Documente Academic
Documente Profesional
Documente Cultură
ScienceDirect
Article history: Background: Numerous thoracoscopic techniques have been used in the management of
Received 18 July 2016 primary spontaneous pneumothorax (PSP), including wedge resection, pleurectomy,
Received in revised form pleural abrasion, chemical pleurodesis, and staple line covering. The purpose of this sys-
16 October 2016 tematic review was to compare outcomes for the most commonly reported techniques.
Accepted 26 October 2016 Materials and methods: A systematic literature search looking at pneumothorax recurrence
Available online 3 November 2016 rate, length of stay, and chest tube duration after surgery was conducted in accordance
with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines using
Keywords: the PubMed database.
Pneumothorax Results: Fifty-one unique studies comprised of 6907 patients published between January
Thoracoscopic surgery 1988 and June 2015 were identified. Heterogeneity among effect sizes was significant
Recurrence for all outcomes. The lowest recurrence rates were observed in the wedge
Pleurodesis resection chemical pleurodesis (1.7%; 95% confidence interval [CI], 1.0%-2.7%) and the
Pleural abrasion wedge resection pleural abrasion chemical pleurodesis (2.8%; 95% CI, 1.7%-4.7%)
Pleurectomy groups. The shortest chest tube duration and length of stay were observed in the wedge
resection staple line covering other group (2.1 d; 95% CI, 1.4-2.9 and 3.3 d; 95% CI, 2.6-
4.0, respectively).
Conclusions: The variability in reported outcomes and the lack of published multicenter
randomized controlled trials highlights a need for more robust investigations into the
optimal surgical technique in the management of PSP. Based on the limited quality studies
available, this systematic review favors wedge resection chemical pleurodesis and wedge
resection pleural abrasion chemical pleurodesis in terms of recurrence rate after
surgery for PSP.
2016 Elsevier Inc. All rights reserved.
This study was presented, in part, at the 2016 Annual meeting of the Southeastern Surgical Congress held in Atlanta, GA on February
21, 2016.
* Corresponding author. Division of Pediatric Surgery, Department of Surgery, Emory University School of Medicine, Childrens
Healthcare of Atlanta, 1405 Clifton Road NE, 3rd Floor Surgical Suite, Atlanta, GA 30322. Tel.: 1 404 785 0781; fax: 1 404 785 0800.
E-mail address: mehulvraval@emory.edu (M.V. Raval).
0022-4804/$ e see front matter 2016 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jss.2016.10.024
sudduth et al surgical technique in pneumothorax 33
Methods Data from individual studies were extracted, checked for ac-
curacy, and entered into an electronic data collection form.
Study selection Two authors (C.S. and J.S.) independently reviewed the studies
and treatment arms. If disagreement existed on whether a
This review was designed, performed, and reported in accor- study met inclusion criteria, the article was reviewed in detail
dance with the Preferred Reporting Items for Systematic Re- and discussed with other authors until reaching a consensus
views and Meta-Analyses (PRISMA) guidelines.8 We conducted (C.S., J.S., and M.V.R.).
a PubMed search for studies published between January 1988
(based on the emergence of thoracoscopic surgery in patients Treatment groups
with pneumothorax) and June 31, 2015, applying English
language and humans filters. The search strategy used the The precise intraoperative intervention reported by the au-
following Medical Subject Headings term: pneumothorax/ thors was recorded in the electronic data collection form. If
surgery. Furthermore, the reference lists of retrieved studies the same intervention was reported in at least two different
were reviewed for inclusion. articles and published results using that intervention met all
our inclusion criteria, then that intervention was identified as
Inclusion and exclusion criteria a treatment group.
STROBE guidelines were applied on an individual study level used the Q test to formally determine if heterogeneity was
in this analysis to achieve and maintain transparency present. To assess the robustness of our results, we conducted
regarding the quality and content of the available reporting. a sensitivity analysis to determine how sensitive the overall ES
Each individual article was searched for each of the recom- was to any one study by repeatedly calculating the overall ES
mended reporting items and assigned a score based on the with one study omitted per iteration. The recalculated ES was
number of recommended items that they included. When then compared with the overall ES to see the effect of a specific
items had multiple recommended components, partial credit study. We assessed the threat of publication bias using the
was given for each component that was included. trim and fill method, a nonparametric method that first esti-
mates the number of missing studies in a meta-analysis and
Statistical analysis then determines the effect they would have on the outcome
had they been present all along. If the added studies signifi-
To calculate effect sizes (ESs), we used means with standard cantly change the result, then publication bias is possible. To
deviations (SD) or frequencies with percentages. For each assess the quality of study reporting, we stratified our analysis
outcome of interest, recurrence rates, LOS, and CTD, we based on the level of bias indicated by our STROBE checklist.
calculated a separate ES estimate across studies for each PSP We determined the median STROBE scores and calculated ESs
treatment group. for studies above and below the median STROBE score. Forest
Data were entered and analyzed using Comprehensive plots were created using the Meta package in R (R Development
Meta-Analysis 3 (Englewood, NJ). All categorical outcomes Core Team: version 3.2.4-revised, 2016, Vienna, Austria).
