consensus conference

Management of Spontaneous
Pneumothorax*
An American College of Chest Physicians Delphi
Consensus Statement
Michael H. Baumann, MD, FCCP; Charlie Strange, MD, FCCP;
John E. Heffner, MD, FCCP; Richard Light, MD, FCCP; Thomas J. Kirby, MD;
Jeffrey Klein, MD, FCCP; James D. Luketich, MD; Edward A. Panacek, MD, FCCP;
and Steven A. Sahn, MD, FCCP; for the ACCP Pneumothorax Consensus Group

Objective: Provide explicit expert-based consensus recommendations for the management of adults with
primary and secondary spontaneous pneumothoraces in an emergency department and inpatient hospital
setting. The use of opinion was made explicit by employing a structured questionnaire, appropriateness
scores, and consensus scores with a Delphi technique. The guideline was designed to be relevant to
physicians who make management decisions for the care of patients with pneumothorax.
Options: Decisions for observation, chest tube placement, surgical interventions, and radiographic
imaging.
Outcomes: Effectiveness of pneumothorax resolution, duration of and patient tolerance of care, and
pneumothorax recurrence.
Evidence: Literature review from 1967 to January 1999 and Delphi questionnaire submitted in three
iterations to a multidisciplinary physician panel.
Values: The guideline development group determined by consensus the relevant outcomes to be
considered in developing the Delphi questionnaire.
Benefits, harms, and costs: The type and magnitude of benefits, harms, and costs expected for patients
from guideline implementation.
Recommendations: Management decisions vary between patients with primary or secondary pneumothoraces, with observation of small pneumothoraces being appropriate only for primary pneumothoraces. The level of consensus varies regarding the specific interventions indicated, but agreement
exists for the general principles of care.
Validation: Recommendations were peer reviewed by physician experts and were reviewed by the
American College of Chest Physicians (ACCP) Health and Science Policy Committee.
Implementation: The guideline recommendations will be published in printed and electronic form
with distribution of synopses for patients and health care providers. Contents of the guideline will be
incorporated into continuing medical education programs.
Sponsors: The ACCP.
(CHEST 2001; 119:590 602)
Key words: consensus; Delphi; guideline; management; pneumothorax; practice guideline; spontaneous pneumothorax
Abbreviations: ACCP American College of Chest Physicians; BTS British Thoracic Society

*From the Division of Pulmonary and Critical Care Medicine

(Dr. Baumann), University of Mississippi Medical Center, Jackson, MS; Division of Pulmonary and Critical Care Medicine (Drs.
Strange and Sahn), Medical University of South Carolina (Dr.
Heffner), Charleston, SC; Vanderbilt University (Dr. Light),
Saint Thomas Hospital, Nashville, TN; Thoracic Surgery (Dr.
Kirby); University Hospitals of Cleveland, Cleveland, OH; Department of Radiology (Dr. Klein), University of Vermont College of Medicine, Burlington, VT; Thoracic Surgery (Dr. Luketich), University of Pittsburgh Medical Center, Pittsburgh, PA;
and Emergency Medicine (Dr. Panacek), University of California-Davis, Sacramento, CA.

A complete list of the consensus group is located in Appendix 1.

Additional information about the questionnaire, consensus tables,
and other data are available at www.chestnet.org/publications/
18098/index.html.
Manuscript received June 2, 2000; revision accepted September
14, 2000.
Correspondence to: Michael H. Baumann, MD, FCCP, Associate
Professor of Medicine, Division of Pulmonary and Critical Care
Medicine, University of Mississippi Medical Center, 2500 North
State St, Jackson, MS 39216-4505; e-mail: mbaumann@medicine.
umsmed.edu

590

Consensus Conference

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

pneumothoraces, which occur in the

S pontaneous
absence of thoracic trauma, are classified as

primary or secondary.1 Primary spontaneous pneumothoraces affect patients who do not have clinically
apparent lung disorders. Secondary pneumothoraces
occur in the setting of underlying pulmonary disease,
which most often is COPD.
Although primary and secondary spontaneous
pneumothoraces affect 20,000 patients per year in
the United States2 and account for nearly
$130,000,000 in health-care expenditures each year,3
generally accepted and methodologically sound
guidelines for the care of these patients do not exist.
Consequently, observational studies demonstrate extensive practice variation in the management of this
relatively common condition.4
To address this variation in care, the American
College of Chest Physicians (ACCP) commissioned
the development of a practice guideline for the
management of spontaneous pneumothorax. The
guideline committee recognized that insufficient
data existed from randomized controlled trials to
develop an evidence-based document and that recommendations would largely derive from expert
opinion. Because informal approaches for developing expert-based statements are subject to extensive
bias, the guideline developers selected the Delphi
technique5 to formalize the expert panels consensus
process and explicitly state opinion. The methodology for this consensus guideline provides clinicians
with a description of the level of consensus achieved
for each treatment recommendation and identifies
clinical settings wherein multiple options for care
exist. The guideline pertains to adult patients with
primary spontaneous pneumothorax and patients with
secondary pneumothorax associated with COPD.
Many of the recommendations will have relevance to
secondary pneumothoraces affecting patients with underlying lung disorders other than COPD.
Materials and Methods
The guideline development process used the Delphi method to
create and quantify group consensus (Fig 1). The Delphi method
was developed by RAND Corporation (Santa Monica, CA)
researchers in the 1950s.5 Characteristics of the Delphi method
are anonymity, controlled feedback, and statistical group response.6 Anonymity derives from the absence of face-to-face
interaction. Participants respond independently to questionnaires, and responses are communicated to other participants
without being attributed to specific individuals. Controlled feedback occurs during several questionnaire iterations. Opinions
expressed during one round of the questionnaire are returned to
the group during the next round in the form of statistical
summaries. The statistical group response is the final stage of the
Delphi method with the group consensus expressed as a statistical score. The results of the questionnaire are expressed using
summary decision rules that quantify the level of consensus and
the appropriateness of management recommendations.7

