94 Hong Kong Med J Vol 16 No 2 # April 2010 # www.hkmj.

org
Introduction
Primary spontaneous pneumothorax (PSP) occurs in healthy people without clinically
apparent lung disease.
1
It may result in absence from school, hospital admission, and
signicant morbidity. While the treatments vary, accurate quantication of its size is
important because it helps determine treatment.
2
For PSPs that are small and clinically
asymptomatic, conservative treatment by observation alone is accepted. For larger
PSPs, both thoracentesis and chest tube insertion are established treatment modalities,
though which is the better option remains unclear. In view of uncertainties in diagnosis
and treatment, we therefore conducted this review in corresponding Chinese patients
to: (1) determine their demographics, (2) compare different radiological methods
for quantifying pneumothorax volume, (3) compare the difference in outcomes after
Objectives To (1) determine the demographics of Chinese children admitted
with primary spontaneous pneumothorax, (2) suggest how they
may be quantied radiologically, (3) compare the difference in
outcomes after their primary management by thoracentesis and
chest tube insertion, and (4) review the local experience with
surgical intervention for such children.
Design Retrospective, descriptive study.
Setting Acute tertiary public hospital, Hong Kong.
Patients Consecutive patients younger than 18 years and admitted with
primary spontaneous pneumothorax between 1 January 1999
and 30 September 2007.
Main outcome measures Hospital stay and risk of recurrence after thoracentesis versus
chest tube insertion.
Results Seventy-seven patients with 114 episodes of primary
spontaneous pneumothorax were reviewed. They were
signicantly taller (P<0.001) and thinner (P<0.001) than the
population mean percentile. Both the Light index and Collins
formula were accurate in quantifying pneumothorax volume, but
as the former was simpler and more user-friendly, this was more
applicable in children. Thoracentesis resulted in shorter hospital
stays (mean, 4.6; standard deviation, 1.9 days) than chest tube
insertion (6.9; 3.0 days), but there was no signicant difference in
the recurrence rates within 6 months (P=1.0), 1 year (P=0.9), and
2 years (P=0.1). Insignicant pneumothorax was treated with
observation alone in 16% of the patients. For patients with a
clinically signicant pneumothorax, thoracentesis and chest tube
insertion were successful in 78% and 67%, respectively (P=0.34).
The success rate of video-assisted thoracoscopic surgery was
89%, and postoperative recurrence occurred more commonly in
patients without a lung bleb.
Conclusion Chinese children with primary spontaneous pneumothorax
exhibited similar demographic characteristics to Caucasian
children. Light index is simple and accurate for quantifying
pneumothorax volume in children. Conservative treatment
including observation, thoracentesis, and chest tube insertion
should sufce for most patients with rst episode of primary
spontaneous pneumothorax. Early surgery is warranted for
any patient who fails conservative treatment, for which video-
assisted thoracoscopic surgery is safe and effective.
Management of primary spontaneous
pneumothorax in Chinese children
O R I G I N A L
A R T I C L E
Key words
Chest tubes; Child; Pneumothorax;
Recurrence; Treatment outcome
Hong Kong Med J 2010;16:94-100
United Christian Hospital, Kwun Tong,
Hong Kong:
Department of Paediatrics and
Adolescent Medicine
LPY Lee, FHKCPaed, FHKAM (Paediatrics)
WK Chiu, FHKCPaed, FHKAM (Paediatrics)
HB Chan, FRCPCH, FHKAM (Paediatrics)
Department of Radiology and Organ
Imaging
MHY Lai, MB, ChB, FRCR
Division of Paediatric Surgery,
Department of Surgery
KKW Liu, FHKAM (Surgery), FRCS
Division of Paediatric Surgery,
Department of Surgery, Queen Elizabeth
Hospital, Hong Kong
MWY Leung, FHKAM (Surgery), FRCS
Correspondence to: Dr LPY Lee
Email: leepyl@ha.org.hk
Lilian PY Lee
Miranda HY Lai
WK Chiu
Michael WY Leung
Kelvin KW Liu
HB Chan

CME
# Primary spontaneous pneumothorax in children #
Hong Kong Med J Vol 16 No 2 # April 2010 # www.hkmj.org 95

