Article, Pulmonology

Small-bore catheter versus chest tube drainage for pneumothorax

Unlabelled imageAmerican Journal of Emergency Medicine (2012) 30, 1407-1413

Original Contribution

Small-bore catheter versus chest tube drainage for pneumothorax?

Damien Contou MDa,?, Keyvan Razazi MDa, Sandrine Katsahian MD, PhDb, Bernard Maitre MD, PhDc,d,e, Armand Mekontso-Dessap MD, PhDa,d,

Christian Brun-Buisson MDa,d,e, Arnaud W. Thille MD, PhDa,d,e

aMedical Intensive Care Unit, AP-HP, Henri Mondor Hospital, Creteil, France bBiostatistics Unit, AP-HP, Henri Mondor Hospital, Creteil, France cPulmonology Unit, AP-HP, Henri Mondor Hospital, Creteil, France dINSERM, U955, Creteil, France

eUniversity of Paris-Est, Creteil, France

Received 7 August 2011; revised 10 October 2011; accepted 11 October 2011

Abstract

Study Objective: The aim of the study was to compare the effectiveness of drainage via a single-lumen (5F catheter) Central venous catheter to a conventional (14-20F catheter) chest tube (CT) for the management of pneumothoraces, including primary Spontaneous pneumothorax (PSP), secondary spontaneous pneumothorax (SSP), and traumatic and iatrogenic pneumothoraces.

Patients: All consecutive patients admitted to the intermediate intensive care unit of a university hospital for pneumothorax were retrospectively screened over an 8-year period. Patients were preferentially treated using CT from 2003 to 2007 and using CVC from 2008 to 2010. Drainage failure was defined as the need for a second drainage procedure or for surgery.

Results: Of 212 patients included, 117 (55%) had PSP, 28 (13%) had SSP associated with chronic

obstructive pulmonary disease, 19 (9%) had Traumatic pneumothorax, and 48 (23%) had iatrogenic pneumothorax. The failure rate was 23% in PSP, 36% in SSP, 16% in traumatic pneumothorax, and only 2% in iatrogenic pneumothorax. After adjustment, iatrogenic pneumothorax was the only factor that had an influence on drainage failure. The failure rate was similar between the 112 patients treated using CVC and the 100 patients treated using CT (18% vs 21%, P = .60). However, the durations of drainage (3.3 +- 1.9 vs 4.6 +- 2.6 days, P b .01) and of hospital stay were significantly shorter in patients treated using CVC as compared with CT.

Conclusion: Our findings suggest that drainage via a catheter or via a CT is similarly effective in the management of pneumothorax. We recommend considering drainage via a small-bore catheter as a first- line treatment in patients with pneumothorax, whatever its cause.

(C) 2012

? The study was conducted in the at the medical intensive care unit of Henri Mondor Hospital, Creteil, France.

* Corresponding author. Reanimation Medicale, Hopital Henri Mondor,

94010 Creteil, France. Tel.: +33 149 812 344; fax: +33 149 814 943.

E-mail address: [email protected] (D. Contou).

Introduction

Pneumothorax can be spontaneous, traumatic, or iatro- genic after the insertion of an intravascular device. Spontaneous pneumothorax can be stratified into primary

0735-6757/$ – see front matter (C) 2012 doi:10.1016/j.ajem.2011.10.014

spontaneous pneumothorax (PSP), occurring in the absence of underlying lung disease, and secondary spontaneous pneumothorax (SSP), occurring in patients having underly- ing chronic lung disease, most commonly chronic obstruc- tive pulmonary disease (COPD).

The initial management for a large and symptomatic pneumothorax is to remove the air from the pleural space, but recommended methods differ across guidelines [1,2]. According to the 2001 American College of Chest Physicians (ACCP) consensus conference [1], a large PSP should be managed with a small-bore pleural catheter (<=14F catheter) or a chest tube (CT) (16-22F catheter). In contrast, the 2010 British Thoracic Society (BTS) guideline [2] recommends that Needle aspiration should be the first-line procedure, whereas a small-bore catheter (<=14F catheter) should be inserted only after the Failure of needle aspiration [2]. Indeed, there is now compelling evidence that needle aspiration should be considered the first-line approach in the manage- ment of spontaneous pneumothorax [3-7]. Despite these recommendations, hospital admissions and insertions of large CTs remain the standard approaches in many institutions.