(recurrence) were converted to proportions with associated 95%
confidence intervals (CIs) using the number of events or pro-
portion of event and the total sample size for each PSP subgroup Results
within a study. For continuous outcomes (LOS and CTD), data
were summarized using means and SDs. When the mean was Systematic review
unavailable, but the median and interquartile range were pro-
vided, the median was used in place of the mean and the SD was A total of 1103 studies were identified using the initial search
extrapolated from the interquartile range. An overall ES was criteria (Fig. 1). After application of initial inclusion and ex-
calculated for each PSP intervention group for each outcome of clusions criteria, 69 studies representing 115 treatment arms
interest and compared between intervention groups using were identified for data extraction. After applying treatment
random effects models. To examine the effect of age on choice arm exclusion criteria, 65 study arms representing 6907 pa-
of intervention and outcome, meta-regression was conducted. tients and 51 studies met final inclusion criteria.
To assess heterogeneity of outcomes within PSP interven- Studies included were published between 1994 and 2015
tion groups and between studies, we calculated ESs and asso- (Table 1). Of the 51 studies, 36 were retrospective reviews and/
ciated 95% CIs for each PSP intervention group. In addition, we or case series and 15 were prospective studies. Of the 15
Fig. 1 e Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram for study selection.
Table 1 e Studies included in the systematic review and meta-analysis.
Author Year Study Intervention Number of Age SD or Follow- SD or CTD SD or LOS SD or Recurrence
design patients (y)* interquartile up (mo)* interquartile (d)* interquartile (d)* range (%)
range (y) range (mo) range (d)
Imperatori 2015 Retrospective Wedge 134 25 7 79 36-187 5.97
resection
pleurectomy
Min 2014 Prospective, Wedge resection 144 22 5 18 6 1 10 2 6.25
randomized
by surgical
technique
35
(continued)
36
Table 1 e (continued )
Author Year Study Intervention Number of Age SD or Follow- SD or CTD SD or LOS SD or Recurrence
design patients (y)* interquartile up (mo)* interquartile (d)* interquartile (d)* range (%)
range (y) range (mo) range (d)
Park 2012 Retrospective Wedge 165 24.5 11.5 66.2 2.82 1.28 7.27
resection pleural
abrasion
Chen 2011 Retrospective Wedge 20 22.97 8.13 3.55 11.5 4.85 1.46 0
resection pleural
abrasion
Murumatsu 2011 Retrospective Wedge resection 357 27.2 51.2 10-108 8.40
j o u r n a l o f s u r g i c a l r e s e a r c h a p r i l 2 0 1 7 ( 2 1 0 ) 3 2 e4 6
Saito 2011 Retrospective Wedge resection 11 17 17-33 11 1-31 0
staple line
covering other
Shaikhrezai 2011 Retrospective Wedge resection 41 28.4 10.1 73 0
pleurectomy
Shaikhrezai 2011 Retrospective Wedge resection 189 28.4 10.9 73 1.06
chemical
pleurodesis
Shaikhrezai 2011 Retrospective Wedge 255 28.4 10.8 73 3.14
resection pleural
abrasion
Chou 2009 Retrospective Wedge 62 20 34 8 3 1 0
resection pleural
abrasion
Nakanishi 2009 Retrospective Wedge resection 157 28.1 12.6 19.96 23.18 9.55
staple line
covering other
Bialas 2008 Retrospective Wedge 31 16.5 46 6-104 4.1 2-19 4.3 2-11 6.45
resection pleural
abrasion