199911200793018

77114

P<0.001P<0.001Light index
Collins

4.61.9
6.93.0
6P=1.01P=0.92
P=0.116%

78%67%
P=0.3489%

Light index

thoracentesis versus chest tube insertion, and (4)

provide recommendations for PSP treatment.
Methods
Patient selection
Patients were recruited from the United Christian
Hospital, a 1400-bed acute tertiary public hospital in
Hong Kong serving a regional population of 900 000
inhabitants. Consecutive patients younger than 18
years and admitted with PSP during the period from
1 January 1999 to 30 September 2007 were included.
Those with secondary spontaneous pneumothorax
(due to pre-existing lung diseases, traumatic and
iatrogenic factors) were excluded.
Data collection
Medical records were retrieved by the Clinical Data
Analysis and Reporting System using the ninth
revision of the International Classication of Diseases,
Clinical Modication codes: 512.8 for spontaneous
pneumothorax, and 512.0 for spontaneous tension
pneumothorax. Data abstracted included patient
demographics, presenting symptoms, treatment
modalities on admission, volume of air aspirated,
length of hospital stay, episodes of recurrence,
indications for surgery, and nal outcome. The
patients were contacted by phone and their electronic
patient record was accessed from the Clinical
Management System to retrieve information on any
subsequent episodes and/or surgery performed in
other hospitals.
Quantication of size of pneumothorax
Chest X-rays on admission were retrieved and were
all commented by the same paediatric radiologist,
who was blinded to all other clinical data. The size of
the pneumothorax was quantied by two methods.
The Light index (Box
3
) calculates the volume from
cubes of the average lung diameter and the average
hemithorax diameter. The Collins formula (Box
4
)
derives the volume by summation of the interpleural
distances measured at three locations, versus the
pneumothorax size measured by helical computed
tomography (CT). The amount of air aspirated
(retrieved from each patients medical record)
was presumed to be the actual volume of the
pneumothorax.
Treatment protocol
The British Thoracic Society (BTS) Research
Committee issued the rst guideline concerning
the treatment of PSP in 1993.
5
They reported
that thoracentesis and chest tube drainage were
equally successful in rst and recurrent episodes
of spontaneous pneumothorax.
5
Our department
adopted the use of chest tube insertion as the initial
treatment of PSP, unless the pneumothorax was
just a small rim of air around the lung, for which
observation was used.
5
Referral for surgery was made
if the pneumothorax persisted for more than 3 days.
This BTS guideline was revised in 2003.
6