For SSP, the BTS recommends the insertion of a small- bore pleural catheter (<=14F catheter) [2], whereas the ACCP consensus conference recommended a larger CT (16-22F catheter) [1]. These variations in guidelines and recommen- dations may explain the wide variations in clinical practice. Indeed, a recent survey of intensivists showed that pneumothorax management was particularly heterogeneous; drainage via a large CT was still very common for the management of PSP, and only half of physicians used a small-bore catheter [8]. However, the evolution of guidelines from 2001 [1] to 2010 [2] has tended toward less invasive management with the use of smaller catheters, which are easier and faster to insert and entail less patient discomfort. We compared the effectiveness of a small-bore (5F catheter) central venous catheter (CVC) used as pleural catheter with a conventional large CT (14-20F catheter) in the management of all types of pneumothoraces, including spontaneous (PSP and SSP), traumatic, and iatrogenic

pneumothoraces.

Materials and methods

Patients and data collection

This study was approved by the institutional review board of the French Society for Respiratory Medicine (Societe de Pneumologie de Langue Francaise, SPLF). All consecutive patients admitted to the intermediate intensive care unit (ICU) of Henri Mondor Teaching Hospital for the manage- ment of pneumothorax from January 2003 to September 2010 were retrospectively analyzed. Patients were identified from the computerized ICU discharge reports, which are Microsoft Word files. Thanks to the “search” function, all

Word files in which the word “pneumothorax” occurred from 2003 to 2010 were identified. Only patients admitted for pneumothorax management were analyzed, and patients with a history of pneumothorax or with a pneumothorax occurring under mechanical ventilation were not analyzed. Once patients were identified, Dr Contou was responsible for data abstraction from these ICU discharge reports, and the data were directly reported using a Microsoft Excel database. All types of pneumothoraces were analyzed, including first episodes and recurrent PSP, SSPs due to COPD (SSP- COPD), and traumatic and iatrogenic pneumothoraces. We included only patients with confirmed radiologic diagnoses of a “large” pneumothorax requiring drainage, according to BTS guidelines [2]. We excluded small pneumothoraces managed conservatively with rest only, recurrent pneu- mothoraces contralateral to a prior pneumothorax with a decision of surgery upon admission, hemopneumothoraces,

and SSPs in non-COPD patients.

The presence of symptoms (chest pain and dyspnea) was determined based on the clinical history and examination results reported in the ICU discharge reports.

Drainage procedure

Whatever the cause of a pneumothorax, a conventional CT was preferentially used from 2003 to 2007, whereas a CVC was preferentially used from 2008 to 2010. Catheter drainage was undertaken with a small-bore single-lumen CVC (ARROW ES-04700, Reading, PA, USA) with a

diameter of 1.6 mm (5F catheter), inserted in the pleural space using the Seldinger technique. Chest tube drainage used a large-bore tube (REDAX Trocar catheter, Mirandola, Italy) with a diameter of 4.7 to 6.6 mm (14-20F catheter) inserted into the pleural space after incision and dissection of the intercostal tissues (Fig. 1).

Fig. 1 Illustration of a 5F catheter (14 gauge) single-lumen CVC (left) and conventional CTs of 16F catheter (middle) and 20F catheter (right) diameter.

PSP (n = 117)

SSP-COPD (n = 28)

Traumatic (n = 19)

Iatrogenic (n = 48)

Characteristics

Age, y

27 +- 7

58 +- 10 ?

36 +- 15 ?

62 +- 19 ?

Male sex, n (%)

100 (85)

25 (89)

17 (89)

30 (62) ?

Smoking status, n (%)

80 (68)

28 (100) ?

8 (42) ?

23 (48) ?

Dyspnea, n (%)

47 (40)

22 (79) ?

5 (26)

24 (50)

Chest pain, n (%)

114 (97)

19 (68) ?

18 (95)

9 (19) ?

SpO2 (%)

96 +- 3

90 +- 6 ?

96 +- 3

93 +- 6 ?

Days between onset and drainage

2.0 +- 2.5

3.1 +- 2.4 ?

1.6 +- 1.5

2.0 +- 1.9

Outcome

Duration of drainage, d

3.8 +- 2.0

5.3 +- 3.7 ?