Guideline Development Committee and Expert Panel Members

The ACCP Health and Science Policy Committee selected the
content chairman, the content co-chairman, and the methodology
chairman. The chairmen selected six members of a multidisciplinary guideline development committee on the basis of the
members previous publications on the topic of pneumothorax.
The chairmen met with the project development committee to
organize the Delphi process and to select members of the expert
panel. Panel members were selected from specialty fields proportionally related to the distribution of publications on the
management of pneumothorax among specialty and subspecialty
journals. This proportionality was determined by a MEDLINE
literature search from 1966 to 1997 (see below). Experts were
eligible for selection if they had published a peer-reviewed article
on pneumothorax during the previous 5 years. Each member
provided a written statement disclosing the existence of any
corporate relationships related to the care of patients with
pneumothoraces. The distribution of panel members among
medical specialties were as follows: pulmonary/critical care, 12
members (38%); thoracic surgery, 12 members (38%); general
surgery, 1 member (3%); interventional radiology, 3 members
(9%); and emergency medicine, 4 members (13%).
Literature Search
A MEDLINE literature search of English language articles
was performed for the period from 1966 to 1997. The MeSH
heading of spontaneous pneumothorax was combined with the
terms randomized controlled trials, meta-analysis, and guidelines.
Recent review articles were searched for additional randomized
controlled trials. Retrieved articles were distributed to panel
members. The literature search was repeated during each of the
three iterations of the Delphi questionnaire, with the last literature search occurring in January 1999. Retrieved articles were
graded by the two content chairmen on the basis of the articles
study designs (Table 1).8 The methodology chairman resolved
grading disagreements with a majority vote.
Additional articles were identified by the panel members and
were communicated to the development group through the
Delphi questionnaire. Abstracts of these articles were distributed
to the panel during the next round of the questionnaire.
Delphi Questionnaire
The guideline development committee met to design a questionnaire that would query panel members about management
decisions in the care of patients with primary spontaneous
pneumothoraces and secondary pneumothoraces due to COPD.
The committee constructed a decision tree for the care of
patients with pneumothoraces and selected decision branch
points for inclusion into the questionnaire that were considered
by the committee to be key management practices. The committee did not discuss the appropriateness of these practices so as to
avoid influencing the questionnaire development or bias members in attendance who would later respond to the questionnaire.
Most questions were case-based scenarios with multiple management options presented as choices (Fig 2). Panel members
were asked to respond to the appropriateness of each option
using a 9-point Likert scale (Table 2). A few questions were
open-ended, multiple choice, or requested a yes or no
response.
The multidisciplinary experts were allowed to skip questionnaire items by indicating that they did not have sufficient
knowledge or experience to respond to a particular question.
Panel members also were asked to indicate whether their reCHEST / 119 / 2 / FEBRUARY, 2001

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

591

Figure 1. Steps of the Delphi method. HSPC Health and Science Policy Committee of ACCP.

sponses were based on opinion or an interpretation of published

investigations. Panel members were provided a space to present
an argument or literature citations in support of their opinions.
The questionnaire listed on its face page definitions of terms
and clinical assumptions (Tables 3, 4).
Administration of the Questionnaire
The first Delphi questionnaire was mailed to the panel members with a request for its completion and return within 2 weeks.
Responses on the returned questionnaires were summarized. A
second questionnaire was developed that included a summary of
the panel members responses to each of the first questionnaires
items, a synopsis of the panel members comments, and a list of
the articles cited by the panel members in support of their
questionnaire responses. Questionnaire items that were identified by the panel members as ambiguous were refined.

Summaries of item responses were placed adjacent to the

specific item and were described as follows. The number of panel
members responding to each item was listed. Bar and dot
symbols were placed adjacent to the Likert scales to indicate
median responses, middle 50% range, and the range for all
responses. Similar summary statistics were presented for openended questions that requested a numeric response. A number
reported the proportion of panel members responding yes or
no to an item. Panel members were provided with a key for
each questionnaire that explained the data summary techniques.
This second questionnaire was mailed to panel members.
Responses to the second questionnaires items, the panel members comments, and cited literature were summarized and
incorporated into a third questionnaire that was mailed to the
panel members. Bar and dot symbols (Fig 2) were placed over
the Likert scales to indicate median responses, the middle 50%
range, the middle 80% range, and outlier responses. The third

592

Consensus Conference

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

Table 1Levels of Evidence for Studies Evaluating

Treatment Effectiveness*
Level of Evidence
and Grade
Level I

Level II

Level III
Level IV
Level V
Grade
Grade
Grade
Grade

A
B
C
D

Grade E

definitions to determine levels of consensus (Table 5). Evidence

cited in the questionnaire by panel members to support their
questionnaire responses also was cited in the guideline text with
an evidence grade.

Study Design
Large, randomized trials with clear-cut results;
low risk of false-positive () error or falsenegative () error
Small, randomized trials with uncertain
results; moderate to high risk of falsepositive and/or false-negative error
Nonrandomized, contemporaneous control
Nonrandomized, historical control subjects,
and expert opinion
Case series, uncontrolled studies, and expert
opinion
Supported by at least two level I investigations
Supported by only one level I investigation
Supported by level II investigations only
Supported by at least one level III
investigation
Supported by level IV or level V evidence

*Adapted from Taylor.8

mailing included printed copies of the abstracts from the articles

cited by panel members in support of their responses to specific
questionnaire items.
Description of Level of Consensus and Degree of EvidenceBased Support
Responses to the third questionnaires items that used the
9-point Likert scale were summarized and applied to a priori

Description of Appropriateness of Management Options

Management options were graded regarding appropriateness
using the summary results of the Likert scale (Table 6). Depending on the panel recommendations and the level of consensus
achieved, the guideline uses the words must, should, and
may to identify recommendations that are standards (must),
guidelines (should), or options (may) for care9 (Table 6). This
language is keyed directly to the panel members scored responses. Management approaches are described as inappropriate
when a high degree of consensus indicated that the intervention
must not be employed for any patient in any clinical circumstance. Because of the lack of high-grade evidence in the
management of pneumothorax and the expert-opinion basis of
the guideline, few interventions are described as inappropriate.
Although based on previously reported approaches,10,11 methods for assessing and reporting the level of consensus and
appropriateness were developed during this project and are
unique to this guideline statement.

Results and Management Recommendations

Literature Search
The literature search retrieved nine articles,1220
which included eight randomized controlled trials1219 (Table 7), no meta-analyses, and one practice

Figure 2. A sample item on the questionnaire showing the statistical summaries of the panel members
responses from the previous round. The solid dots above and below the Likert scales indicate the
median responses. The bars above and below the Likert scale show the middle 50% and the middle
80% responses, respectively. The open dots represent outlier responses.
CHEST / 119 / 2 / FEBRUARY, 2001

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

593

Table 2Expert Opinion Ratings

Likert Scale
9
7 and 8
46
2 and 3
1

Definition
Extremely appropriate: treatment of choice
(may have more than one per question).
Appropriate: a first-line treatment you would
often use.
Equivocal: a second-line treatment you would
sometimes use (eg, after first line had failed).
Usually inappropriate: at most, a third-line
treatment you would rarely use.
Extremely inappropriate: a treatment you would
never use.

guideline.20 The analysis of the retrieved articles

indicated that all of the guideline recommendations
were grade E (lowest grade of evidence).
Delphi Technique
Three questionnaire iterations were completed
with 100% participation in the first iteration, 97%
participation (31 of 32) in the second iteration (a
thoracic surgeon dropped out), and 94% participation (30 of 32) in the third iteration (two thoracic
surgeons dropped out). The guideline incorporates
the consensus opinions of the 30 members who

completed all three questionnaires. The degree of

consensus increased or remained stable during the
Delphi process for most questionnaire items (68%).
Primary Spontaneous Pneumothorax
Clinically Stable Patients With Small Pneumothoraces: Clinically stable patients with small pneumothoraces should be observed in the emergency department for 3 to 6 h and discharged home if a
repeat cheat radiograph excludes progression of the
pneumothorax (good consensus). Patients should be
provided with careful instructions for follow-up
within 12 h to 2 days, depending on circumstances. A
chest radiograph should be obtained at the follow-up
appointment to document resolution of the pneumothorax. Patients may be admitted for observation if
they live distant from emergency services or follow-up care is considered unreliable (good consensus). Simple aspiration of the pneumothorax or
insertion of a chest tube is not appropriate for most
patients (good consensus), unless the pneumothorax
enlarges. The presence of symptoms for 24 h does
not alter the treatment recommendations.
Clinically Stable Patients With Large Pneumothoraces: Clinically stable patients with large pneumo-

