Treatment of primary spontaneous pneumothorax in Switzerland: results of a survey
http://www.100md.com
《血管的通路杂志》
Division of General Thoracic Surgery, University Hospital Berne, CH – 3010 Berne, Switzerland
Abstract
Few trials to guide clinical management of primary spontaneous pneumothorax (PSP) exist. This study aims to reveal current practice in the management of PSP in Switzerland, to define the level of consensus and to provide evidence in guiding clinical practice. Questionnaires were sent to 355 departments of internal medicine and surgery in Switzerland and 114 (32.1%) were available for analysis. Recommendations based on the highest consensus are extracted. Good to very good consensus is reached in 63% of all management options. There is very good consensus for the management of clinically stable or unstable patients with small or large PSP, for the treatment of patients without chest tubes and the operative management of persistent air leaks and recurrence prevention. There is good consensus concerning the role of suction, the size of chest tubes, and the use of CT-scans. However, there is little consensus concerning chest tube removal and chemical pleurodesis. Good and very good consensus exists for most management options in the treatment of PSP in Switzerland. The given recommendations can be used as evidence in guiding clinical practice in circumstances where no evidence of higher levels exists.
Key Words: Evidence-based medicine; Management guidelines; Primary spontaneous pneumothorax; Questionnaire
1. Introduction
Primary spontaneous pneumothorax (PSP) is defined as a pneumothorax that occurs without a precipitating event in a person with no clinically apparent lung disease. Therapy of PSP has two objectives, first to provide re-expansion of the lung and second to decrease the likelihood of recurrence. To achieve this, wide variation in treatment and management exists [1,2]. A literature review yielded only eight randomised controlled trials (RCTs), no meta-analyses, and only two practice guidelines [3,4].
We chose an evidence-based approach to determine actual clinical practice in the management of PSP and to provide a basis for further research. A nationwide survey was undertaken using a questionnaire.
The aim of this survey was to reveal current practice in Switzerland, to define the level of consensus in the management of PSP and to provide evidence in guiding clinical practice.
2. Materials and methods
2.1. Development of the questionnaire
The questionnaire developed by our group used similar questions for management options and the same description for the level of consensus (Table 2) and for the appropriateness of management options (Table 3) as defined in the American College of Chest Physicians Delphi Consensus Statement [3].
The participants were asked to respond using a 9-point Likert scale (Table 1). A few questions requested a numerical answer.
The questionnaire addressed the topics of clinically stable patients with a small or large pneumothorax, clinically unstable patients with a large pneumothorax, management recommendations for observation of patients without chest tube, the role of suction, chest tube size, chest tube removal, persistent air leak, recurrence prevention, chemical pleurodesis, operative procedures, pleurectomy vs. pleurabrasio and computed tomography in the management of PSP.
The following clinical definitions were given:
PSP: No antecedent traumatic or iatrogenic cause, no clinically apparent underlying lung abnormalities or underlying conditions. Clinical stable patient: respiratory rate 24 breaths/min, heart rate >60 beats/min or <120 beats/min, normal blood pressure. Room air O2 saturation 90%, patient can speak in whole sentences between breaths. Unstable patient: any patient not fulfilling the definition of stable. Small pneumothorax: <3 cm apex-to-cupola distance. Large pneumothorax: >3 cm apex-to-cupola distance.
2.2. Administration of the questionnaire
The questionnaire was sent to all medical and surgical departments that provide postgraduate training in Switzerland in January 2002 (355 questionnaires to 151 surgical and 204 medical departments).
2.3. Statistical analysis
For each question answered according to the 9-point Likert scale the median, 25%- and 75%- quartile were calculated. According to the definitions (Tables 2, 3) the level of consensus, management options and status of recommendation were given.
3. Results
Of 355 questionnaires mailed to 151 surgical and 204 medical departments, 114 (32%) questionnaires were finally available for analysis.