3.6 +- 1.9

3.4 +- 1.8

Length of stay, d

4.7 +- 2.8

7.4 +- 5.1 ?

4.5 +- 2.2

4.3 +- 2.0

Primary drainage failure, n (%)

27 (23)

10 (36)

3 (16)

1 (2) ?

Surgery, n (%)

25 (21)

9 (32)

2 (10)

0 (0) ?

* P b .05 as compared with PSP.

Management protocol and definitions

Table 1 Comparison of patients’ characteristics and outcome according to the cause of pneumothorax

The follow-up and decision-making processes regarding drainage withdrawal and discharge were similar whatever the procedure. After insertion, a chest x-ray was performed to confirm the correct position of the CT/CVC and lung re- expansion. In cases of incomplete lung re-expansion, negative pressure (-20 cm H2O) was applied to the water- seal device until the next morning. Once complete lung re- expansion was obtained, the CT or CVC was clamped until the next morning. If lung expansion persisted, then the catheter or tube was withdrawn, and a final chest x-ray was performed to check for the absence of recurrence before patient discharge. In cases of persistent air leaks for more than 5 days or recurrent pneumothorax after clamping the CVC or CT, a surgical cure was performed, or a CT was inserted as a second-line therapy, according to the decision of the attending physician. Drainage failure was defined as the need for a second drainage procedure or for surgery.

Table 2 Comparison of patients’ characteristics and outcome according to the type of drainage

CVC

(n = 112)

CT

(n = 100)

P

Characteristics

Age, y

38 +- 19

42 +- 21

.12

Smoking status, n (%)

72 (64%)

67 (67%)

.77

PSP, n (%)

71 (63%)

46 (46%)

.01

COPD, n (%)

9 (8%)

19 (19%)

.02

Traumatic, n (%)

6 (5%)

13 (13%)

.06

Iatrogenic, n (%)

26 (23%)

22 (22%)

.87

Days before drainage

2.0 +- 2.4

2.3 +- 2.1

.28

Period, 2008-2010, n (%)

83 (74%)

6 (6%)

b.001

Outcome

Duration of drainage, d

3.3 +- 1.9

4.6 +- 2.6

b.01

Length of stay, d

4.5 +- 3.2

5.5 +- 3.0

.02

Primary drainage failure, n (%)

20 (18%)

21 (21%)

.60

Surgery, n (%)

16 (14%)

20 (20%)

.28

Statistical analysis

Demographic and outcome data were expressed as the means (+-SDs), and dichotomous variables were reported as numbers (percentages). Qualitative data were compared using Fisher exact test, and quantitative data, using the unpaired Student t or the Mann-Whitney U test when appropriate. P b .05 was considered significant.

First, we performed a multivariate analysis using a stepwise logistic regression model according to the Akaike criteria. All variables in Table 4 were included in the model. For each step, the variable maximizing the Akaike criteria was excluded from the model. The final model contained variables minimizing the criteria. Then, to validate the results, we adjusted the propensity score using the method introduced by Rosenbaum and Runbin [9] in 1983. It was generated using the characteristics of the univariate analyses in Table 2 with a significant P value. The final propensity score model included the following covariates: PSP, SSP- COPD, and the 2008 to 2010 period. Finally, the multivariate stepwise logistic regression model was repeated by adding the propensity score to the original model. Statistical analyses were performed using the nonrandom package 1.1, using R 2.13.0 (online at http://www.R-project.org, The R Foundation for Statistical Computing, Vienna, Austria).

Results

Patients

Over an 8-year period, 268 patients were admitted for pneumothorax. Among them, 56 patients were excluded because of conservative treatment without drainage (n = 11), direct surgical intervention without drainage (n = 23), hemopneumothorax (n = 13), and non-COPD SSP (n = 9) (Fig. 2). These 9 SSPs were due to nonspecific interstitial pneumonia (n = 2), sarcoidosis fibrosis (n = 2), sequellar

Fig. 2 Flow chart. Among the 268 patients admitted for pneumothorax to our unit, 56 patients were excluded because of conservative treatment without drainage (n = 11), decision of surgery upon admission for recurrent or contralateral pneumothorax (n = 23), SSP not due to COPD (n = 9), and hemopneumothorax (n = 13). Ultimately, 212 patients were included in the analysis, 112 of whom were initially treated with CVCs and 100 with conventional CTs.

tuberculosis (n = 2), X histiocytosis (n = 1), hamartochon- droma (n = 1), and lung abscess (n = 1). Ultimately, 212 patients were included in the analysis, of whom 112 were initially treated with a CVC and 100 with a conventional CT.