Table 3Questionnaire Definition of Terms

Terms

Definition

Spontaneous pneumothorax
Primary spontaneous pneumothorax
Secondary spontaneous pneumothorax
Pneumothorax size
Small pneumothorax
Large pneumothorax
Patient age groups, yr
Young
Older
Clinical stability
Stable patient

No antecedent traumatic or iatrogenic cause

No clinically apparent underlying lung abnormalities or underlying conditions
known to promote pneumothorax (eg, HIV disease)
Clinically apparent underlying lung disease
Determined by distance from the lung apex to the ipsilateral thoracic cupola at
the parietal surface as determined by an upright standard radiograph
3 cm apex-to-cupola distance
3 cm apex-to-cupola distance
1840
40

Unstable patient
Drainage tubes
Small chest tube or small percutaneous
catheter
Moderate-sized chest tube
Large chest tube
Simple aspiration
Sclerosis (pleurodesis) procedure
Chemical pleurodesis
Open or surgical pleurodesis

All of the following present: respiratory rate, 24 breaths/min; heart rate, 60

beats/min or 120 beats/min; normal BP; room air O2 saturation, 90%; and
patient can speak in whole sentences between breaths
Any patient not fulfilling the definition of stable
14F
16F to 22F
24F to 36F
Insertion of a needle or cannula with removal of pleural air followed by
immediate removal of the needle or cannula
Intrapleural instillation of a sclerosing agent through a chest tube or percutaneous
catheter
Pleurodesis performed with a thoracoscope or through a limited or full
thoracotomy

594

Consensus Conference

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

Table 4 Clinical Assumptions

Patients will comply with treatment recommendations and can
obtain prompt emergency medical care
Questions related to secondary spontaneous pneumothoraces
pertain to patients with underlying COPD
Patients have no comorbidities not mentioned in the case scenarios
Pneumothorax is the cause of the patients presenting clinical
manifestations
Care recommendations do not consider patient preferences
First-time pneumothorax unless otherwise indicated

thoraces should undergo a procedure to reexpand

the lung and should be hospitalized in most instances
(very good consensus). The lung should be reexpanded by using a small-bore catheter ( 14F) or
placement of a 16F to 22F chest tube (good consensus). Catheters or tubes may be attached either to a
Heimlich valve (good consensus) or to a water seal
device (good consensus) and may be left in place
until the lung expands against the chest wall and air
leaks have resolved. If the lung fails to reexpand
quickly, suction should be applied to a water-seal
device. Alternatively, suction may be applied immediately after chest tube placement for all patients
managed with a water seal system (some consensus).
Reliable patients who are unwilling to undergo
hospitalization may be discharged home from the
emergency department with a small-bore catheter
attached to a Heimlich valve if the lung has reexpanded after the removal of pleural air (good consensus). Follow-up should be arranged within 2 days.
The presence of symptoms for 24 h does not alter
management recommendations.
Clinically Unstable Patients With Large Pneumothoraces: Unstable patients with large pneumothoraces should undergo hospitalization with insertion
of a chest catheter to reexpand the lung (very good
consensus). Most patients should be treated with a

16F to 22F standard chest tube (good consensus) or

with a small-bore catheter (good consensus), depending on the degree of clinical instability. A 24F to
28F standard chest tube may be used if the patient is
anticipated to have a bronchopleural fistula with a
large air leak or requires positive-pressure ventilation (good consensus). A water seal device can be
used without suction initially (good consensus), but
suction should be applied if the lung fails to reexpand with water seal drainage. Suction may be
alternatively applied immediately after chest tube
placement (some consensus). Some patients may be
managed with a small-bore catheter attached to a
Heimlich valve if clinical stability can be obtained
with immediate evacuation of the pleural space
(good consensus). A water seal device should be
substituted for the Heimlich valve and suction applied if the lung fails to reexpand (good consensus).
Chest Tube Removal: Chest tubes should be removed in a staged manner so as to ensure that the air
leak into the pleural space has resolved (good consensus). The first stage requires that a chest radiograph demonstrates complete resolution of the
pneumothorax and that there is no clinical evidence
of an ongoing air leak. Any suction applied to the
chest tube should be discontinued (good consensus).
Fifty-three percent of panel members would
never clamp a chest tube to detect the presence of an
air leak after reexpansion of the lung. The remaining
panel members would clamp the chest tube approximately 4 h after the last evidence of an air leak.
Regardless of whether the tube was or was not
clamped, panel members would repeat a chest radiograph 5 to 12 h after the last evidence of an air
leak (62% of members) to ensure that the pneumothorax had not reoccurred in preparation for pulling
the chest tube. Other panel members would wait
4 h (10%), 13 to 23 h (10%), or 24 h (17%) before
repeating a chest radiograph.

Table 5Consensus Definitions*

Term

Definition

Perfect consensus
Very good consensus

All respondents agree on an answer

Median and middle 50% (interquartile range) of respondents are found at one integer (eg, median and
interquartile range are both at 8) or 80% of respondents are within one integer of the median (eg, median
is 8, 80% respondents are from 7 to 9)
50% of respondents are within one integer of the median (eg, median is 8, 50% of respondents are from 7 to
9) or 80% of the respondents are within two integers of the median (eg, median is 7, 80% of respondents
are from 5 to 9).
50% or respondents are within two integers of the median (eg, median is 7, 50% of respondents are from 5
to 9) or 80% of respondents are within three integers of the median (eg, median is 6, 80% of respondents
are from 3 to 9).
All other responses.

Good consensus

Some consensus

No consensus

*Definitions refer to Likert scale (Nos. 1 to 9) for responses. See Table 2.

CHEST / 119 / 2 / FEBRUARY, 2001

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

595

Table 6 Management Definitions*

Management Options
Preferred management in most circumstances
Acceptable management in many
circumstances
Acceptable management in certain
circumstances
Acceptable management in rare
circumstances
Inappropriate management
Indeterminate

Median
(Middle 50% Range)

Strength of Recommendation

79
(79)
79
(49)
46
(49)
2 and 3
(1 4)
(13)
All other median and range
combinations including no
consensus

Must if perfect consensus; should otherwise

Should if no preferred management exists; may if a
preferred management exists
May
May
Must not if perfect consensus; should not otherwise
No management recommendation

*Median scores for responses to questionnaire items asking for ranking of appropriateness are given on a scale of 1 to 9. See Table 2 for definitions.
See Table 5.