Results are given as recommendations based on treatment options with the highest consensus.
3.1. Clinically stable patients with a small pneumothorax, first episode (Table 4)
Clinically stable patients with a small pneumothorax (less than 3 cm apex-to-cupola distance) should be observed in an emergency department for 3 to 6 h and discharged home with close follow up if repeat X-ray shows a stable pneumothorax or excludes progression (good consensus). Patients may also be admitted for observation (some consensus). Acceptable management in only rare circumstances is the use of needle aspiration with discharge of the patient (good consensus) (Table 4).
3.2. Clinically stable patients with a large pneumothorax, first episode (Table 4)
Clinically stable patients with a large pneumothorax (more than 3 cm apex-to-cupola distance) should be hospitalised and have a chest tube inserted (very good consensus). They should not be treated as an outpatient, even with close follow up (very good consensus). Acceptable in only rare circumstances is both the admission for observation or needle aspiration with discharge (some consensus).
3.3. Clinically unstable patients with a large pneumothorax, first episode (Table 4)
Clinically unstable patients with a large pneumothorax should be hospitalised and have a chest tube inserted (very good consensus). It is inappropriate to observe and discharge these patients (perfect consensus), to admit for observation only (very good consensus) or to use needle aspiration (good consensus).
3.4. Recommendations for observation of patients without chest tubes (Table 5)
Patients treated with observation only may be observed for 3 to 6 h and have a second chest X-ray before discharge (good consensus). They should have a follow up with repeat chest X-ray within 12 h to 2 days after discharge from the emergency department (very good consensus) (Table 5).
3.5. Chest tube size (Table 5)
The insertion of a medium sized chest tube of size 16 to 22 charriere is acceptable (good consensus). A large bore chest tube (size 30 to 36) is inappropriate (very good consensus) and there is only some consensus in using a Mathys drain (size <16) and not using size 24 to 28 chest tubes.
3.6. Role of suction (Table 5)
A patient without suspicion of a large air leak and who is not intubated should be put on suction with an underwater seal (very good consensus).
Because of wide variations no recommendations can be given for the options of water seal without suction (some consensus) or for the use of the Heimlich valve (some consensus).
3.7. Chest tube removal (Table 5)
The precise method of chest tube removal is controversial. However, 80.9% of all physicians agree that chest tubes should be removed in a staged manner with an X-ray before removal to ensure that the air leak has resolved. Whether clamping or continuation without suction is used, there is consensus to wait at least 12 h before ordering the X-ray.
3.8. Persistent air leak (Table 6)
For the treatment of persistent air leaks the patient should be operated (very good consensus). Of all physicians, 22.4% wait for two days before performing any intervention to close a persistent air leak. 23.30% wait for three days, 15.5% wait for 4 days, and 18.9% wait for 5 days. 3.5% of all physicians would wait only one day whereas the remaining 16.4% would wait longer than 5 days (Table 6).
Bronchoscopically directed attempts to seal the air leak are inappropriate (good consensus). A second chest tube is rarely acceptable (very good consensus) and chemical pleurodesis should not be used (good consensus).
3.9. Recurrence prevention (Table 7)
Of all physicians, 82% recommend a treatment for recurrence prevention after the second PSP (the first recurrence). 1.6% of all physicians recommend an intervention after the first PSP, 14% after the third PSP, and 2.4% after the fourth PSP.
3.10. Chemical pleurodesis (Table 7)
Because of wide treatment variation no recommendation can be given for the option of performing a chemical pleurodesis through the chest tube. There was also no consensus on which chemical agent may be used (Table 7).
3.11. Operative procedure (Table 7)
An operative procedure is the preferred method for recurrence prevention (very good consensus) (Table 8). The preferred method is the video assisted thoracoscopic surgery (VATS) (very good consensus). Both the median sternotomy and the (antero)lateral thoracotomy are inappropriate (very good consensus and good consensus). The axillary approach is rarely acceptable (some consensus).