Thoracentesis was suggested as the rst step in
BOX.
Light index
3
Y = (A
3
B
3
)/A
3
Percentage volume of a pneumothorax (Y) approximates to the ratio of the
cube of the hemithorax diameter (A) to the collapsed lung diameter (B).
Collins formula
4
Y = 4.2 + [4.7 x (A + B + C)]
Y = percentage pneumothorax size
A = distance between lung apex and the uppermost point of the pleural cavity
B = distance between the midpoint of the upper half of collapsed lung to the
hemithorax
C = distance between the midpoint of the lower half of collapsed lung to the
hemithorax
# Lee et al #
96 Hong Kong Med J Vol 16 No 2 # April 2010 # www.hkmj.org
the management of PSP for patients who required
intervention.
6
Since January 2005, our department
adopted the 2003 guideline and managed all patients
requiring intervention with thoracentesis. Signicant
pneumothorax was dened as the presence of a
visible rim of more than 2 cm between the lung
margin and the chest wall.
6
Thoracentesis was
repeated for persistent signicant air collection. A
chest drain was inserted after failure of the second
thoracentesis. Referral for surgery was mandated if
the pneumothorax persisted for more than 3 days
despite a chest drain.
Successful outcome was dened as sustained
complete or near-complete re-expansion of the lung
for at least 24 hours after treatment. All patients were
followed up after discharge. The mean duration of
follow-up was 4.2 years.
Statistical analysis
Patient demographic characteristics and progress
summaries were compiled from descriptive data
analysis. For analysing continuous variables, one-
and two-sample t tests were used; whereas the Chi
squared test was used to compare categorical data.
Pearson correlation coefcients were computed
for measuring the association between continuous
variables.
This study was approved by our Institutional
Review Board for research.
Results
Seventy-seven patients with 114 episodes of PSP were
admitted between 1 January 1999 and 30 September
2007. The numbers of patients admitted before and
after January 2005 (ie when the new BTS guideline was
adopted) were 58 and 56, respectively. At the time of
diagnosis, the mean patient age was 16 years (range,
14-18 years), the male-to-female ratio was 10:1, and 11
(14%) were smokers. Chest pain (n=113, 99%) with or
without dyspnoea (n=31, 27%) was the most common
presenting symptom. About two thirds (n=75, 66%)
suffered a left-sided pneumothorax. Using the
middle 50th percentile information and applying
the one-sample t test, our patients were signicantly
taller (P<0.001) and thinner (P<0.001) than the
population mean. Similarly, the body mass index was
also signicantly lower for the corresponding mean
population percentile (P<0.001) [Table 1].
Of the 114 admissions of pneumothorax,
85 admission chest X-rays could be retrieved and
analysed by the same paediatric radiologist. The size
of the pneumothorax was quantied by both the
Light index and the Collins formula, both correlated
well with the actual volume; Pearsons correlation
coefcients being 0.846 and 0.841, respectively
(P<0.001).
Twelve (16%) of the patients with a clinically
insignicant pneumothorax were treated by
observation alone. Of the 65 patients not deemed
suitable for observation, six (9%) had absolute
indications for surgery (four had haemopneumothorax
and two had tension pneumothorax). The remaining
59 patients were treated with either thoracentesis
(n=23, 30%) or chest tube insertion (n=36, 47%),
depending on whether the admission was after or
before January 2005 (Fig).
All 12 patients treated by observation were
successfully discharged within 7 days. In all, four
patients had surgery; three had clinically signicant
recurrence with persistent air leak despite chest tube
insertion, and one for bilateral apical bullae. Surgery
was performed between 3 months and 3 years after
the rst occurrence of these children. The child with
bilateral apical bullae was electively admitted for
surgery 6 months after the rst PSP.
Among the 23 patients having thoracentesis as
the initial primary intervention, 18 were successfully
treated. Signicant recurrence occurred in four
patients, who were treated with thoracentesis and
chest tube insertion. The recurrences occurred
between 19 days and 7 months after the rst PSP.
Two of these patients received surgery for persistent
air leak, and two were successfully treated by
thoracentesis (thus surgery having been refused
by the parents). Five of the 23 patients required
an additional chest drain after failure of primary
thoracentesis. Three of these resolved successfully,
whereas two with persistent air leaks underwent
subsequent surgery.
For the 36 patients who primarily treated with
chest tube insertion, 24 were successfully treated, but
the remaining 12 underwent surgery for persistent
air leak. Eleven of the 24 patients successfully
treated with a primary chest drain had signicant
recurrences, with persistent air leak and surgery was
performed subsequently. Nine patients experienced
recurrences within 2 years of the rst PSP and two
after 4 years.
Demographics (test value: 50th percentile) Mean (%) Standard deviation (%) Mean difference (95% condence interval) P value
Mean height percentile 64 26 14 (8 to 20) <0.001
Mean weight percentile 37 24 -13 (-19 to -8) <0.001
Mean body mass index 30 23 -20 (-26 to -15) <0.001
TABLE 1. Body height, body weight, and body mass index of patients with primary spontaneous pneumothorax
# Primary spontaneous pneumothorax in children #
Hong Kong Med J Vol 16 No 2 # April 2010 # www.hkmj.org 97
The success rates of thoracentesis and chest
tube insertion as primary treatments were 78% (18/23)
and 67% (24/36), respectively (P=0.34), indicating
comparable success rates with both interventions.
The overall success rate of conservative treatment
(including observation, thoracentesis, and chest tube
insertion either alone or in combination) for the rst
episode of PSP was 80% (57/71).
The baseline characteristics of patients
primarily treated with thoracentesis and chest tubes
were comparable (Table 2). Corresponding outcomes
including length of hospital stay, and recurrence
within 6 months, 1 year, and 2 years are shown in
Table 3. In our series, the length of hospital stay
was signicantly shorter in the thoracentesis group
(P=0.006). The overall recurrence rate after successful
primary conservative treatment was 27% (16/59). As
shown in Table 3, there was no statistically signicant
difference in recurrence rates between the two
groups at 6 months, 1 year, and 2 years, indicating
that thoracentesis as the initial intervention was not
associated with a higher recurrence rate.
A total of 37 patients underwent surgery
because of persistent air leak (despite chest drain),
haemopneumothorax, tension pneumothorax,
and bilateral pneumothorax. Apical pleurodesis or
pleurectomy with or without bullectomy via video-
assisted thoracoscopic surgery (VATS) was the
procedure used in our unit (Table 4).
Bleb(s) was/were found in 70% (26/37) of the
patients. Four patients had postoperative recurrence,
FIG. Flowchart showing outcomes of different treatment groups
Patients admitted for 1st episode of primary spontaneous pneumothorax (n=77)
Observation (n=12) Intervention (n=65)
Surgery (n=6)
4 Haemopneumothorax
2 Tension pneumothorax
Thoracentesis (n=23) Chest drain (n=36)
Success (n=12) Success (n=18) Failed thoracentesis, with subsequent
chest drain insertion (n=5)
Success (n=24) Failure and required
surgery (n=12)
Signicant
recurrence and
received surgery
(n=3)
Surgery for bilateral
apical bullae (n=1)
Signicant
recurrence (n=4)
2 Received
surgery
2 Refused
surgery
Signicant
recurrence (n=1),
but refused surgery
Success (n=3) Failure and required
surgery (n=2)
Signicant
recurrence and
received surgery
(n=11)
* SD denotes standard deviation