Comparison of patients’ characteristics and outcomes according to the cause of pneumothorax

Among the 212 patients included, 117 (55%) were PSP,

28 (13%) were SSP-COPD, 19 (9%) were traumatic, and 48

(23%) were iatrogenic (Table 1).

Iatrogenic pneumothorax caused few respiratory symp- toms (b20% of the patients had chest pain), and the first drainage was almost always successful because only 1 (2%) in 48 patients required a second drainage, and no patients required surgery (Table 1).

Chest pain was almost always (97%) reported upon admission by patients having PSP, whereas dyspnea occurred in only 40% of the patients, and only 3 of 117 patients had SpO2 below 90% at admission (Table 1). Despite more frequent dyspnea and lower SpO2, the time that elapsed between the onset of symptoms and drainage was longer in patients with SSP-COPD compared with that of patients with PSP. The rate of drainage success did not differ in SSP- COPD compared with that of PSP (64% vs 77%, P = .23), but the duration of drainage was longer (Table 1).

Comparison of patients’ characteristics and outcomes according to the type of drainage

Whatever the cause of pneumothorax, 24% of the patients (29/123) were treated using CVC from 2003 to 2007, compared with 93% (83/89) from 2008 to 2010 (P b

.001). Catheter drainage was most often used in cases of PSP, whereas CT was most often used in SSP-COPD (Table 2). The failure rate was similar between the 112 patients treated using CVC and the 100 patients treated using CT (18% vs 21%, P = .60). However, the durations of drainage and hospital stay were shorter when patients received catheter drainage. Among the 117 patients with PSP, there were no significant differences between those receiving catheter or CT drainage in terms of demographic charac- teristics (Table 3). Again, the duration of drainage was shorter for those patients treated with CVC compared with CT, but their lengths of stay were similar. Rates of drainage failure and of surgery were similar in patients treated with CVC or CT.

Risk factors for drainage failure

Drainage failure occurred in 23% of PSP, 36% of SSP- COPD, 16% of traumatic pneumothorax, and 2% of iatrogenic pneumothorax (Table 1). For spontaneous pneu- mothoraces (PSP or SSP), the rate of drainage failure was 24% (29/121) in cases of a first episode and 33% (8/24) in cases of recurrence (P = .44). The type of drainage had no influence on success or failure (Table 4). For catheter and CT drainage, the Failure rates were 20% (14/71) and 28% (13/46), respectively, in PSP (P = .37); 33% (3/9) and 37% (7/19), respectively, in SSP-COPD (P = .99); 33 (2/6) and 8% (1/13), respectively, in traumatic pneumothorax (P =

.22); and 4 (1/26) and 0% (0/22), respectively, in iatrogenic pneumothorax (P = .99). In univariate analysis, iatrogenic pneumothorax was associated with a higher success rate, whereas SSP-COPD was associated with a higher failure rate (Table 4). After adjustment using the propensity score, iatrogenic pneumothorax was a strong independent factor for

successful drainage (odds ratio [OR], 0.05), and low SpO

Table 3 Comparison of patients’ characteristics and outcome according to the type of drainage in PSP

CVC

(n = 71)

CT

(n = 46)

P

Characteristics

Age, y

27 +- 6

27 +- 8

.77

Smoking status, n (%)

49 (69%)

31 (67%)

.99

First episode, n (%)

59 (83%)

38 (82%)

.99

Delay before drainage, d

1.9 +- 2.7

2.2 +- 2.1

.61

Outcome

Duration of drainage, d

3.3 +- 1.9

4.5 +- 2.0

b.01

Length of stay, d

4.5 +- 3.1

5.2 +- 2.2

.20

Primary drainage failure, n (%)

14 (20%)

13 (28%)

.37

Surgery, n (%)

13 (18%)

12 (26%)

.36

2

was the only factor that tended to increase the risk of drainage failure (OR, 1.08 for each percentage decrease of SpO2), although the increase was not significant (Table 4).