Cho 2008 Retrospecitve Wedge resection 424 24.9 29.4 13-45 1.7 1.8 3.2 2 4.72
staple line
covering other
Nathan 2008 Retrospective Wedge resection 40 30 24.1 1-72 3.6 3.4 2.5
pleurectomy
Butterworth 2007 Retrospective Wedge 10 14.6 11-17 19 12-23 4.7 1-9 6.7 2-13 10
resection pleural
abrasion
Marcheix 2007 Retrospective Wedge resection 603 30.1 11.5 8 5.4 1.49
chemical
pleurodesis
Bobbio 2006 Retrospective Wedge 67 27 14-40 76 1-106 2.99
resection pleural
abrasion
Santillan- 2006 Retrospective Wedge resection 55 24 17-47 48 6-72 3.5 1.7 0
Doherty pleurectomy
Chen 2004 Retrospective Wedge 51 26.2 10.9 39 1-120 7.7 3.2 9.80
resection pleural
abrasion
Chen 2004 Retrospective Wedge 313 24.9 39 1-120 5.8 3.7 2.88
resection pleural
abrasion
chemical
37
(continued)
38
Table 1 e (continued )
Author Year Study Intervention Number of Age SD or Follow- SD or CTD SD or LOS SD or Recurrence
design patients (y)* interquartile up (mo)* interquartile (d)* interquartile (d)* range (%)
range (y) range (mo) range (d)
Chan 2001 Retrospective Wedge 88 27 15-45 44.4 8-87 5.68
resection pleural
abrasion
Ayed 2000 Prospective Wedge resection 33 25 6 42 35-45 3 1 4.1 1 0
pleurectomy
Ayed 2000 Prospective Wedge 39 25 6 42 35-45 3.5 2 4.5 2.1 10.26
resection pleural
j o u r n a l o f s u r g i c a l r e s e a r c h a p r i l 2 0 1 7 ( 2 1 0 ) 3 2 e4 6
abrasion
Cardillo 2000 Retrospective Wedge resection 122 28.4 12-69 38 2-72 9.8
pleurectomy
Cardillo 2000 Retrospective Wedge resection 217 28.4 12-69 38 2-72 0.9
chemical
pleurodesis
Loubani 2000 Retrospective Wedge resection 26 31.8 3.1 38 36-40 4.7 1 6.76 1.09 19.23
Loubani 2000 Retrospective Wedge resection 26 29 3.2 38 36-40 3.1 1.09 4.8 1.08 3.85
chemical
pleurodesis
Miller 2000 Retrospective Wedge resection 45 27 8 39 10.7 4.7 4.4 6.67
pleurectomy
Zijl 2000 Retrospective Wedge 34 22 17-24 12 2 1-3 3 2-5 2.94
resection pleural
abrasion
chemical
pleurodesis
Horio 1998 Retrospective Wedge resection 51 34.8 16.4 3-42 2.2 3.1 6.4 5.2 13.73
Rieger 1998 Retrospective Wedge resection 21 31.7 17-54 34 3.2 2-5 5.8 4-8 9.52
pleurectomy
Rieger 1998 Retrospective Wedge resection 29 31.7 17-54 34 2.9 2-5 5.4 3-7 6.90
Dumont 1997 Retrospective Wedge 101 16 35 17-80 6.5 2-16 9.5 5-96 2.97
resection pleural
abrasion
Freixnet 1997 Retrospective Wedge 234 26 13-38 10-36 5.6 2-15 3.42
resection pleural
abrasion
McCarthy 1997 Retrospective Wedge resection 42 36.7 17-84 18 1-31 2.7 .5-9 5.1 0
pleurectomy
Yim 1997 Retrospective Wedge 196 13-102 20 1-36 1 1-17 2.5 1-19 1.02
resection pleural
abrasion
sudduth et al surgical technique in pneumothorax 39
11.11
2.90
1.82
0.07
0
0
Quality of reporting
3.2
3
7.5
8.2
3
similar between studies. Three studies performed either
sensitivity analyses or analyses of subgroups and in-
teractions, and only four acknowledged biases. Likewise, six
1.41
1.5
2.1
5.44
4.4
Treatment groups
49-70
49-70
8-20
10
15.1
24
30
30
12
16
15-58
13-81
16.8
17
17
37.2
37.2
34
43
36
20
69
25
110
30
resection pleural
resection pleural
resection pleural
Wedge resection
Wedge resection
Wedge resection
pleurectomy
pleurectomy
abrasion
abrasion
abrasion
Wedge
Wedge
Wedge
Retrospective
Retrospective
Retrospective
Retrospective
Prospective
Prospective
two treatment arms used talc and two treatment arms used