3.12. The method of bleb or bulla resection (Table 8)
Blebs or bullae should be resected with a stapler (very good consensus). Blebs or bullae should never be left if found during surgery (very good consensus) (Table 8).
Hand sewing, laser ablation or electro-coagulation yielded no consensus. There was also no consensus on performing an apex resection routinely even if no blebs or bullae are found with almost as many surgeons preferring as opposing this option.
3.13. Performing pleurodesis (Table 8)
A parietal pleurectomy or pleurabrasio should be used to achieve pleurodesis (good consensus). The pleurectomy or pleurabrasio should be done from the apex to the sixth rib (good consensus).
No recommendation can be given for the intraoperative use of talc poudrage (no consensus). Other chemical agents should not be used (very good consensus).
3.14. Computed Tomography in the management of PSP
A CT-scan should be done for planning an operative procedure and in cases of persisting air leaks (good consensus, median 8). A CT-scan should not be done in the first episode of PSP (good consensus, median 1).
No recommendation can be given for the use of the CT-scan in patients with recurrent pneumothorax when no operation is intended.
4. Discussion
Practice variation is common in the management of PSP. In the present study several management options for specific clinical situations were presented. In almost all of these clinical situations every single option was chosen by some clinicians as the preferred treatment option. However, for most clinical situations a good or very good consensus was reached for a specific treatment option.
For 60 management options the consensus was perfect in one, very good in 16 and good in 21. Almost two thirds (63%) of all options reached a consensus that was good or even better. This is in contradiction to other questionnaires [1,4] and may be a consequence of published guidelines [3,5]. It may be due to the fact that with the use of video assisted thoracoscopic surgery (VATS) a less invasive treatment method has become available [6].
Several open questions in the management of the PSP are difficult to test in clinical trials because patient preferences play a major role. Simple endpoints like death are not an issue and recurrence rates after VATS are in the range of 5% at 5 years.
The present study does not represent substitute for evidence of higher levels. Evaluating the question of aspiration versus chest tube insertion, a recent RCT [7] has shown that a former treatment option with a high level of consensus for being regarded as inappropriate [3] was indeed shown to be highly effective in a subgroup of patients.
What are the limitations of the study At first the return-rate was only 35%. However, the return-rate for all surgical departments was 50% and it is likely that more departments with a special interest in that issue were responding. A second limitation is the fact that only PSP was considered and the results cannot be applied to the secondary spontaneous pneumothorax (SSP). On the other hand, this limitation is also its strength because the SSP represents a different group of patient with more comorbidities, higher mortality of the SSP itself and of interventions. The results of several clinical trials and even RCTs may not be applicable because patients with both PSP and SSP were included [8].
To provide more evidence for the treatment of PSP two other approaches have been tried. The ACCP conducted a consensus conference according to the Delphi technique [3], in which evidence was created by reiterating rounds of expert questioning to create various standard treatment options. Proposals of these rounds were afterwards submitted to a larger circle of experts to achieve a consensus.
A second approach was undertaken by the British Thoracic Society. A literature research with critical appraisal was done for the available literature and for the management of PSP and was presented with comments on conflicting aspects [5].
What is the significance of the study Speaking in terms of evidence-based medicine the present study represents evidence of level V. This is the lowest form of evidence but still has to be regarded higher than personal experience or anecdotal reports.
The clinician should be aware of clinical trials or randomised controlled trials addressing questions in the management of PSP. But, up to the present time there is lack of evidence in many aspects. In these specific situations, the clinician can use the presented results as a guide in the management of PSP.
5. Conclusions
This survey demonstrates management variation in the treatment of PSP, but also good to very good consensus for the majority of the questions. Only few practice habits with good to very good consensus are in contradiction to the literature. Although in cases where no literature exists the consensus can be used as a guide for treatment, but it has to be regarded as the weakest form of evidence. More RCTs are needed to resolve some of the described problems in order to provide the form of evidence, which is needed for the development of guidelines for the treatment of PSP. In addition, this survey shows some controversial issues with contradictory practice habits, which can be used to justify and plan randomised trials.
