
medical devices with very good fit at time of surgery or
The study has been performed retrospectively in only
one health center in Canada. This population may not be
anatomically representative of other trauma systems with
variation in height, build, and body mass index.
The catheter used in needle decompression of a TPT may
not reach the pleural space in 10% to 19% of men and a
fourth to a third of women, depending on age. However,
several authors,3,12,20do not advise a longer catheter because
of the possibility of subclavian or pulmonary artery injury
and cardiac tamponade. One author actually did recommend
a longer needle if the 14-gauge needle fails.15
In cases of subcutaneous air or chest wall hematoma,
pressure on the catheter may collapse the lumen. To prevent
this, the needle can be left in place3,10with use of a stabili-
zation device,21which also creates more length and prevents
kinking. But leaving the needle in place can be a danger,11
potentially leading to damage to lung tissue or vital struc-
tures. Another suggestion is tousethelateralmidanterioraxillary
line (used for chest tubes) for needle decompression,2,12,22,23
since there is less fat and muscle tissue and less chance of
damage to vessels and heart. However, pleural adhesions are
more likely to be encountered and may increase the risk of
lung injury when a large pneumothorax is not present.24
Considering the tools available now and their advantages and
NT in the Treatment of a TPT in Trauma Patients
Volume 64 • Number 1
disadvantages, the 4.5-cm needle remains the best choice, but
one should be aware of its potential shortcomings.
This study suggests that 4.5-cm needles for decompres-
sion may be inadequate. Interestingly, no device for needle
catheter decompression of TPT has yet been developed to
overcome the conflicting challenges posed by variable chest
thickness versus the potential risk of damage to lung, heart,
and great vessels. These results support performance of a
clinical trial to evaluate effectiveness and risk of a longer
catheter. Development of a longer introducer needle with a
blunt retracting tip may be useful in the clinical setting of
1. Dudley H. Hamilton Bailey’s Emergency Surgery. Bristol: Wright;
2. Leigh-Smith S, Harris T. Tension pneumothorax-time for a re-think?
Emerg Med J. 2005;22:8–16.
3. Cullinane DC, Morris JA Jr, Bass JG, Rutherford EJ. Needle
thoracostomy may not be indicated in the trauma patient. Injury.
4. American College of Surgeons, Committee on Trauma. Advanced
Trauma Life Support Course Manual. Chicago, IL: ACSCOT; 2004;
5. Friend KD. Prehospital recognition of tension pneumothorax.
Prehosp Emerg Care. 2000;4:75–77.
6.Barton ED, Epperson M, Hoyt DB, Fortlage D, Rosen P. Prehospital
needle aspiration and tube thoracostomy in trauma victims: a six-
year experience with aeromedical crews. J Emerg Med. 1995;
7. Davis DP, Pettit K, Rom CD, et al. The safety and efficacy of
prehospital needle and tube thoracostomy by aeromedical personnel.
Prehosp Emerg Care. 2005;9:191–197.
8.Heng K, Bystrzycki A, Fitzgerald M, et al. Complications of
intercostal catheter insertion using EMST techniques for chest
trauma. ANZ J Surg. 2004;74:420–423.
9. Britten S, Palmer SH. Chest wall thickness may limit adequate
drainage of tension pneumothorax by needle thoracocentesis. J Accid
Emerg Med. 1996;13:426–427.
10.Pattison GT. Needle thoracocentesis in tension pneumothorax:
insufficient cannula length and potential failure. Injury. 1996;27:758.
Jones R, Hollingsworth J. Tension pneumothoraces not responding to
needle thoracocentesis. Emerg Med J. 2002;19:176–177.
Jenkins C, Sudheer PS. Needle thoracocentesis fails to diagnose a
large pneumothorax. Anaesthesia. 2000;55:925–926.
Britten S, Palmer SH, Snow TM. Needle thoracocentesis in tension
pneumothorax: insufficient cannula length and potential failure.
Givens ML, Ayotte K, Manifold C. Needle thoracostomy:
implications of computed tomography chest wall thickness [see
comment]. Acad Emerg Med. 2004;11:211–213.
Marinaro J, Kenny CV, Smith SR, et al. Needle thoracostomy in
trauma patients: what catheter length is adequate? Acad Emerg Med.
Mines D, Abbuhl S. Needle thoracostomy fails to detect a fatal
tension pneumothorax. Ann Emerg Med. 1993;22:863–866.
Janssens V, Thys P, Clarys JP, et al. Post-mortem limitations of
body composition analysis by computed tomography. Ergonomics.
Mitsiopoulos N, Baumgartner RN, Heymsfield SB, et al. Cadaver
validation of skeletal muscle measurement by magnetic resonance
imaging and computerized tomography. J Appl Physiol. 1998;
Warriner RK, Haddad M, Hendry PJ, Mussivand T. Virtual
anatomical three-dimensional fit trial for intra-thoracically implanted
medical devices. ASAIO J. 2004;50:354–359.
Butler KL, Best IM, Weaver WL, Bumpers HL. Pulmonary artery
injury and cardiac tamponade after needle decompression of a
suspected tension pneumothorax. J Trauma. 2003;54:610–611.
Allison K, Porter KM, Mason AM. Use of the Asherman chest seal
as a stabilisation device for needle thoracostomy. Emerg Med J.
Biffl WL. Needle thoracostomy: a cautionary note [comment]. Acad
Emerg Med. 2004;11:795–796.
Rawlins R, Brown KM, Carr CS, Cameron CR. Life threatening
haemorrhage after anterior needle aspiration of pneumothoraces. A role
for lateral needle aspiration in emergency decompression of spontaneous
pneumothorax. [comment]. Emerg Med J. 2003;20:383–384.
Goodman LR, Putman CE, eds. Intensive Care Radiology, Imaging
of the Critically Ill. Philadelphia: W.B. Saunders; 1982:99–100.
The Journal of TRAUMA?Injury, Infection, and Critical Care