minocycline.11-18
Reinforcement of the staple line was preformed using
polyglycolic acid sheets (three treatment arms), fibrinogen-
based collagen fleece (one treatment arm), cellulose mesh
1996
1996
1996
1996
1995
1995
1994
Recurrence rates
Mouroux
Mouroux
Bertrand
Radberg
Waller
Yim
Kim
j o u r n a l o f s u r g i c a l r e s e a r c h a p r i l 2 0 1 7 ( 2 1 0 ) 3 2 e4 6
Park 2012 1 1 1 1 1 1 1 1 0 1 1 0.4 0 0.667 1 1 0 1 1 1 1 0 17.07
Chen 2011 1 1 1 1 1 1 1 1 0 1 1 0.2 0.5 0.333 1 1 0 1 0 1 0 0 15.03
Murumatsu 2011 1 1 1 1 1 1 1 1 0 1 1 0.4 1 0.667 1 1 0 1 0 1 1 0 17.07
Saito 2011 1 1 1 1 1 0.5 1 1 0 0.5 1 0.4 0.5 0.667 1 1 0 1 0 1 1 0 15.57
Shaikhrezai 2011 1 1 1 1 1 1 1 1 0 1 1 0.2 1 0.667 1 1 0 1 1 1 1 0 17.87
Chou 2009 1 1 1 1 1 1 1 1 0 1 1 0.2 1 0.667 1 1 0 1 0 1 0 0 15.87
Nakanishi 2009 1 1 1 1 1 1 1 0.5 0 1 1 0.2 0.5 0.667 1 1 0 1 0 1 1 0 15.87
Bialas 2008 1 1 1 1 1 1 1 1 0 1 1 0.4 0 0.667 1 1 0 1 0 1 1 0 16.07
Cho 2008 1 1 1 1 1 1 1 1 0 1 1 0.4 1 1 1 1 0 1 1 1 1 0 18.40
Nathan 2008 1 1 1 1 1 0.5 1 0.5 0 1 1 0 0.5 0.667 1 1 0 1 1 1 1 0 16.17
Butterworth 2007 1 1 1 1 1 1 1 1 0 1 1 0.2 0.667 0.667 1 1 0 1 0 1 1 0 16.53
Marcheix 2007 1 1 1 1 1 1 1 1 0 1 1 0.4 1 0.667 1 1 0 1 1 1 1 0 18.07
Bobbio 2006 1 1 1 1 1 1 1 1 0 1 1 0.2 1 0.667 1 1 0 1 0 1 0 0 15.87
Santillan-Doherty 2006 1 1 1 1 1 0.5 1 0 0 1 1 0 1 1 1 1 0 1 0 1 1 0 15.50
Chen 2004 1 1 1 1 1 1 1 1 0 1 1 0.2 0.5 0.667 1 1 0 1 0 1 0 0 15.37
Cheng 2004 1 1 1 1 1 1 1 1 0 1 1 0.2 0.5 0.667 1 1 0 1 0 1 0 0 15.37
Czerny 2004 1 1 1 1 1 1 1 1 0 1 1 0.4 1 1 1 1 0 1 1 1 1 0 18.40
Freixinet 2004 1 1 1 1 1 1 1 1 0 1 1 0.2 1 0.667 1 1 0 1 0 1 0 1 16.87
Gossot 2004 1 1 1 1 1 1 1 1 0 1 1 0.4 1 1 1 1 0 1 0 1 1 0 17.40
Sakamoto 2004 1 1 1 1 1 1 1 1 0 1 1 0.2 1 0.667 1 1 0 1 0 1 1 0 16.87
Chen 2003 1 1 1 1 1 1 1 1 1 1 1 0.4 1 1 1 1 0 1 1 1 0 0 18.40
Lang-Lazdunski 2003 1 1 1 1 1 1 1 1 0 0.5 1 0.2 0.5 1 1 1 0 1 0 1 1 0 16.20
Margolis 2003 1 1 1 1 1 1 1 1 0 1 1 0 1 0.333 1 1 0 1 0 1 1 0 16.33
Horio 2002 1 1 1 1 1 0.5 1 1 0 1 1 0.4 0 0.667 1 1 0 1 0 1 1 0 15.57
Sawabata 2002 1 1 1 1 1 1 1 1 0 1 1 0.6 1 1 1 1 0 1 1 1 1 0 18.60
Chan 2001 1 1 1 1 1 1 1 1 0 1 1 0.4 1 1 1 1 0 1 0 1 1 0 17.40
Ayed 2000 1 1 1 1 1 0 1 1 0 1 1 0.2 0 0.667 1 1 0 1 0 1 1 0 14.87
Cardillo 2000 1 1 1 1 1 0.5 1 1 0 1 1 0.4 0 0.667 1 1 0 1 0 1 1 0 15.57
Loubani 2000 1 1 1 1 1 1 1 0.5 0 1 1 0.2 1 0.667 1 1 0 1 0 1 1 0 16.37
Zijl 2000 1 1 1 1 1 0 1 1 0 0 1 0 0 0.667 1 1 0 1 0 1 0 0 12.67
Miller 2000 1 1 1 1 1 1 1 1 0 1 1 0.2 1 0.667 1 1 0 1 0 1 1 0 16.87
Horio 1998 1 1 1 1 1 1 1 1 0 0 1 0.2 0 0.667 1 1 0 1 1 1 1 0 15.87
Rieger 1998 1 1 1 1 1 1 1 1 0 1 1 0.6 1 0.667 1 1 0 1 0 1 1 0 17.27
Freixnet 1997 1 1 1 1 1 0.5 1 1 0 0 1 0 0 0.33 1 1 0 1 0 1 0 0 12.83
McCarthy 1997 1 1 0 1 1 1 1 0.5 0 1 1 0 0.5 0.667 1 1 0 1 1 1 1 1 16.67
Yim 1997 1 1 1 1 1 0.5 1 0.5 0 0.5 1 0.2 0.5 0.667 1 1 0 1 1 1 1 1 16.87
Dumont 1997 1 1 1 1 1 1 1 1 0 1 1 0.2 1 0.667 1 1 0 1 0 1 0 0 15.87
Ba background; Bi bias; DD descriptive data; DS data sources; F funding; G generalizability; KR key results; L limitations; MR main results; O objective; OA other analyses;
P participants; QV quantitative variables; S setting; SD study design; SM statistical methods; SS study size; T total; T&A title and abstract; V variables.