Failure (n = 41)

Success (n = 171)

P

Propensity score-adjusted OR (95% CI); P value

Characteristics

Age, y

36 +- 16

41 +- 20

.22

Male sex, n (%)

36 (88%)

136 (80%)

.27

Smoking status, n (%)

31 (76%)

108 (63%)

.15

Recurrent episode, n (%)

8 (20%)

16 (9%)

.09

Chest pain, n (%)

37 (90%)

123 (72%)

.01

Dyspnea, n (%)

24 (59%)

74 (43%)

.08

SpO2 (%)

94 +- 6

95 +- 5

.27

OR, 1.08 (0.99-1.17); P = .07 a

Delay before drainage

2.0 +- 2.5

2.2 +- 2.2

.36

Type of drainage

Drainage using CVC, n (%)

20 (49%)

92 (54%)

.60

OR, 0.90 (0.44-1.87); P = .78

Period, 2008-2010, n (%)

16 (39%)

73 (43%)

.73

Cause of pneumothorax

PSP, n (%)

27 (66%)

90 (53%)

.16

SSP-COPD, n (%)

10 (24%)

18 (11%)

.04

OR, 1.23 (0.44-3.44); P = .69

Iatrogenic, n (%)

1 (2%)

47 (27%)

.0003

OR, 0.05 (0.01-0.44); P b .01

Traumatic, n (%)

3 (7%)

16 (9%)

N.99

CI indicates confidence interval.

a For each percentage of desaturation.

After failure of first-line catheter drainage (n = 20), a CT was placed as a second-line therapy in 7 patients, avoiding surgery in 4 of them. After failure of a first-line CT (n = 21), a second CT was successful in only 1 patient. Overall, surgery was ultimately necessary in 16 (14%) of 112 patients and in 20 (20%) of 100 patients (P = .27), respectively, when catheter drainage or CT was the First-line treatment (Table 2).

Table 4 Risk factors for drainage failure

Discussion

In this study, we found that small-bore catheter (5F catheter) drainage was at least as effective as CT (14-20F catheter) drainage in the initial management of pneumotho- rax, irrespective of its cause. The rates of drainage failure and of second-line surgery were similar using primary catheter or CT drainage, whether the pneumothorax was spontaneous (PSP and SSP-COPD), traumatic, or iatrogenic. The type of drainage had no influence on the success or failure of drainage, but the durations of drainage and of hospital stay were significantly shorter using catheter drainage.

Small-bore pleural catheter

A few small and noncomparative series have demonstrat- ed the safety and effectiveness of small-bore catheters in patients having first episodes of spontaneous [10-18] or iatrogenic pneumothorax [11-13,16,19-22]. Theses results were supported by 4 retrospective comparative studies that found that small-bore catheters (8-14F catheter) were as effective as large CTs in treating spontaneous pneumothorax [23-26], whether primary [23-25] or secondary [24-26]. In

these previous studies, small-bore catheters were used in 50 PSPs [23], 69 SSPs [26], and 24 and 36 spontaneous pneumothoraces, respectively [24,25]. However, these studies used catheters no smaller than 8F catheter and only focused on spontaneous pneumothoraces, excluding patients with traumatic or iatrogenic pneumothorax. We reported on 112 patients treated with a 5F catheter single-lumen pleural catheter, including 80 patients having spontaneous pneumo- thorax. Although our study was also retrospective, this was, to our knowledge, the largest comparative study on the management of pneumothorax that included pneu- mothoraces from diverse mechanisms.

Use of a simple CVC

Only 2 noncomparative studies have evaluated the effectiveness of a simple CVC [18,19]. The success rates were 96% among 28 iatrogenic pneumothoraces [19] and

86% among 127 spontaneous pneumothoraces [18]. In the latter study, the success rate of the first drainage attempt was 77%, and some patients required a second catheter for complete re-expansion of the lung [18]. In our study, the primary success rate using CVC drainage was comparable with the 96% and 79% success rates among the 26 iatrogenic and 80 spontaneous pneumothoraces, respectively.