Acknowledgements
The manuscript was presented at the Swiss Society of Surgery (Schweizerische Gesellschaft für Chirurgie, SGC), Davos, Switzerland, 23–26 June 2004.
References
Baumann MH, Strange C. The clinician's perspective on pneumothorax management. Chest 1997; 112:822–828.
Sahn SA, Heffner JE. Spontaneous pneumothorax. N Engl J Med 2000; 342:868–874.
Baumann MH, Strange C, Heffner JE, Light R, Kirby TJ, Klein J, Luketich JD, Panacek EA, Sahn SA. Management of spontaneous pneumothorax: an American College of Chest Physicians Delphi consensus statement. Chest 2001; 119:590–602.
Miller AC, Harvey JE. Guidelines for the management of spontaneous pneumothorax. Standards of Care Committee, British Thoracic Society. Br Med J 1993; 307:6896114–116.
Henry M, Arnold T, Harvey J. BTS guidelines for the management of spontaneous pneumothorax. Thorax 2003; 58:Suppl_2ii39–ii52.
Inderbitzi RG, Leiser A, Furrer M, Althaus U. Three years' experience in video-assisted thoracic surgery (VATS) for spontaneous pneumothorax. J Thorac Cardiovasc Surg 1994; 107:1410–1415.
Noppen M, Alexander P, Driesen P, Slabbynck H, Verstraeten A. Manual aspiration versus chest tube drainage in first episodes of primary spontaneous pneumothorax: a multicenter, prospective, randomized pilot study. Am J Respir Crit Care Med 2002; 165:1240–1244.
Waller DA, Forty J, Morritt GN. Video-assisted thoracoscopic surgery versus thoracotomy for spontaneous pneumothorax. Ann Thorac Surg 1994; 58:372–376.(Jan R. Kuester, Steffen F)
Abstract
Few trials to guide clinical management of primary spontaneous pneumothorax (PSP) exist. This study aims to reveal current practice in the management of PSP in Switzerland, to define the level of consensus and to provide evidence in guiding clinical practice. Questionnaires were sent to 355 departments of internal medicine and surgery in Switzerland and 114 (32.1%) were available for analysis. Recommendations based on the highest consensus are extracted. Good to very good consensus is reached in 63% of all management options. There is very good consensus for the management of clinically stable or unstable patients with small or large PSP, for the treatment of patients without chest tubes and the operative management of persistent air leaks and recurrence prevention. There is good consensus concerning the role of suction, the size of chest tubes, and the use of CT-scans. However, there is little consensus concerning chest tube removal and chemical pleurodesis. Good and very good consensus exists for most management options in the treatment of PSP in Switzerland. The given recommendations can be used as evidence in guiding clinical practice in circumstances where no evidence of higher levels exists.
Key Words: Evidence-based medicine; Management guidelines; Primary spontaneous pneumothorax; Questionnaire
1. Introduction
Primary spontaneous pneumothorax (PSP) is defined as a pneumothorax that occurs without a precipitating event in a person with no clinically apparent lung disease. Therapy of PSP has two objectives, first to provide re-expansion of the lung and second to decrease the likelihood of recurrence. To achieve this, wide variation in treatment and management exists [1,2]. A literature review yielded only eight randomised controlled trials (RCTs), no meta-analyses, and only two practice guidelines [3,4].
We chose an evidence-based approach to determine actual clinical practice in the management of PSP and to provide a basis for further research. A nationwide survey was undertaken using a questionnaire.
The aim of this survey was to reveal current practice in Switzerland, to define the level of consensus in the management of PSP and to provide evidence in guiding clinical practice.