Owing to missing chest X-ray, the number of chest X-rays retrieved and analysed was 17
and 15 for the thoracentesis group and chest drain group, respectively
Characteristic Thoracentesis
(n=18)
Chest drain
(n=24)
P value
Sex
Female
Male
4 (22%)
14 (78%)
1 (4%)
23 (96%)
0.074
Mean (SD*) age (years) 16 (1) 16 (1) 0.263
Mean (SD) body height (cm) 169 (7) 173 (6) 0.112
Mean (SD) body weight (kg) 52 (6) 52 (5) 0.898
Mean (SD) body mass index (kg/m
2
) 18 (2) 17 (2) 0.209
Light index 50%

7 (39%) 8 (33%) 0.710

TABLE 2. Baseline characteristics of patients with primary spontaneous
pneumothorax having thoracentesis and chest drains
* SD denotes standard deviation
TABLE 3. Outcomes of patients primarily treated by thoracentesis and chest drains
Outcomes Thoracentesis
(n=18)
Chest drain
(n=24)
P value
Mean (SD*) length of hospital
stay (days)
4.6 (1.9) 6.9 (3.0) 0.006
Recurrence in 6 months 3 (17%) 4 (17%) 1.000
Recurrence in 1 year 4 (22%) 5 (21%) 0.914
Recurrence in 2 years 4 (22%) 11 (46%) 0.114
# Lee et al #
98 Hong Kong Med J Vol 16 No 2 # April 2010 # www.hkmj.org
giving a success rate of 89% (33/37). Among the latter,
one was found to have another upper lobe bleb on the
ipsilateral side 2 years after surgery. This patient was
treated by bullectomy with extensive mechanical and
chemical pleurodesis. In the remaining three patients,
no bleb was found during the second surgery, all
of whom were treated by extensive mechanical and
chemical pleurodesis. Postoperative recurrence risk
was signicantly greater in without-bleb than with-
bleb patients (27% vs 4%, P=0.036; Table 5).
Discussion
Demographic characteristics
The annual incidence of PSP is 7.4 to 18 per 100 000
men and 1.2 to 6 per 100 000 women.
1
It generally
occurs in tall, thin males between the ages of 10 and
30 years
1
; the reported male-to-female ratio of PSP
being 6.9:1.
7
In our series of 77 patients (114 episodes)
with PSP encountered within a period of 8 years and
9 months, the male-to-female ratio was 10:1. The
subjects were also taller, thinner, and had lower body
mass index values than the general population.
The percentage having chest pain and shortness
of breath as presenting symptoms was similar to that
in other paediatric series.
7
About one seventh of our
patients were smokers. In our series, there was also
a predilection for left-sided pneumothorax, as in
previously reported studies.
7,8
Radiological quantication of pneumothorax for
children
The Collins formula is derived from the summation
of the interpleural distances measured at three
locations versus the pneumothorax size measured
by helical CT.
4
The Light index is calculated from the
cubes of the average lung diameter and the average
hemithorax diameter.
3
Both are widely used for
estimating pneumothorax volume.
Using linear regression and correlation
analyses, a study by Noppen et al
9
provided proof that
the size of PSP estimated by Light index associates
well with the amount of air manually aspirated.
9
The
quoted correlation (r=0.84, P<0.0001) corresponds
to that derived in our study. However, Noppens
study
9
might be biased towards moderate and large
pneumothoraces, since small pneumothoraces are
often treated by observation (without drainage). The
same limitation also applied to our study.
The accuracy of the Collins formula was
validated by the excellent correlation (r=0.98, P<0.0001)
between percentage pneumothorax size estimated
by the formula and that calculated using helical CT.
4