White corresponds to high-quality study criteria, light gray corresponds to moderate quality, and dark gray corresponds to low quality.
41
42 j o u r n a l o f s u r g i c a l r e s e a r c h a p r i l 2 0 1 7 ( 2 1 0 ) 3 2 e4 6
Fig. 2 e Outcome forest plots. (A) Recurrence rate. (B) LOS. (C) CTD.
recurrence rates ranged from 0% to 23% with pooled ESs es- studies with younger patients tended to have higher recur-
timates ranging from 2% to 10% (Fig. 2A). For the treatment rence rates relative to studies with older patients. In contrast,
groups wedge resection pleural abrasion and wedge for the treatment group wedge resection pleurectomy, this
resection staple line covering other, the significant het- effect was not observed (slope 0.01; 95% CI, 0.09 to 0.06;
erogeneity in treatment ESs implies that there was significant
variability in recurrence rates across the group of studies
(Q values of 54.6 and 19.1 with P < 0.01, respectively). Thus, a
random effect model was justified when calculating the
within treatment group ES. Among the six treatment groups,
there was significant heterogeneity in treatment ESs (Q 52.6;
P < 0.001) indicating significant differences in recurrence rates
among the treatment groups. Recurrence rate was lowest in
the wedge resection chemical pleurodesis (1.7%; 95% CI,
1.0%-2.7%) and wedge resection pleural abrasion chemical
pleurodesis (2.8%; 95% CI, 1.7%-4.7%) groups. Recurrence rate
was highest in the wedge resection alone group (9.7%; 95% CI,
7.7%-12.2%).
For the treatment groups wedge resection pleural abra-
sion (26) and wedge resection pleurectomy (15), we possessed
a sufficient number of studies to perform a meta-regression
analysis to determine the impact of age at treatment on the
outcome recurrence rate. When regressing the studies average
age on the logit recurrence rate, age was significantly associ- Fig. 3 e Recurrence rate by age for the two most common
ated with recurrence rates in patients undergoing wedge intervention groups, wedge resection D pleural abrasion,
resection pleural abrasion (slope 0.09; 95% CI, 0.15 to and wedge resection D pleurectomy. (Color version of
0.05, P < 0.001; Fig. 3). The negative slope indicates that figure is available online.)
sudduth et al surgical technique in pneumothorax 43
P 0.820). The average age of patients studied ranged from 14.6 Therefore, the results appear robust in spite of possible pub-
to 43 y. lication bias. For the outcome LOS, Duvall and Tweedies trim
and fill method indicated that some studies could be imputed
Length of stay for the treatments wedge resection alone and wedge
resection pleurectomy (Supplemental Table S2). Like
A total of 26 studies reported data on 33 treatment arms for recurrence rates, none of the adjusted ESs were significantly
LOS (days). Individual treatment group average LOS ranged different than the original ESs. Funnel plots for the treatment
from 2.4 to 10 d with pooled ESs estimates ranging from 3.3 to groups with possible publication bias are presented in
7.0 d (Fig. 2B). For all treatment groups, heterogeneity was Supplemental Figure S2A and B. For CTD, none of the treat-
present and statistically significant (P < 0.001); therefore, a ment groups that were examined required imputation of
random effect model was justified when calculating the missing studies (Supplemental Table S3).