Persistent Air Leaks: For patients with persistent

air leaks, the panel recommended continued observation for 4 days for spontaneous closure of bronchopleural fistula (median, 4 days; interquartile
range, 3 to 5 days; middle 80% range, 2 to 6 days).
Patients with air leaks persisting beyond 4 days
should be evaluated for surgery to close the air leak
and to perform a pleurodesis procedure to prevent
pneumothorax recurrence (very good consensus).
Although the relative value of thoracoscopy compared to a limited thoracotomy has not been clearly
defined, the panel selected thoracoscopy as the
preferred management (very good consensus). Patients should not undergo the placement of an
additional chest tube or bronchoscopy with attempts
to seal endobronchial sites of air leaks (very good
consensus). Most patients should not be managed
with chemical pleurodesis by instilling sclerosing
agents through the chest tube except in special
circumstances in which surgery is contraindicated or
patients refuse an operative procedure (very good
consensus). If chemical pleurodesis is performed,
doxycycline or talc slurry is the preferred sclerosing
agent (good consensus).
Pneumothorax Recurrence Prevention: Except for
patients with persistent air leaks, procedures to
prevent the recurrence of a primary spontaneous
pneumothorax should be reserved for the second
pneumothorax occurrence (85% of panel members).
Fifteen percent of panel members, however, would
offer patients an intervention to prevent a recurrence after the first pneumothorax. Patients preferences and interests in continuing activities that
would place them at high risk if a pneumothorax
reoccurred (eg, scuba diving or flying) should be
considered in deciding the timing of the intervention. Thoracoscopy is the preferred intervention for

preventing pneumothorax recurrence (very good

consensus). The instillation of sclerosing agents
through a chest tube is an acceptable approach for
pneumothorax prevention in patients who wish to
avoid surgery and for patients who present increased
surgical risk (eg, bleeding diathesis) (good consensus). Success rates with chemical pleurodesis, however, are only 78 to 91% compared to success rates of
95 to 100% with surgical interventions.3
Patients selected for surgical prevention of pneumothorax recurrence should be managed by thoracoscopy (very good consensus). The panel could not
agree on the utility of limited (axillary) thoracotomy
in recurrence prevention. The panel noted that
clinical trials that include patients with primary
spontaneous pneumothorax do not demonstrate the
superiority of thoracoscopy vs limited thoracotomy in
pneumothorax prevention12,21; the panels preference for thoracoscopy was based on practice preferences.
Thoracoscopy can be performed with or without
video assistance. Patients with apical bullae visualized at surgery should undergo intraoperative bullectomy (very good consensus). Bullectomy should be
performed by staple bullectomy (very good consensus). Options for eliminating bullae include electrocoagulation, laser ablation, or hand sewing, depending on institutional expertise and experience with
these procedures. Intraoperative pleurodesis should
be performed in most patients with parietal pleural
abrasion limited to the upper half of the hemithorax
(good consensus). Parietal pleurectomy (some consensus) is an acceptable alternative pleurodesis technique. No consensus was reached regarding the
utility of talc poudrage in primary spontaneous pneumothorax recurrence prevention.
Chest Imaging With CT: The panel did not rec-

596

Consensus Conference

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

Table 7Characteristics of Randomized Controlled Spontaneous Pneumothorax Trials*

Study
General treatment trials
Ma et al14
Andrivet et al17

Harvey and Prescott19

Engdahl et al16

Pleurodesis trials
Light et al15

Almind et al18

van den Brande and

Staelens13
Surgical trial
Waller et al12

Cohorts and Patient Characteristics

Results Summary

Short-wave diathermy treatment vs observation;

n 11 in each group
28 patients treated with thoracic drainage and
33 with needle aspiration

Air absorption rate significantly greater with shortwave diathermy

Higher success rate for patients undergoing
thoracic drainage than with needle aspiration;
no difference in mean length of hospital stay
Longer hospital stay and greater daily pain in
patients with intercostal tube drains
Visual analog pain scale scores lower in the
bupivacaine group

Simple aspiration (n 35) vs intercostal tube

drainage (n 38)
Indwelling chest drains with interpleural
bupivacaine (n 11) vs saline solution
placebo (n 11)
Spontaneous pneumothorax patients with chest
tubes randomized to intrapleural tetracycline
(n 113) vs control group (n 116)
Spontaneous pneumothorax patients in three
treatment groups: simple drainage (n 34);
drainage/tetracycline (n 33); and drainage/
talc (n 29)
Primary spontaneous pneumothorax only: 10
patients with drainage tetracycline/glucose;
10 patients with drainage alone

5-year study period: pneumothorax recurrence

rates lower in the tetracycline group

30 patients with VATS vs 30 patients with

posterolateral thoracotomy for persistent air
leak or pneumothorax recurrence

Greater postoperative decline in lung function in

thoracotomy group; no difference in
postoperative stay, recurrence, or morphine use;
longer operating time for VATS

Talc with significant pneumothorax recurrence

reduction compared to simple drainage;
tetracycline recurrence reduction no different
than other two groups.
Pleurodesis with reduction in early but not late
recurrences

*VATS video-assisted thoracoscopic surgery.

ommend the routine use of chest CT imaging for

patients with a first-time pneumothorax (good consensus). The panel did not achieve consensus regarding the utility of chest CT scans for evaluating
patients with recurrent pneumothoraces, persistent
air leaks, or planned surgical interventions. Chest CT
may be indicated to evaluate the presence of pulmonary disorders, such as interstitial lung disease, that
are suspected clinically but are not apparent on
standard radiographs.
Age Considerations: The questionnaire did not
query the panel regarding the importance of age in
making management decisions.
Secondary Spontaneous Pneumothorax
Clinically Stable Patients With Small Pneumothoraces: Clinically stable patients with small pneumothoraces should be hospitalized (good consensus).
Patients should not be managed in the emergency
department with observation or simple aspiration
without hospitalization (very good consensus). Hospitalized patients may be observed (good consensus)
or treated with a chest tube (some consensus),
depending on the extent of their symptoms and the
course of their pneumothorax. Some of the panel

members argued against observation alone because

of a report of deaths with this approach.22 Patients
should not be referred for thoracoscopy without
prior stabilization (very good consensus). The presence of symptoms for 24 h did not alter the panel
members recommendations.
Clinically Stable Patients With Large Pneumothoraces: Clinically stable patients with large pneumothoraces should undergo the placement of a chest
tube to reexpand the lung and should be hospitalized
(very good consensus). Patients should not be referred for thoracoscopy without prior stabilization
with a chest tube (very good consensus). The presence of symptoms for 24 h did not alter the panel
members recommendations.
Clinically Unstable Patients With Pneumothoraces
of Any Size: Patients should undergo placement of a
chest tube to reexpand the lung and should be
hospitalized (very good consensus). Patients should
not be referred for thoracoscopy without prior stabilization with a chest tube (very good consensus).
Chest Tube Management: The size of chest tubes
used for patients with secondary pneumothoraces
CHEST / 119 / 2 / FEBRUARY, 2001

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

597

depends on clinical circumstances.