However, we observed that the Collins formula was
generally associated with a larger pneumothorax size
than that estimated by the Light index, and some of
the values were greater than 100% (up to 160% without
evidence of tension pneumothorax on chest X-ray).
The most likely reason was that the Collins formula
was derived from a sample of adults and did not take
account of thoracic size. Our paediatric patients had
smaller thoraces and were signicantly taller and
thinner than the general population, leading to a
possible overestimation of pneumothorax volume.
In comparison with the Collins formula, the Light
index is simpler and more user-friendly. Given its
high correlation with PSP volume, we propose that
it should be the method of choice in paediatric
practice.
Because of record retrieval failure, only 85 pre-
aspiration chest X-rays were accessed. The correlation
might have been better still if all of the chest X-rays could
have been analysed, which was a limitation due to the
retrospective nature of our study. Another limitation
was the use of volume of aspirated air as a reection
of pneumothorax volume. The amount of aspirated air
may not truly reect the actual volume, because there
was usually small amount of residual air in the thoracic
cavity after the procedure. In this series, we did not
review post-aspiration chest X-rays to estimate the size
of any residual pneumothorax. Comparison between
the pre-aspiration and post-aspiration chest X-rays
taken immediately after thoracentesis may give a more
accurate correlation with pneumothorax volumes
extracted. However, this would have been technically
difcult and not at all practical.
Outcomes following thoracentesis versus
chest drain as primary treatment for primary
spontaneous pneumothorax
Patients treated by thoracentesis had a shorter
hospital stay compared with those treated by chest
TABLE 4. Indications for surgery following primary spontaneous
pneumothorax
Indications for surgery No. of patients (n=37)
Persistent air leak 30 (81%)
Haemopneumothorax 4 (11%)
Tension pneumothorax 2 (5%)
Bilateral pneumothorax 1 (3%)
* P=0.036
TABLE 5. Patient numbers with pathologically conrmed bleb and postoperative
recurrence
Postoperative recurrence With bleb
(n=26)
Without bleb
(n=11)
Total (n=37)
Yes* 1 (4%) 3 (27%) 4 (11%)
No 25 (96%) 8 (73%) 33 (89%)
# Primary spontaneous pneumothorax in children #
Hong Kong Med J Vol 16 No 2 # April 2010 # www.hkmj.org 99
tube insertion. Thoracentesis is safe and simple to
perform, entails a lower dose of systemic analgesia
and inicts less pain. Our results conformed with
a previous report that thoracentesis was more
advantageous than chest tube insertion based on
lower total pain scores and shorter hospital stays.
10