Clinical implications

Consistent with the recent BTS guidelines [2], our study supported that CT drainage should not be used as the first-line approach in PSP management. The success rate of drainage in our patients with PSP was 80%, similar to previous studies

reporting a 65% to 88% success rate using either CT or small- bore catheter drainage in PSP [23-25]. Moreover, our study showed that CVC drainage was as effective as CT drainage for the management of iatrogenic pneumothorax–with a partic- ularly high success rate, close to 100%, as already reported in the literature [19]–as well as in patients with traumatic pneumothorax or SSP-COPD. In the population of patients with COPD, we found a primary success rate of 64%, similar to the 65% success rate of catheter drainage reported by Maury et al [18] and close to the 70% rate reported by Tsai et al [26] using small-bore catheters or CTs and the 75% rate reported by Chen et al [27] using a Pigtail catheter (10-16F catheter). One of these studies found a higher rate of drainage failure in SSP- COPD than in PSP in an unadjusted analysis [18]. By contrast, we found that the rate of drainage failure was not significantly different between SSP-COPD and PSP. After adjustment using the propensity score, COPD did not influence drainage failure, and iatrogenic pneumothorax was the only cause indepen- dently associated with a high rate of successful drainage.

The insertion of a CVC catheter into the pleural space is easier, safer, and faster to implement than CT insertion. Catheter drainage appears as effective as conventional CT drainage, is minimally invasive and associated with less pain and discomfort, and results in a smaller scar after removal of the catheter. For all these reasons, we no longer use CT insertion in the initial management of pneumothorax, and a pleural catheter is almost always placed first, whatever the cause of the pneumothorax.

Limitations

First, the data were collected retrospectively from a medical record review, and the study was conducted in a single unit. Our results may, therefore, not be applicable in all centers. Furthermore, a well-performed, randomized, controlled trial would provide confirmation of the results reported here. However, these results appear applicable to patients hospitalized immediately or after the failure of needle aspiration.

Second, we evaluated an overlap period during which both techniques were used, suggesting that a specific technique was preferentially selected in some patients. This limitation probably explains why patients with COPD more frequently received CT drainage, whereas those patients having PSP were more frequently treated with a CVC. However, we adjusted our analysis with a propensity score for using one technique or the other, and we found that the presence of COPD and the type of drainage had no influence on its outcome. Despite the use of a propensity score, the retrospective nature of our study makes it likely that unmeasured confounding variables affected the choice of drainage modality, and these confounders may have affected the observed results.

Third, we excluded SSP because of non-COPD, and the results cannot be extrapolated to patients with underlying

chronic restrictive pulmonary disease or patients with reduced pulmonary compliance. However, the main cause was due to COPD, which represents most SSPs [26-29].

Fourth, we did not analyze the recurrence rate of pneumothorax in our patients; however, it has been shown that the risk of recurrence is similar among patients treated using a small-bore catheter or a CT [24,26].

Finally, it should be noted that our study did not answer the question of the management of drainage failure after a CVC placement and whether it is better to insert a CT as second-line treatment or to refer the patient to surgery.

Conclusion

We reported here the largest series on the management of pneumothorax, comparing drainage using a large CT or a CVC with a much smaller caliber. Management of pneumothorax using simple CVC (5F catheter) drainage appeared as effective as drainage using a larger (14-20F catheter) CT. No matter which drainage procedure, the failure rate was close to zero in iatrogenic pneumothoraces. Simple CVC drainage seemed effective, both in spontaneous pneumothorax (primary or secondary due to COPD) and in traumatic and iatrogenic pneumothoraces. Because the procedure is minimally invasive and less painful, we recommend its use for all types of large pneumothoraces requiring drainage; CT drainage should be used only as a second line of therapy.