Unstable patients (very good consensus) and patients who appear to be at risk for large pleural air
leaks because they require mechanical ventilation
(good consensus) should be managed with a 24F to
28F chest tube. Larger chest tubes were not considered necessary by the panel members (some consensus).
Stable patients who are not at risk for large air
leaks should be managed with 16F to 22F chest
tubes (good consensus), although a small-bore catheter ( 14F) may be acceptable in certain circumstances, which would include small pneumothoraces
and patient preference (good consensus). Some
members of the panel were concerned with the risk
for occlusion of a small-bore catheter.
Attachment of the chest tube to a water seal
device with (some consensus) or without (good
consensus) suction is acceptable management for
most patients. Patients treated with water seal alone
should be managed with suction if the lung fails to
reexpand (good consensus). A Heimlich valve (good
consensus) may be used, although the panel considered a water seal device to be a better option for
most patients.
Pneumothorax Recurrence Prevention: Most
members (81%) of the panel recommend an intervention to prevent pneumothorax recurrence after
the first occurrence because of the potential lethality
of secondary pneumothoraces.2326 The remaining
19% of panel members would perform an intervention to prevent recurrence after the second spontaneous pneumothorax. The preferred management
for pneumothorax recurrence prevention is surgical
(very good consensus) because of the lower recurrence rates with these interventions compared to the
instillation of a sclerosing agent through a chest
tube.3 The instillation of a sclerosant through a chest
tube, however, may be used in certain circumstances
(good consensus) based on patients contraindications to surgery, management preferences, or a poor
prognosis from the patients underlying disease.
Medical or surgical thoracoscopy is preferred
management (very good consensus), although a muscle-sparing (axillary) thoracotomy is an acceptable
alternative (good consensus). A standard thoracotomy through a lateral or median sternotomy approach is not appropriate therapy for most patients
(good consensus).
Most members of the panel recommend bullectomy and a procedure to produce pleural symphysis
during the surgical intervention. Staple bullectomy
was the preferred procedure for bullectomy (very
good consensus). Other methods for bullectomy

were ranked as indeterminate to inappropriate, with

levels of consensus that ranged from no consensus to
good consensus. Acceptable interventions to produce pleural symphysis included parietal pleurectomy (some consensus), talc insufflation (poudrage)
(some consensus), and parietal pleural abrasion
(good consensus). Fibrin pleurodesis and intraoperative instillation of sclerosing agents other than talc
were considered to be rarely acceptable (some consensus). Parietal pleurectomy or parietal pleural
abrasion limited to the upper half of the hemithorax
constitutes the preferred therapy for most patients
(good consensus).
For producing pleural symphysis by instillation of
a sclerosing agent through a chest tube, doxycycline
(good consensus) and talc slurry (very good consensus) were the preferred agents. Minocycline was
considered to be an acceptable alternative agent for
some patients (good consensus), but bleomycin was
considered rarely acceptable (good consensus).
Assessment of Pulmonary Function: The performance of pulmonary function tests to assist management decisions is considered inappropriate (perfect
consensus) for patients presenting with secondary
pneumothoraces. Performing forced expiratory maneuvers during the acute phase of a pneumothorax is
ill-advised and may produce inaccurate results.
Seventy-seven percent of the panel members indicated, however, that results from previously performed pulmonary function tests would assist patient
selection for an intervention to prevent a pneumothorax recurrence in special circumstances (good
consensus). Those circumstances include patients
with relatively good pulmonary function with a
strong desire to avoid a procedure to prevent a
recurrence. Such patients would most likely tolerate
another spontaneous pneumothorax with a low risk
of death. Conversely, a patient with poor lung function who decides to avoid recurrence prevention
should be counseled that such a decision would be
ill-advised.
Persistent Air Leaks: For patients with persistent
air leaks who are selected for observation with
prolonged chest tube drainage because they initially
refuse a surgical procedure, the panel recommended
continued observation for 5 days (median, 5 days;
interquartile range, 4 to 7 days; middle 80% range, 2
to 7 days) before encouraging the patient to accept a
surgical intervention. More prolonged delays may
decrease the effectiveness of thoracoscopy27 and
increase the cost of care.28 The recommendations for
surgical interventions for patients with prolonged air
leaks are similar to those for recurrence prevention

598

Consensus Conference

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

(see above). The panel concluded that the instillation

of chemical agents through a chest tube to produce
a pleural symphysis in managing persistent air leaks
was appropriate management for patients who were
not operative candidates (good consensus). If this
technique was used, doxycycline (good consensus)
and talc (very good consensus) were the preferred
agents.
Chest Tube Removal: For patients treated with a
chest tube without referral for a surgical intervention
to prevent a recurrence, management decisions for
removing the chest tube are similar with a few
exceptions to those for patients with a primary
pneumothorax.
Forty-one percent of panel members would never
clamp a chest tube to detect the presence of an air
leak after reexpansion of the lung. The remaining
panel members would clamp the chest tube 5 to 12 h
after the last evidence of an air leak. Regardless of
whether the tube was or was not clamped, panel
members would repeat a chest radiograph 13 to 23 h
after the last evidence of an air leak (63% of
members) to ensure that the pneumothorax had not
reoccurred in preparation for pulling the chest tube.
Other panel members would wait 4 h (4%), 5 to
12 h (18%), or 24 h (15%).
Chest Imaging With CT: The panel could not
develop recommendations for the use of chest CT
scanning after the first occurrence of a pneumothorax. Obtaining a chest CT scan was considered
acceptable management for patients with pneumothorax recurrence (good consensus), during management of a persistent air leak (some consensus), and
for planning a surgical intervention (some consensus). Chest CT scans might be especially useful if
lung volume reduction surgery was being considered
as an adjunctive procedure.
Age Considerations: Ninety percent of the panel
members did not incorporate the patients age into
the determination of management decisions.

Comparison to Previous Guidelines

Only one previous guideline exists for the management of pneumothorax.20 A panel of two physicians
representing the Standards of Care Committee of
the British Thoracic Society (BTS) developed this
guideline by disseminating a draft to 450 physician
members of the BTS. The two authors modified the
draft on the basis of the 1,052 comments received
from 150 responding physicians. The guideline
methodology did not use a formal literature search.