Another study showed that chest drain insertion
involved more pain and more analgesic use.
11
Other
possible complications of chest tube drainage
include: malposition, kinking or blockage of chest
drains, surgical emphysema, pleural infection,
6
and
wound problems.
The risk of recurrence of primary spontaneous
pneumothorax following conservative treatment is
about 54% within the rst 4 years.
12
According to a
survey of several studies involving widely varying
follow-up periods (0.25-10 years), the average
recurrence rate was 30% (range, 16-52%).
13
Most
recurrences occurred within 6 months to 2 years after
the initial pneumothorax.
1
The recurrence rate after
successful primary conservative treatment in our
patient series without a bleb was 27%.
Recurrence rates after PSPs initially treated
by thoracentesis versus chest tube insertion had
been compared. A systematic review by Devanand
et al
14
summarised the results of three randomised
controlled trials
10,15,16
and concluded that there was
no signicant difference in PSP recurrence rates
at 3 and 12 months. A more recent study with the
largest patient pool in 2006 by Ayed et al
17
also found
no signicant difference in recurrence rates 1 and 2
years after either form of initial treatment. A similar
nding was also noted in our series.
In our series, the success rate of thoracentesis
was 78%. Devanand et al
14
reported that the success
rate of simple aspiration varied from 67 to 93%,
which concurred with our results. Noppen et al
16