within treatment group ES. Among the six treatment groups, We further stratified our analysis based on the mean
there was significant heterogeneity in overall treatment ESs STROBE guidelines score of our 51 studies (16.5 versus >16.5)
(Q 22.38; P < 0.001), indicating significant variability in the for each of the outcomes (Supplemental Tables S4-S6). We
LOS across the group of studies. LOS was the shortest in the compared studies with higher quality reporting (STROBE score
wedge resection staple line covering other (3.3 d; 95% CI, >16.5) versus those with poor reporting (16.5). For the
2.6-4.1 d) and the wedge pleural abrasion chemical pleu- outcome recurrence rate, only the treatment group wedge
rodesis (3.9 d; 95% CI, 2.2-5.7) treatment groups. LOS was resection alone showed a significant difference in the recur-
longest in the wedge resection alone (7.0 d; 95% CI, 5.1-8.8) and rence rate when stratifying by quality of reporting. For the
the wedge resection chemical pleurodesis (6.4 d; 95% CI, 3.3- groups with the most omissions in their reporting, the esti-
9.5) treatment groups. mated ES for recurrence rate was 15.9% (95% CI, 10.5%-23.3%)
compared with a lower recurrence rate for the studies with
Chest tube duration better reporting (8.2%; 95% CI: 6.5%-10.3%). There were some
differences seen for the outcomes LOS and CTD for the
A total of 21 studies reported data on 29 treatment arms for treatment groups wedge resection chemical pleurodesis and
CTD (days). Individual study treatment group CTD means wedge resection pleural abrasion chemical pleurodesis;
ranged from 1.7 to 6.0 d with pooled ESs estimates ranging however, this is likely because of the small number of studies
from 2.1 to 4.0 d (Fig. 2C). For all treatment groups, heteroge- reporting for the treatment group (n 3). For the other treat-
neity was present and statistically significant (P < 0.001), ment groups, no difference was seen in ESs when stratifying
justifying the use of a random effects model when calculating by quality of reporting.
the within treatment group ES and overall ES. Significant
heterogeneity in treatment ESs was observed (Q 13.05;
P 0.023). This implies that there is significant variability in Discussion
CTD. CTD was shortest in the wedge resection staple line
covering other group (2.1 d; 95% CI, 1.4-2.9). CTD was longest The purpose of this review was to determine which inter-
in the wedge resection chemical pleurodesis (4.1 d; 95% CI, vention, or combination of interventions, performed best for
2.19-5.92) and wedge resection pleural abrasion groups each of our three outcomes of interest for patients undergoing
(4.0 d; 95% CI, 3.10-4.93). surgical management of PSP. The main finding of this review
was the paucity of high-quality, prospective, randomized tri-
Sensitivity analysis als in the literature. The current literature, comprised mainly
of single-institution case series, suggests that recurrence rates
Sensitivity analysis was performed to determine whether one for pneumothorax are lowest in patients undergoing wedge
study was overly influential in estimating the overall ES resection chemical pleurodesis (1.7%; 95% CI, 1.0-2.7) or
within a given treatment group. This consisted of removing wedge resection pleural abrasion chemical pleurodesis
one study a time and recalculating the overall ES. Sensitivity (2.8%; 95% CI, 1.7-4.7). Patients are at the highest risk of
analysis for recurrence rates, LOS, and CTD did not show any suffering a recurrence after wedge resection alone (9.7%; 95%
one study to be overly influential. CI, 7.7-12.2). However, it is difficult to draw definitive conclu-
sions from this comparison because of the heterogeneity of
Publication bias the treatment groups.
To the authors knowledge, this is the first systematic re-
For the outcome recurrence rate, Duvall and Tweedies trim view of reported literature assessing outcomes after video-
and fill method indicated that some studies could be imputed assisted thoracoscopic surgery (VATS) for pneumothorax.
and thus there is the possibility of publication bias because of The high recurrence rate after stapler wedge resection alone is
missing studies. The adjusted ESs are presented in in agreement with another, nonsystematic review that esti-
Supplemental Table S1. In addition, forest plots showing the mated the rate to be between 10% and 20%.26 In that review,
four treatment groups with possible threat of publication bias the lowest recurrence rates were also seen in the chemical
are provided in Supplemental Figure S1A-D. For the four pleurodesis groups (1.1%-1.9%).12,13,16 Interestingly, our data
treatment groups identified as having missing studies, none showed that the addition of pleural abrasion to chemical
of the adjusted ESs were significantly different than the orig- pleurodesis and wedge resection was associated with an
inal ESs and the adjusted ES were well within the 95% CI. increased recurrence rate. Incomplete abrasion or differences
44 j o u r n a l o f s u r g i c a l r e s e a r c h a p r i l 2 0 1 7 ( 2 1 0 ) 3 2 e4 6
in surgeon-specific techniques may explain variances in tended to be longer in the wedge resection alone, wedge
recurrence rate. However, CIs overlapped significantly and so resection chemical pleurodesis and wedge resection pleural
definitive conclusions should not be drawn from this abrasion groups. However, reported outcomes varied signifi-
distinction. cantly, and CIs were wide.