The audience of the BTS guideline was hospitalbased doctors who were not respiratory specialists
but who directed the initial management of patients
with pneumothoraces.
Both the present ACCP and the BTS guidelines
base treatment recommendations on the severity of
symptoms and the degree of lung collapse, as determined by chest radiographs. Symptom assessments
in the BTS guideline, however, are based only on the
presence or absence of obvious deterioration in usual
exercise tolerance (termed significant dyspnea). The
BTS statement also uses a different method for
grading the degree of lung collapse that includes
levels of small (small rim of air around lung), moderate (lung collapsed halfway toward heart border),
and complete pneumothorax (airless lung).
The BTS statement emphasizes the utility of observation without pleural drainage as initial management for patients without significant dyspnea who
have (1) small or moderately sized primary pneumothoraces or (2) small secondary pneumothoraces.
Simple aspiration with immediate catheter removal
is the initial intervention recommended for the
remaining patients. The placement of a chest tube
with water-seal drainage without suction is recommended only for patients who fail simple aspiration.
The present ACCP guideline consensus process
found simple aspiration to be appropriate rarely in
any clinical circumstance, although two panel members argued that simple aspiration is usually effective
for stable patients.
In the BTS statement, hospitalization is recommended only for patients with secondary pneumothoraces. No specific recommendations are provided
for patients with persistent air leaks or for patients
who require surgery.

Strengths and Limitations of the

Guideline
The present guideline used the Delphi method,
which combines limited evidence with expert opinion and inference in a manner that limits group bias
to the greatest degree possible.29 The guideline
adhered to evidence-based medicine principles of
being relevant to specific circumstances and patients.30 Because the recommendations are largely
expert opinion based, however, they do not represent sufficiently strong evidence to form the basis for
health-care policy.31 Physicians applying these recommendations in patient care should consider the
assumptions presented to the panel members (Table
4) and the unique problems presented by individual
patients who require a flexible clinical approach.
CHEST / 119 / 2 / FEBRUARY, 2001

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

599

Guideline Implementation and Consensus

Data
The complete guideline and the consensus tables
for the entire questionnaire are available on the internet (www.chestnet.org/publications/18098/index.html).
A summary of the guideline and algorithms are available on the Internet and are available for distribution
by the ACCP. A quick reference guide also will be
available.
Priorities for Future Research
The extensive search of the literature underscores
the paucity of high-grade data from clinical trials on
which recommendations for the care of patients with
pneumothoraces can be based. Major limitations of
the literature include the following: pooling of patients with primary and secondary pneumothoraces;
nonstandardized interventions; lack of information
on clinical course (natural history of untreated pneumothorax in different clinical settings); risk stratification on the basis of factors such as the severity of
underlying lung disease, age, and comorbidities;
absence of health-related quality-of-life outcomes
and the patients perspective regarding different
treatment options; and the relative cost-effectiveness
of approaches to care.
These issues call for prospective studies that have
adequate sample sizes and follow-up periods to show
effects. Study end points should include the relevant
clinical outcomes, such as mortality, morbidity, patient perceptions, functional status, and cost.
ACKNOWLEDGMENTS: The guideline developers acknowledge Sydney Parker, PhD, and Beth Welch of the ACCP for their
assistance in coordinating the panels activities; Celia Barbieri,
MS, of the University of Arizona Phoenix-based campuses for
managing the database and questionnaire analyses; and Carolyn
Davis of the University of Mississippi for secretarial support.

Appendix 1: ACCP Pneumothorax Consensus

Group
Content Chairman
Michael H. Baumann, MD, FCCP (Pulmonary), Associate Professor of Medicine, Division of Pulmonary and Critical Care Medicine, University of Mississippi Medical Center, Jackson, MS.
Content Co-Chairman
Charlie Strange, MD, FCCP (Pulmonary), Associate Professor
of Medicine, Division of Pulmonary and Critical Care Medicine,
Medical University of South Carolina, Charleston, SC.
Methodology Chairman
John E. Heffner, MD, FCCP (Pulmonary), Professor of Medicine, Associate Dean, Medical University of South Carolina,
Charleston, SC.

Project Development Committee Members

Richard Light, MD, FCCP (Pulmonary), Professor of Medicine, Vanderbilt University, Director, Pulmonary Diseases, Saint
Thomas Hospital, Nashville, TN; Steven A. Sahn, MD, FCCP
(Pulmonary), Professor of Medicine and Director, Division of
Pulmonary and Critical Care Medicine, Allergy and Clinical
Immunology, Medical University of South Carolina, Charleston,
SC; Thomas J. Kirby, MD (Thoracic Surgery), Co-Chairman and
Professor of Surgery, Director, Thoracic Surgery, University
Hospitals of Cleveland, Cleveland, OH; Jeffrey Klein, MD,
FCCP (Interventional Radiology), Associate Professor and Vice
Chairman, Department of Radiology, University of Vermont
College of Medicine, Burlington, VT; James D. Luketich, MD
(Thoracic Surgery), Assistant Professor of Surgery, Section Head,
Thoracic Surgery, Co-Director, Minimally Invasive Surgery Center, Co-Director, Lung Cancer Center, University of Pittsburgh
Medical Center, Pittsburgh, PA; and Edward A. Panacek, MD,
FCCP (Emergency Medicine), Professor of Medicine, Emergency Medicine Residency Director, University of CaliforniaDavis, Sacramento, CA.
Expert Panel
Pulmonologist Participants: Andrew C. Miller, MD, PhD,
FCCP, Consultant Physician, Mayday University Hospital, Croydon, UK; Colin Selby, DM, Consultant Physician, Queen Margaret Hospital, Dunfermline, Fife, UK; Ian Johnston, MD,
Consultant Physician, University Hospital, Queens Medical Center, Nottingham, UK; John Harvey, MD, Consultant Respiratory
Physician, Southmead Hospital, Bristol, UK; Marc Noppen, MD,
PhD, Head, Interventional Endoscopy Unit, Respiratory Division, Academic Hospital AZ-VUB, Brussels, Belgium;
F. M. N. H. Schramel, MD, PhD, Department of Pulmonary
Diseases, St. Antonius Hospital, The Netherlands; and Harmeet
S. Gill, MD, FCCP, Desert Pulmonary Consultants, PLC, Desert
Samaritan Regional Medical Center, Mesa, AZ.
Surgeon Participants: David Waller, FRCS (Cth), Consultant
Thoracic Surgeon, Glenfield Hospital, Leicester, UK; Richard
Berrisford, ChM (CTh), Consultant Thoracic Surgeon, Royal
Devon and Exeter NHS Trust, Exeter, UK; Anthony P. C. Yim,
MD, FCCP, Chief, Division of Cardiothoracic Surgery, Prince of
Wales Hospital, The Chinese University of Hong Kong, Satin,
New Territories, Hong Kong; Jose Ribas Milanez de Campos,
MD, Associate Professor of Division of Thoracic Surgery, Hospital Das Clinicas, University of Sao Paulo Medical School, Sao
Paulo, Brazil; Mark Ferguson, MD, FCCP, Professor, Department of Surgery, The University of Chicago, Chicago, IL; Andrus
Votik, MD, MSc, PhD, Surgeon-in-Chief, The Salvation Army
Scarborough Grace Hospital, Scarborough, Ontario, Canada;
Kerry Lynn Paape, MD, Cardiovascular Institute of the South,
Houma, LA; James E. Stephenson, MD, Palmetto State Surgical
Associates, Greenville Memorial Hospital, Department of Surgery, Greenville, SC; Averel B. Snyder, MD, Peachtree Cardiovascular and Thoracic Surgeons, PA, Atlanta, GA; Steven M.
Keller, MD, FCCP, Associate Professor, Chief of Thoracic
Surgery, Beth Israel Medical Center, New York, NY; and Rodney
J. Landreneau, MD, Professor of Surgery, Head, Division of
General Thoracic Surgery, Allegheny General Hospital Campus,
Medical College of Pennsylvania, Pittsburgh, PA.
Radiology Participants: Edward F. Patz, Jr, MD, Professor of
Radiology, Department of Radiology, Duke University Medical
Center, Durham, NC; and Dewey Conces, MD, Professor of
Radiology, Indiana University Hospital, Indianapolis, IN.
Emergency Medicine Participants: Kenneth Frumkin, PhD,
MD, Emergency Department, Chesapeake General Hospital,
Chesapeake, VA; Joseph Varon, MD, FCCP, Associate Professor