2. Materials and methods
2.1. Development of the questionnaire
The questionnaire developed by our group used similar questions for management options and the same description for the level of consensus (Table 2) and for the appropriateness of management options (Table 3) as defined in the American College of Chest Physicians Delphi Consensus Statement [3].
The participants were asked to respond using a 9-point Likert scale (Table 1). A few questions requested a numerical answer.
The questionnaire addressed the topics of clinically stable patients with a small or large pneumothorax, clinically unstable patients with a large pneumothorax, management recommendations for observation of patients without chest tube, the role of suction, chest tube size, chest tube removal, persistent air leak, recurrence prevention, chemical pleurodesis, operative procedures, pleurectomy vs. pleurabrasio and computed tomography in the management of PSP.
The following clinical definitions were given:
PSP: No antecedent traumatic or iatrogenic cause, no clinically apparent underlying lung abnormalities or underlying conditions. Clinical stable patient: respiratory rate 24 breaths/min, heart rate >60 beats/min or <120 beats/min, normal blood pressure. Room air O2 saturation 90%, patient can speak in whole sentences between breaths. Unstable patient: any patient not fulfilling the definition of stable. Small pneumothorax: <3 cm apex-to-cupola distance. Large pneumothorax: >3 cm apex-to-cupola distance.
2.2. Administration of the questionnaire
The questionnaire was sent to all medical and surgical departments that provide postgraduate training in Switzerland in January 2002 (355 questionnaires to 151 surgical and 204 medical departments).
2.3. Statistical analysis
For each question answered according to the 9-point Likert scale the median, 25%- and 75%- quartile were calculated. According to the definitions (Tables 2, 3) the level of consensus, management options and status of recommendation were given.
3. Results
Of 355 questionnaires mailed to 151 surgical and 204 medical departments, 114 (32%) questionnaires were finally available for analysis.
Results are given as recommendations based on treatment options with the highest consensus.
3.1. Clinically stable patients with a small pneumothorax, first episode (Table 4)
Clinically stable patients with a small pneumothorax (less than 3 cm apex-to-cupola distance) should be observed in an emergency department for 3 to 6 h and discharged home with close follow up if repeat X-ray shows a stable pneumothorax or excludes progression (good consensus). Patients may also be admitted for observation (some consensus). Acceptable management in only rare circumstances is the use of needle aspiration with discharge of the patient (good consensus) (Table 4).
3.2. Clinically stable patients with a large pneumothorax, first episode (Table 4)
Clinically stable patients with a large pneumothorax (more than 3 cm apex-to-cupola distance) should be hospitalised and have a chest tube inserted (very good consensus). They should not be treated as an outpatient, even with close follow up (very good consensus). Acceptable in only rare circumstances is both the admission for observation or needle aspiration with discharge (some consensus).
3.3. Clinically unstable patients with a large pneumothorax, first episode (Table 4)
Clinically unstable patients with a large pneumothorax should be hospitalised and have a chest tube inserted (very good consensus). It is inappropriate to observe and discharge these patients (perfect consensus), to admit for observation only (very good consensus) or to use needle aspiration (good consensus).
3.4. Recommendations for observation of patients without chest tubes (Table 5)
Patients treated with observation only may be observed for 3 to 6 h and have a second chest X-ray before discharge (good consensus). They should have a follow up with repeat chest X-ray within 12 h to 2 days after discharge from the emergency department (very good consensus) (Table 5).
3.5. Chest tube size (Table 5)
The insertion of a medium sized chest tube of size 16 to 22 charriere is acceptable (good consensus). A large bore chest tube (size 30 to 36) is inappropriate (very good consensus) and there is only some consensus in using a Mathys drain (size <16) and not using size 24 to 28 chest tubes.
3.6. Role of suction (Table 5)
A patient without suspicion of a large air leak and who is not intubated should be put on suction with an underwater seal (very good consensus).