References

  1. Baumann MH, Strange C, Heffner JE, et al. Management of spontaneous pneumothorax: an American College of Chest Physicians Delphi consensus statement. Chest 2001;119:590-602.
  2. MacDuff A, Arnold A, Harvey J. Management of spontaneous pneumothorax: British Thoracic Society Pleural Disease Guideline 2010. Thorax 2010;65(Suppl 2):ii18-31.
  3. Harvey J, Prescott RJ. Simple aspiration versus intercostal tube drainage for spontaneous pneumothorax in patients with normal lungs. British Thoracic Society Research Committee. BMJ 1994;309:1338-9.
  4. Andrivet P, Djedaini K, Teboul JL, et al. Spontaneous pneumothorax. Comparison of thoracic drainage vs immediate or delayed needle aspiration. Chest 1995;108:335-9.
  5. Noppen M, Alexander P, Driesen P, et al. 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;165:1240-4.
  6. Ayed AK, Chandrasekaran C, Sukumar M. Aspiration versus tube drainage in primary spontaneous pneumothorax: a randomised study. Eur Respir J 2006;27:477-82.
  7. Ho KK, Ong ME, Koh MS, et al. A randomized controlled trial comparing minichest tube and needle aspiration in outpatient management of primary spontaneous pneumothorax. Am J Emerg Med.
  8. Mongardon N, Tremey B, Marty J. Thoracentesis and chest tube management in critical care medicine: a multicenter survey of current practices. Chest 2010;138:1524-5.
  9. Rosenbaum PR, Rubin DB. The central role of propensity scores in observational studies for causal effects. Biometrika 1983;70:41-55.
  10. Horsley A, Jones L, White J, et al. Efficacy and complications of small-bore, wire-guided chest drains. Chest 2006;130:1857-63.
  11. Martin T, Fontana G, Olak J, et al. Use of pleural catheter for the management of simple pneumothorax. Chest 1996;110:1169-72.
  12. Samelson SL, Goldberg EM, Ferguson MK. The thoracic vent. Clinical experience with a new device for treating simple pneumo- thorax. Chest 1991;100:880-2.
  13. Peters J, Kubitschek KR. Clinical evaluation of a percutaneous pneumothorax catheter. Chest 1984;86:714-7.
  14. Conces Jr DJ, Tarver RD, Gray WC, et al. Treatment of pneumothoraces utilizing small caliber Chest tubes. Chest 1988;94: 55-7.
  15. Minami H, Saka H, Senda K, et al. Small caliber catheter drainage for spontaneous pneumothorax. Am J Med Sci 1992;304:345-7.
  16. Sargent EN, Turner AF. Emergency treatment of pneumothorax. A simple catheter technique for use in the radiology department. Am J Roentgenol Radium Ther Nucl Med 1970;109:531-5.
  17. Marquette CH, Marx A, Leroy S, et al. Simplified stepwise management of primary spontaneous pneumothorax: a pilot study. Eur Respir J 2006;27:470-6.
  18. Maury E, Doyon F, Guidet B, et al. Drainage of spontaneous pneumothorax using an intravenous catheter. Immediate and long-term results. Respir Med 1998;92:961-2.
  19. Laub M, Milman N, Muller D, et al. Role of small calibre chest tube drainage for iatrogenic pneumothorax. Thorax 1990;45:748-9.
  20. Pancione L. The treatment of iatrogenic pneumothorax with small- gauge catheters. The author’s personal experience in 30 cases. Radiol Med 2000;100:42-7.
  21. Casola G, vanSonnenberg E, Keightley A, et al. Pneumothorax: radiologic treatment with small catheters. Radiology 1988;166:89-91.
  22. Perlmutt LM, Braun SD, Newman GE, et al. Transthoracic needle aspiration: use of a small chest tube to treat pneumothorax. AJR Am J Roentgenol 1987;148:849-51.
  23. Liu CM, Hang LW, Chen WK, et al. Pigtail tube drainage in the treatment of spontaneous pneumothorax. Am J Emerg Med 2003;21: 241-4.
  24. Vedam H, Barnes DJ. Comparison of large- and small-bore intercostal catheters in the management of spontaneous pneumothorax. Intern Med J 2003;33:495-9.
  25. Benton IJ, Benfield GF. Comparison of a large and small-calibre tube drain for managing spontaneous pneumothoraces. Respir Med 2009;103:1436-40.
  26. Tsai WK, Chen W, Lee JC, et al. Pigtail catheters vs Large-bore chest tubes for management of Secondary spontaneous pneumothoraces in adults. Am J Emerg Med 2006;24:795-800.
  27. Chen CH, Liao WC, Liu YH, et al. Secondary spontaneous pneumothorax: which associated condition is benefit for pigtail catheter treatment? Am J Emerg Med 2010.
  28. Gupta D, Hansell A, Nichols T, et al. Epidemiology of pneumothorax in England. Thorax 2000;55:666-71.
  29. Sahn SA, Heffner JE. Spontaneous pneumothorax. N Engl J Med 2000;342:868-74.