A surgical procedure is performed in a minority of patients Looking over the past 20 years, only two published pro-
hospitalized for spontaneous pneumothorax.27 Nonoperative spective randomized trials on surgical technique in the
management options for these patients include observation, management of PSP were available for inclusion in this anal-
supplemental oxygen, aspiration, and intercostal drainage ysis.15,44 The need for high-quality, multicenter, prospective
with a pig tail catheter or chest tube. Recurrence rates in these randomized trials assessing the efficacy of different in-
cases range from 48.7% to 54.2%.4,28 In addition, chemical terventions has been established.26,45 The variability in re-
pleurodesis can also be performed alone, with or without ported outcomes and overlapping CIs in this systematic
VATS. In the absence of VATS, this is done by instilling a review further supports the idea that a clear optimal tech-
chemical in the pleural space using an intercostal drainage nique will not be elucidated by the continued reporting of case
tube. Although data are limited on recurrence using this series alone.
method, reported rates range from 13% to 29.2%.29,30 Using The ideal prospective randomized trial would include
VATS, chemical pleurodesis is performed under direct thor- intervention arms that are safe, commonly performed, and
acoscopic guidance. A single, small study found a recurrence effective at reducing future risk of recurrence. The findings
rate of 5% using this method.31 However, after selecting for from this review would suggest a treatment group comprised,
patients at low risk of suffering a recurrence, other, larger at least in part, of wedge resection chemical pleurodesis. It
studies have reported recurrence rates between 0.3% and would be interesting to see the effect of adding chemical
2.41%.32,33 Specific surgical technique for these patients was pleurodesis to the other intervention groups (i.e., wedge
determined at the time of VATS based on thoracoscopic resection pleurectomy and wedge resection staple line
findings. covering other). Furthermore, as demonstrated by the
Reductions in recurrence risk must be weighed against significantly higher recurrence rate, wedge resection alone
negative side effects. Cases of acute respiratory distress syn- does not seem worthy of further analysis in future studies. As
drome and empyema have been reported after chemical for outcomes, given that recurrence is most common in the
pleurodesis with talc.34,35 However, more recent studies show first 2 y, patients should be followed for at least this long.33
this risk to be minimal, if not nonexistent, when large-particle Finally, data on potential long-term complications associ-
talc is used.36 In addition, concerns exist regarding the feasi- ated with the different types of pleurodesis are scarce; so
bility of subsequent thoracic surgery after chemical pleurod- careful monitoring would be strongly recommended.
esis because of fears of excessive thoracic sclerosis.37 There This study has several limitations. First, it is a systematic
seems to be little evidence in the literature supporting this review of mainly case series and is therefore at an increased
concern. In fact, in 39 patients with recurrent PSP after initial risk of selection bias. There is likely heterogeneity within
talc pleurodesis who were managed with a VATS, 69% were patient populations and intervention groups: the indications
managed successfully.38 As for other long-term risks of for surgery were variable among studies and exact surgical
chemical pleurodesis, a structured review of the literature techniques for the same intervention groups can differ be-
found no increased incidence of lung cancer or decreased tween surgeons and between institutions. This has implica-
pulmonary function in patients who underwent talc pleu- tions for the external validity of the study. Some treatment
rodesis.39 In summary, chemical pleurodesis, particularly groups were comprised of a small number of published
with talc, is considered a safe and effective option for PSP by studies and were therefore subject to reporting bias. Finally,
the thoracic surgery community.29-33,36,38-41 CIs for many treatment groups were wide indicating impre-
Younger age is a known risk factor for recurrence after cise measures of effect.
surgery for PSP.42,43 The effect of age on recurrence risk was
analyzed for the two treatment groups with sufficient data
to complete the analysis: (1) wedge resection pleural Conclusion
abrasion and (2) wedge resection pleurectomy (Fig. 3). In
the wedge resection pleural abrasion group, younger age Recurrence rates after pneumothorax can be minimized with
was unexpectedly associated with an increased risk of recur- VATS, but there is a paucity of high-quality, prospective ran-
rence. However, this effect was not seen in the wedge domized trials to guide decisions on surgical technique. This
resection pleurectomy group. Although the overall recur- meta-analysis, comprised mainly of retrospective reviews,
rence rate was comparable between the two groups in older suggests that wedge resection chemical pleurodesis and
patients (<5%), recurrence increased in younger patients un- wedge resection pleural abrasion chemical pleurodesis
dergoing wedge resection pleural abrasion but not wedge are associated with the lowest recurrence rates. However, a
resection pleurectomy. Although the intervention of choice definitive conclusion on surgical technique cannot be drawn
for PSP is generally surgeon or institution dependent, this is from this comparison because of the nature of the studies that
preliminary data that wedge resection pleural abrasion may comprise the meta-analysis. In addition, as seen by the in-
not be the optimal intervention to prevent recurrence in crease in recurrence rate in younger patients undergoing
younger patients. wedge resection pleural abrasion, patient age may be an
LOS and CTD were both shortest in the wedge important consideration in the optimal surgical technique for
resection staple line covering other group. CTD and LOS pneumothorax recurrence prevention.
sudduth et al surgical technique in pneumothorax 45
24. Sakamoto K, Takei H, Nishii T, et al. Staple line coverage with 36. Janssen JP, Collier G, Astoul P, et al. Safety of pleurodesis with
absorbable mesh after thoracoscopic bullectomy for talc poudrage in malignant pleural effusion: a prospective
spontaneous pneumothorax. Surg Endosc. 2004;18:478e481. cohort study. Lancet. 2007;369:1535e1539.