Because of wide variations no recommendations can be given for the options of water seal without suction (some consensus) or for the use of the Heimlich valve (some consensus).
3.7. Chest tube removal (Table 5)
The precise method of chest tube removal is controversial. However, 80.9% of all physicians agree that chest tubes should be removed in a staged manner with an X-ray before removal to ensure that the air leak has resolved. Whether clamping or continuation without suction is used, there is consensus to wait at least 12 h before ordering the X-ray.
3.8. Persistent air leak (Table 6)
For the treatment of persistent air leaks the patient should be operated (very good consensus). Of all physicians, 22.4% wait for two days before performing any intervention to close a persistent air leak. 23.30% wait for three days, 15.5% wait for 4 days, and 18.9% wait for 5 days. 3.5% of all physicians would wait only one day whereas the remaining 16.4% would wait longer than 5 days (Table 6).
Bronchoscopically directed attempts to seal the air leak are inappropriate (good consensus). A second chest tube is rarely acceptable (very good consensus) and chemical pleurodesis should not be used (good consensus).
3.9. Recurrence prevention (Table 7)
Of all physicians, 82% recommend a treatment for recurrence prevention after the second PSP (the first recurrence). 1.6% of all physicians recommend an intervention after the first PSP, 14% after the third PSP, and 2.4% after the fourth PSP.
3.10. Chemical pleurodesis (Table 7)
Because of wide treatment variation no recommendation can be given for the option of performing a chemical pleurodesis through the chest tube. There was also no consensus on which chemical agent may be used (Table 7).
3.11. Operative procedure (Table 7)
An operative procedure is the preferred method for recurrence prevention (very good consensus) (Table 8). The preferred method is the video assisted thoracoscopic surgery (VATS) (very good consensus). Both the median sternotomy and the (antero)lateral thoracotomy are inappropriate (very good consensus and good consensus). The axillary approach is rarely acceptable (some consensus).
3.12. The method of bleb or bulla resection (Table 8)
Blebs or bullae should be resected with a stapler (very good consensus). Blebs or bullae should never be left if found during surgery (very good consensus) (Table 8).
Hand sewing, laser ablation or electro-coagulation yielded no consensus. There was also no consensus on performing an apex resection routinely even if no blebs or bullae are found with almost as many surgeons preferring as opposing this option.
3.13. Performing pleurodesis (Table 8)
A parietal pleurectomy or pleurabrasio should be used to achieve pleurodesis (good consensus). The pleurectomy or pleurabrasio should be done from the apex to the sixth rib (good consensus).
No recommendation can be given for the intraoperative use of talc poudrage (no consensus). Other chemical agents should not be used (very good consensus).
3.14. Computed Tomography in the management of PSP
A CT-scan should be done for planning an operative procedure and in cases of persisting air leaks (good consensus, median 8). A CT-scan should not be done in the first episode of PSP (good consensus, median 1).
No recommendation can be given for the use of the CT-scan in patients with recurrent pneumothorax when no operation is intended.
4. Discussion
Practice variation is common in the management of PSP. In the present study several management options for specific clinical situations were presented. In almost all of these clinical situations every single option was chosen by some clinicians as the preferred treatment option. However, for most clinical situations a good or very good consensus was reached for a specific treatment option.
For 60 management options the consensus was perfect in one, very good in 16 and good in 21. Almost two thirds (63%) of all options reached a consensus that was good or even better. This is in contradiction to other questionnaires [1,4] and may be a consequence of published guidelines [3,5]. It may be due to the fact that with the use of video assisted thoracoscopic surgery (VATS) a less invasive treatment method has become available [6].
Several open questions in the management of the PSP are difficult to test in clinical trials because patient preferences play a major role. Simple endpoints like death are not an issue and recurrence rates after VATS are in the range of 5% at 5 years.