600

Consensus Conference

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

of Medicine, Pulmonary and Critical Care Section, Baylor College of Medicine, Research Director, Emergency Services, The
Methodist Hospital, Houston, TX; and Seth Wright, MD, MPH,
Associate Professor of Emergency Medicine, Department of
Emergency Medicine, Vanderbilt University, Nashville, TN.
Questionnaire Design Review: Gerard Silvestri, MD, Associate
Professor of Medicine, Division of Pulmonary and Critical Care
Medicine, Medical University of South Carolina, Charleston, SC.

Appendix 2: Additional References Cited by

the Panel in the Questionnaire Iterations
Bense L, Lewander R, Eklund G, et al. Nonsmoking, non-1antitrypsin deficiency-induced emphysema in nonsmokers with
healed spontaneous pneumothorax, identified by computed tomography of the lungs. Chest 1993; 103:433 438
Bertrand P, Regnard J, Spaggiari L, et al. Immediate and
long-term results after surgical treatment of primary spontaneous
pneumothorax by VATS. Ann Thorac Surg 1996; 61:16411645
Campisi P, Voitk AJ. Outpatient treatment of spontaneous pneumothorax in a community hospital using a Heimlich flutter valve:
a case series. J Emerg Med 1997; 15:115119
Cannon W, Mark J, Jamplis R. Pneumothorax: a therapeutic
update. Am J Surg 1981; 142:26 29
Conces DJ, Tarver RD, Gray WC, et al. Treatment of pneumothoraces utilizing small caliber chest tubes. Chest 1988; 94:5557
Gammie JS, Banks MC, Fuhrman CR, et al. The pigtail catheter
for pleural drainage: a less invasive alternative to tube thoracostomy. JSLS (J Soc Laparoendosc Surg) 1999; 3:57 61
Gobbel W. Spontaneous pneumothorax. J Thorac Cardiovasc
Surg 1963; 46:331345
Graham A. Videothoracoscopy and spontaneous pneumothorax.
Ann Thorac Surg 1995; 59:266 267
Janssen J, Schramel F, Sutedja T, et al. Videothoracoscopic
appearance of first and recurrent pneumothorax. Chest 1995;
108:330 334
Jones J. A place for aspiration in the treatment of spontaneous
pneumothorax. Thorax 1985; 40:66 67
Kennedy L, Sahn SA. Talc pleurodesis for the treatment of
pneumothorax and pleural effusion. Chest 1994; 106:12151222
Lesur O, Delorme N, Fromaget J, et al. Computed tomography
in the etiologic assessment of idiopathic spontaneous pneumothorax. Chest 1990; 98:341347
Liu H, Lin P, Hsieh M, et al. Thoracoscopic surgery as a routine
procedure for spontaneous pneumothorax: results from 82 patients. Chest 1995; 107:559 562
Luketich J, Kiss M, Hershey J, et al. Chest tube insertion: a
prospective evaluation of pain management. Clin J Pain 1998;
14:152154
Massard G, Thomas P, Wihlm J. Minimally invasive management
for first and recurrent pneumothorax. Ann Thorac Surg 1998;
66:592599
Mathur R, Cullen J, Kinnear WJM, et al. Time course of
resolution of persistent air leak in spontaneous pneumothorax.
Respir Med 1995; 89:129 132
Mercier C, Page A, Verdant A, et al. Outpatient management of
intercostal tube drainage in spontaneous pneumothorax. Ann
Thorac Surg 1976; 22:163165
Milanez J, Vargas F, Filomeno L, et al. Intrapleural talc for the
prevention of recurrent pneumothorax. Chest 1994:11621165
Minami H, Saka H, Senda K, et al. Small caliber catheter
drainage for spontaneous pneumothorax. Am J Med Sci 1992;
304:345347
Mitlehner W, Friedrich M, Dissmann W. Value of computed

tomography in the detection of bullae and blebs in patients with

primary spontaneous pneumothorax. Respiration 1992; 59:221
227
Naunheim K, Mack M, Hazelrigg S, et al. Safety and efficacy of
video-assisted thoracic surgical techniques for the treatment of
spontaneous pneumothorax. J Thorac Cardiovasc Surg 1995;
109:1198 1203
Ng AW, Chan KW, Lee SK. Simple aspiration of pneumothorax.
Singapore Med J 1994; 35:50 52
Nickoladze GD. Treatment of spontaneous pneumothorax with
Heimlich flutter valve. J Thorac Cardiovasc Surg 1990; 99:757
758
Noppen M, Meysman M, DHaese J, et al. Comparison of
video-assisted thoracoscopic talcage for recurrent primary versus
persistent secondary spontaneous pneumothorax. Eur Respir J
1997; 10:412 416
Schoenenberger RA, Haefeli WE, Weiss P, et al. Timing of
invasive procedures in therapy for primary and secondary spontaneous pneumothoraces. Arch Surg 1991; 126:764 766
Schoenenberger R, Haefeli W, Weiss P, et al. Evaluation of
conventional chest tube therapy for iatrogenic pneumothorax.
Chest 1993; 104:1770 1772
Schramel F, Sutedja T, Braber J, et al. Cost-effectiveness of
video-assisted thoracoscopic surgery versus conservative treatment for first time or recurrent spontaneous pneumothorax. Eur
Respir J 1996; 9:18211885
Seaton D, Yoganathan K, Coady T, et al. Spontaneous pneumothorax: marker gas technique for predicting outcome of manual
aspiration. BMJ 1991; 302:262265
Seremetis M. The management of spontaneous pneumothorax.
Chest 1970; 57:65 68
Sharma T, Agnihotri S, Jain N, et al. Intercostal tube thoracostomy in pneumothorax. Factors influencing re-expansion of lung.
Indian J Chest Dis Allied Sci 1988; 30:3235
So S, Yu D. Catheter drainage of spontaneous pneumothorax:
suction or no suction, early or late removal? Thorax 1982;
37:46 48
Stradling P, Poole G. Conservative management of spontaneous
pneumothorax. Thorax 1966; 21:145149
Van de Brekel J, Duurkens V, Vanderschueren R. Pneumothorax:
results of thoracoscopy and pleurodesis with talc poudrage and
thoracotomy. Chest 1993; 103:345347
Voge V, Anthracite R. Spontaneous pneumothorax in the USAF
aircrew population: a retrospective study. Aviat Space Environ
Med 1986; 57:939 949
Wait MA, Estrera A. Changing clinical spectrum of spontaneous
pneumothorax. Am J Surg 1992; 164:528 531
Waller DA, Forty J, Morritt GN. Video-assisted thoracoscopic
surgery versus thoracotomy for spontaneous pneumothorax. Ann
Thorac Surg 1994; 58:372377
Waller D, Forty J, Soni A, et al. Videothoracoscopic operation for
secondary spontaneous pneumothorax. Ann Thorac Surg 1994;
57:16121615
Warner B, Bailey W, Shipley T. Value of computed tomography
of the lung in the management of primary spontaneous pneumothorax. Am J Surg 1991; 162:39 42
Weeden D, Smith G. Surgical experience in the management of
spontaneous pneumothorax. Thorax 1983; 38:737743
Yim A, Ho J, Lai C, et al. Primary spontaneous pneumothorax
treated by video assisted thoracoscopic surgery-results of intermediate follow up. Aust N Z J Med 1995; 25:146 150
Yim A. Video-assisted thoracoscopic suturing of apical bullae: an
alternative to staple resection in the management of primary
spontaneous pneumothorax. Surg Endosc 1995; 9:10131016
Yim A, Liu H. Video assisted thoracoscopic management of
primary spontaneous pneumothorax. Surg Laparosc Endosc
1997; 7:236 240
CHEST / 119 / 2 / FEBRUARY, 2001