found no statistically signicant difference between
recurrence rates after thoracentesis and chest tube
insertion, indicating they were similar in efcacy.
Effectiveness of video-assisted thoracoscopic
surgery for primary spontaneous pneumothorax
Open thoracotomy has been the gold standard for
surgery following pneumothorax. Contemporary
advances in surgical technique favour minimally
invasive procedures in the form of VATS. It is used
to resect blebs (if any), the likely underlying cause of
PSP, as well as to create pleural adhesions to prevent
recurrence. In a Korean study,
18
blebs were conrmed
to be present in 85% of patients undergoing
thoracotomy. In another series, 78 to 95% of patients
presenting with PSP had endoscopically visible blebs
or bullae.
19
In our series, 70% of patients undergoing
VATS had blebs.
The reported recurrence rate of spontaneous
pneumothorax after VATS is 2 to 14%,
1
and was 11%
in our series. Ayed and Al-Din
20
reported that the rate
was low (0-3%) after surgery if blebs were identied.
In our series, after surgery, of 26 patients with blebs,
one had a recurrence (4%). The latter underwent
bullectomy with apical pleurectomy yet endured a
recurrence 2 years later; a bleb was found again at the
upper lobe, though the recurrence might have been
caused by a new bleb. This patient was subsequently
treated with bullectomy and more extensive
mechanical and chemical pleurodesis.
Postoperative recurrence was more frequent in
patients without blebs or bullae being noted, the rate
being 27%.
20
Our study also supported this nding in
that a signicantly higher percentage of non-bleb
patients (27%) had postoperative recurrence (Table
5). The exact reason for this observation remains
unknown. The underlying pathology of the non-
bleb group is obscure and may not be denitely
amendable to surgery. The with-bleb group had an
underlying pathology eradicated by apical bullectomy
plus apical pleurectomy and thus the risk of further
recurrence was minimised.
Our overall experience in managing primary
spontaneous pneumothorax and treatment
recommendations
For the rst episode of PSP, 16% of our patients were
treated with observation alone. Over half (58%)
were successfully treated by thoracentesis and/or
chest tube insertion. The remaining 26% underwent
surgery for persistent air leak, haemopneumothorax,
and tension pneumothorax.
For all patients presenting with PSP, the
treatment ow advocated in the BTS guideline 2003
is recommended. Thoracentesis is suggested as
the initial treatment for signicant volumes, as it is
less invasive and results in comparable recurrence
rates and efcacy as chest tube insertion. Early
referral for surgery is recommended for patients
who fail conservative treatment. For patients with
haemopneumothorax and tension pneumothorax,
urgent surgical referral is suggested after initial
stabilisation.
Conclusion
Our patients with PSP showed similar demographic
characteristics as in other studies.
1,7
For quantication
of the pneumothorax volume (as suggested by the
BTS guideline), the modied Light index was accurate
and simple to perform. Thoracentesis is a well-
accepted initial intervention for PSP. Compared with
chest tube drainage, it leads to shorter hospital stays
without an increase in recurrence rate. Conservative
treatment including observation, thoracentesis and
chest tube insertion sufce for most patients with rst
# Lee et al #
100 Hong Kong Med J Vol 16 No 2 # April 2010 # www.hkmj.org
episode of PSP. Early referral to surgery is warranted
for those whose PSP fails to resolve with conservative
treatment. Video-assisted thoracoscopic surgery is
a safe and effective therapeutic approach for such
patients.
Acknowledgement
The authors would like to sincerely thank Prof Kelvin
KW Yau of the City University of Hong Kong for the
statistical support.
1. Sahn SA, Heffner JE. Spontaneous pneumothorax. N Engl J
Med 2000;342:868-74.
2. Chan SS, Lam PK. Simple aspiration as initial treatment
for primary spontaneous pneumothorax: results of 91
consecutive cases. J Emerg Med 2005;28:133-8.
3. Light R. Pleural diseases. 3rd ed. Philadelphia: Lea &
Febinger; 1990.
4. Collins CD, Lopez A, Mathie A, Wood V, Jackson JE,
Roddie ME. Quantication of pneumothorax size on chest
radiographs using interpleural distances: regression analysis
based on volume measurements from helical CT. AJR Am J
Roentgenol 1995;165:1127-30.
5. Miller AC, Harvey JE. Guidelines for the management of
spontaneous pneumothorax. Standards of Care Committee,
British Thoracic Society. BMJ 1993;307:114-6.
6. Henry M, Arnold T, Harvey J; Pleural Diseases Group,
Standards of Care Committee, British Thoracic Society.
BTS guidelines for the management of spontaneous
pneumothorax. Thorax 2003;58 Suppl 2:ii39-52.
7. Hui Y. Adolescent primary spontaneous pneumothorax: a
hospitals experience. Hong Kong J Paediatr 2006;11:128-
32.
8. Poenaru D, Yazbeck S, Murphy S. Primary spontaneous
pneumothorax in children. J Pediatr Surg 1994;29:1183-5.
9. Noppen M, Alexander P, Driesen P, Slabbynck H, Verstraete
A; Vlaamse Werkgroep voor Medische Thoracoscopie en
Interventionele Bronchoscopie. Quantication of the size of
primary spontaneous pneumothorax: accuracy of the Light
index. Respiration 2001;68:396-9.
10. Harvey J, Prescott RJ. Simple aspiration versus intercostal
tube drainage for spontaneous pneumothorax in patients
with normal lungs. British Thoracic Society Research
Committee. BMJ 1994;309:1338-9.
11. Luketich JD, Kiss M, Hershey J, et al. Chest tube insertion:
a prospective evaluation of pain management. Clin J Pain
1998;14:152-4.
12. Sadikot RT, Greene T, Meadows K, Arnold AG. Recurrence of
primary spontaneous pneumothorax. Thorax 1997;52:805-
9.
13. Schramel FM, Postmus PE, Vanderschueren RG. Current
aspects of spontaneous pneumothorax. Eur Respir J
1997;10:1372-9.
14. Devanand A, Koh MS, Ong TH, et al. Simple aspiration
versus chest-tube insertion in the management of primary
spontaneous pneumothorax: a systematic review. Respir
Med 2004;98:579-90.
15. Andrivet P, Djedaini K, Teboul JL, Brochard L, Dreyfuss
D. Spontaneous pneumothorax. Comparison of thoracic
drainage vs immediate or delayed needle aspiration. Chest
1995;108:335-9.
16. Noppen M, Alexander P, Driesen P, Slabbynck H,
Verstraeten A. Manual aspiration versus chest tube drainage
in rst episodes of primary spontaneous pneumothorax:
a multicenter, prospective, randomized pilot study. Am J
Respir Crit Care Med 2002;165:1240-4.
17. Ayed AK, Chandrasekaran C, Sukumar M. Aspiration versus
tube drainage in primary spontaneous pneumothorax: a
randomised study. Eur Respir J 2006;27:477-82.
18. Kim J, Kim K, Shim YM, et al. Video-assisted thoracic surgery
as a primary therapy for primary spontaneous pneumothorax.
Decision making by the guideline of high-resolution
computed tomography. Surg Endosc 1998;12:1290-3.
19. Ng CS, Lee TW, Wan S, Yim AP. Video assisted thoracic
surgery in the management of spontaneous pneumothorax:
the current status. Postgrad Med J 2006;82:179-85.
20. Ayed AK, Al-Din HJ. The results of thoracoscopic surgery for
primary spontaneous pneumothorax. Chest 2000;118:235-
8.
References