The present study does not represent substitute for evidence of higher levels. Evaluating the question of aspiration versus chest tube insertion, a recent RCT [7] has shown that a former treatment option with a high level of consensus for being regarded as inappropriate [3] was indeed shown to be highly effective in a subgroup of patients.
What are the limitations of the study At first the return-rate was only 35%. However, the return-rate for all surgical departments was 50% and it is likely that more departments with a special interest in that issue were responding. A second limitation is the fact that only PSP was considered and the results cannot be applied to the secondary spontaneous pneumothorax (SSP). On the other hand, this limitation is also its strength because the SSP represents a different group of patient with more comorbidities, higher mortality of the SSP itself and of interventions. The results of several clinical trials and even RCTs may not be applicable because patients with both PSP and SSP were included [8].
To provide more evidence for the treatment of PSP two other approaches have been tried. The ACCP conducted a consensus conference according to the Delphi technique [3], in which evidence was created by reiterating rounds of expert questioning to create various standard treatment options. Proposals of these rounds were afterwards submitted to a larger circle of experts to achieve a consensus.
A second approach was undertaken by the British Thoracic Society. A literature research with critical appraisal was done for the available literature and for the management of PSP and was presented with comments on conflicting aspects [5].
What is the significance of the study Speaking in terms of evidence-based medicine the present study represents evidence of level V. This is the lowest form of evidence but still has to be regarded higher than personal experience or anecdotal reports.
The clinician should be aware of clinical trials or randomised controlled trials addressing questions in the management of PSP. But, up to the present time there is lack of evidence in many aspects. In these specific situations, the clinician can use the presented results as a guide in the management of PSP.
5. Conclusions
This survey demonstrates management variation in the treatment of PSP, but also good to very good consensus for the majority of the questions. Only few practice habits with good to very good consensus are in contradiction to the literature. Although in cases where no literature exists the consensus can be used as a guide for treatment, but it has to be regarded as the weakest form of evidence. More RCTs are needed to resolve some of the described problems in order to provide the form of evidence, which is needed for the development of guidelines for the treatment of PSP. In addition, this survey shows some controversial issues with contradictory practice habits, which can be used to justify and plan randomised trials.
Acknowledgements
The manuscript was presented at the Swiss Society of Surgery (Schweizerische Gesellschaft für Chirurgie, SGC), Davos, Switzerland, 23–26 June 2004.
References
Baumann MH, Strange C. The clinician's perspective on pneumothorax management. Chest 1997; 112:822–828.
Sahn SA, Heffner JE. Spontaneous pneumothorax. N Engl J Med 2000; 342:868–874.
Baumann MH, Strange C, Heffner JE, Light R, Kirby TJ, Klein J, Luketich JD, Panacek EA, Sahn SA. Management of spontaneous pneumothorax: an American College of Chest Physicians Delphi consensus statement. Chest 2001; 119:590–602.
Miller AC, Harvey JE. Guidelines for the management of spontaneous pneumothorax. Standards of Care Committee, British Thoracic Society. Br Med J 1993; 307:6896114–116.
Henry M, Arnold T, Harvey J. BTS guidelines for the management of spontaneous pneumothorax. Thorax 2003; 58:Suppl_2ii39–ii52.
Inderbitzi RG, Leiser A, Furrer M, Althaus U. Three years' experience in video-assisted thoracic surgery (VATS) for spontaneous pneumothorax. J Thorac Cardiovasc Surg 1994; 107:1410–1415.
Noppen M, Alexander P, Driesen P, Slabbynck H, Verstraeten A. Manual aspiration versus chest tube drainage in first episodes of primary spontaneous pneumothorax: a multicenter, prospective, randomized pilot study. Am J Respir Crit Care Med 2002; 165:1240–1244.
Waller DA, Forty J, Morritt GN. Video-assisted thoracoscopic surgery versus thoracotomy for spontaneous pneumothorax. Ann Thorac Surg 1994; 58:372–376.(Jan R. Kuester, Steffen F)