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians

601

References
1 Light RW. Management of spontaneous pneumothorax. Am
Rev Respir Dis 1993; 148:245248
2 Melton LJ, Hepper NGG, Offord KP. Incidence of spontaneous pneumothorax in Olmsted County, Minnesota: 1950
1974; Am Rev Respir Dis 1979 1974:120:1379 1382
3 Baumann MH, Strange C. Treatment of spontaneous pneumothorax: a more aggressive approach? Chest 1997; 112:789
804
4 Baumann MH, Strange C. The clinicians perspective on
pneumothorax management. Chest 1997; 112:822 828
5 Dalkey NC, Helmer O. An experimental application of the
Delphi method to the use of experts. Publication RM-727PR. Santa Monica, CA: Rand Corp, 1962
6 Delbecq AL, Van de Ven AH, Gustafson DH. Group techniques for program planning. Glenview, IL: Scott Foresman
Co., 1975; 83107
7 Jones J, Hunter D. Consensus methods for medical and
health services research. BMJ 1995; 311:376 380
8 Taylor RW. Pulmonary artery catheter consensus conference:
consensus statement. New Horizons 1997; 5:175194
9 Heffner JE, Aitken M, Geist L, et al. Attributes of ATS
documents that guide clinical practice. Am J Respir Crit Care
Med 1997; 156:20152025
10 Brooks RH, Chassin M, Fink A, et al. A method for the
detailed assessment of the appropriateness of medical technologies: a Rand note. Publication N-3376- HHS. Santa
Monica, CA: Rand Corp, 1991
11 Kahn DA, Docherty JP, Carpenter D, et al. Consensus
methods in practice guideline development: a review and
description of a new method. Psychopharmacol Bull 1997;
33:631 639
12 Waller DA, Forty J, Morritt GN. Video-assisted thoracoscopic
surgery versus thoracotomy for spontaneous pneumothorax.
Ann Thorac Surg 1994; 58:372376
13 van den Brande P, Staelens I. Chemical pleurodesis in
primary spontaneous pneumothorax. Thorac Cardiovasc Surg
1989; 37:180 182
14 Ma Y, Li J, Liu Y. Short wave diathermy for small spontaneous pneumothorax. Thorax 1997; 52:561566
15 Light RW, OHara VS, Moritz TE, et al. Intrapleural tetracycline for the prevention of recurrent spontaneous pneumothorax. JAMA 1990; 264:2224 2230
16 Engdahl O, Boe J, Sandstedt S. Interpleural bupivacaine for
analgesia during chest drainage treatment for pneumothorax:
a randomized double-blind study. Acta Anaesthesiol Scand
1993; 37:149 153

17 Andrivet P, Djedaini K, Teboul J-L, et al. Spontaneous

pneumothorax: comparison of thoracic drainage vs immediate
or delayed needle aspiration. Chest 1995; 108:335340
18 Almind M, Lange P, Viskum K. Spontaneous pneumothorax:
comparison of simple drainage, talc pleurodesis, and tetracycline pleurodesis. Thorax 1989; 44:627 630
19 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 1339
20 Miller AC, Harvey JE. Guidelines for the management of
spontaneous pneumothorax: Standards of Care Committee,
British Thoracic Society. BMJ 1993; 307:114 116
21 Kim KH, Kim HK, Han JY, et al. Transaxillary minithoracotomy versus video-assisted thoracic surgery for spontaneous
pneumothorax. Ann Thorac Surg 1996; 61:1510 1512
22 ORourke JP, Yee ES. Civilian spontaneous pneumothorax:
treatment options and long-term results. Chest 1989; 96:
13021306
23 Videm V, Pillgram-Larsen J, Ellingsen O, et al. Spontaneous
pneumothorax in chronic obstructive pulmonary disease:
complications, treatment and recurrences. Eur J Respir Dis
1987; 71:365371
24 Shields TW, Oilschlager GA. Spontaneous pneumothorax in
patients 40 years of age and older. Ann Thorac Surg 1966;
2:377383
25 George RB, Herbert SJ, Shames JM, et al. Pneumothorax
complicating pulmonary emphysema. JAMA 1975; 234:389
393
26 Dines DE, Clagett OT, Payne WS. Spontaneous pneumothorax in emphysema. Mayo Clin Proc 1970; 45:481 487
27 Waller DA, McConnell SA, Rajesh PB. Delayed referral
reduces the success of video-assisted thoracoscopic surgery
for spontaneous pneumothorax. Respir Med 1998; 92:246
249
28 Schramel FM, Sutedja TG, Braber JC, et al. Cost-effectiveness of video-assisted thoracoscopic surgery versus conservative treatment for first time or recurrent spontaneous pneumothorax. Eur Respir J 1996; 9:18211825
29 Naylor CD, Guyatt GH. Users guide to the medical literature: XI. How to use an article about a clinical utilization
review. JAMA 1996; 275:14351439
30 Jackson R, Feder G. Guidelines for clinical guidelines: a
simple, pragmatic strategy for guideline development. BMJ
1998; 317:427 428
31 Naylor CD. What is appropriate care? N Engl J Med 1998;
338:1918 1920

602

Consensus Conference

Downloaded from chestjournal.org on September 18, 2007

Copyright 2001 by American College of Chest Physicians