25. Yang HC, Cho S, Jheon S. Single-incision thoracoscopic 37. Naunheim KS, Hazelrigg SR, Ferguson MK, et al. Safety
surgery for primary spontaneous pneumothorax using the and efficacy of video-assisted thoracic surgical
SILS port compared with conventional three-port surgery. techniques for the treatment of spontaneous
Surg Endosc. 2013;27:139e145. pneumothorax. J Thorac Cardiovasc Surg.
26. Muramatsu T, Nishii T, Takeshita S, Ishimoto S, Morooka H, 1995;109:1198e1204.
Shiono M. Preventing recurrence of spontaneous 38. Doddoli C, Barlesi F, Fraticelli A, et al. Video-assisted
pneumothorax after thoracoscopic surgery: a review of recent thoracoscopic management of recurrent primary
results. Surg Today. 2010;40:696e699. spontaneous pneumothorax after prior talc pleurodesis: a
27. Bobbio A, Dechartres A, Bouam S, et al. Epidemiology of feasible, safe and efficient treatment option. Eur J
spontaneous pneumothorax: gender-related differences. Cardiothoracic Surg. 2004;26:889e892.
Thorax. 2015;70:653e658. 39. Hunt I, Barber B, Southon R, Treasure T. Is talc pleurodesis
28. Noh D, Lee S, Haam SJ, Paik HC, Lee DY. Recurrence of safe for young patients following primary spontaneous
primary spontaneous pneumothorax in young adults and pneumothorax? Interact Cardiovasc Thorac Surg.
children. Interact Cardiovasc Thorac Surg. 2015;21:195e199. 2007;6:117e120.
29. Chen JS, Chan WK, Tsai KT, et al. Simple aspiration and 40. Chen JS, Hsu HH, Chen R, et al. Additional minocycline
drainage and intrapleural minocycline pleurodesis versus pleurodesis after thoracoscopic surgery for primary
simple aspiration and drainage for the initial treatment of spontaneous pneumothorax. Am J Respir Crit Care Med.
primary spontaneous pneumothorax: an open-label, parallel- 2006;173:548e554.
group, prospective, randomised, controlled trial. Lancet. 41. Moreno-Merino S, Congregado M, Gallardo G, et al.
2013;381:1277e1282. Comparative study of talc poudrage versus pleural abrasion
30. Almind M, Lange P, Viskum K. Spontaneous pneumothorax: for the treatment of primary spontaneous pneumothorax.
comparison of simple drainage, talc pleurodesis, and Interact Cardiovasc Thorac Surg. 2012;15:81e85.
tetracycline pleurodesis. Thorax. 1989;44:627e630. 42. Chiu CY, Chen TP, Wang CJ, Tsai MH, Wong KS. Factors
31. Tschopp JM, Boutin C, Astoul P, et al. Talcage by medical associated with proceeding to surgical intervention and
thoracoscopy for primary spontaneous pneumothorax is recurrence of primary spontaneous pneumothorax in
more cost-effective than drainage: a randomised study. Eur adolescent patients. Eur J Pediatr. 2014;173:1483e1490.
Respir J. 2002;20:1003e1009. 43. Huang H, Ji H, Tian H. Risk factors for recurrence of primary
32. Cardillo G, Carleo F, Giunti R, et al. Videothoracoscopic talc spontaneous pneumothorax after thoracoscopic surgery.
poudrage in primary spontaneous pneumothorax: a single- Biosci Trends. 2015;9:193e197.
institution experience in 861 cases. J Thorac Cardiovasc Surg. 44. Min X, Huang Y, Yang Y, et al. Mechanical pleurodesis
2006;131:322e328. does not reduce recurrence of spontaneous
33. Cardillo G, Bintcliffe OJ, Carleo F, et al. Primary spontaneous pneumothorax: a randomized trial. Ann Thorac Surg.
pneumothorax: a cohort study of VATS with talc poudrage. 2014;98:1790e1796.
Thorax. 2016;71:847e853. 45. Bille A, Barker A, Maratos EC, Edmonds L, Lim E. Surgical
34. How C, Hsu H, Chen J. Chemical pleurodesis for spontaneous access rather than method of pleurodesis (pleurectomy or
pneumothorax. J Formos Med Assoc. 2013;112:749e755. pleural abrasion) influences recurrence rates for
35. Rehse DH, Aye RW, Florence MG. Respiratory failure following pneumothorax surgery: systematic review and meta-
talc pleurodesis. Am J Surg. 1999;177:437e440. analysis. Gen Thorac Cardiovasc Surg. 2012;60:321e325.