General Thoracic and Cardiovascular Surgery

https://doi.org/10.1007/s11748-019-01207-2

ORIGINAL ARTICLE

Reduction of air leakage using linear staple device with bioabsorbable

polyglycolic acid felt for pulmonary lobectomy
Hiroyuki Deguchi1 · Makoto Tomoyasu1 · Wataru Shigeeda1 · Yuka Kaneko1 · Hironaga Kanno1 · Hajime Saito1

Received: 29 January 2019 / Accepted: 9 September 2019

© The Japanese Association for Thoracic Surgery 2019

Abstract
Objective  Pulmonary fistula is a common complication in pulmonary lobectomy for non-small cell lung cancer (NSCLC).
Although a linear staple device with bioabsorbable polyglycolic acid (PGA) is used for pulmonary wedge resection with
fragile pulmonary parenchyma, the efficacy of the stapler with PGA for dividing incomplete interlobular fissure in pulmo-
nary lobectomy has not been elucidated. This study aimed to evaluate the usefulness of the stapler with PGA in reducing
postoperative air leakage when dividing incomplete interlobular fissure in pulmonary lobectomy for NSCLC.
Methods  A total of 546 patients who underwent radical lobectomy for NSCLC were analyzed retrospectively. Propensity
score analysis generated two matched pairs of 125 patients in both stapler and stapler with PGA groups.
Results  After propensity score matching, postoperative air leakage following pulmonary lobectomy was significantly less
frequent in the stapler with PGA group (9.6%) than in the stapler group (22.4%, p = 0.006). Intraoperative additional man-
agement of PGA and/or fibrin glue was decreased in the stapler with PGA group (56.0% vs. 70.4%, p = 0.018, 54.4% vs.
69.6%, p = 0.013, respectively). On logistic regression analysis, stapler with PGA was an independent factor for preventing
postoperative air leakage (odds ratio, 0.38; p = 0.015).
Conclusions  Using the stapler with PGA to divide the incomplete interlobular fissure in pulmonary lobectomy reduced
postoperative air leakage, and decreased the need for additional intraoperative management using fibrin glue.

Keywords  Lung cancer · Pulmonary lobectomy · Air leakage · Linear stapler · Bioabsorbable polyglycolic acid

Introduction polyglycolic acid (PGA) or fibrin glue, to prevent air leak-

Although opportunities to perform radical surgery for non-
small cell lung cancer (NSCLC) have been increasing in
recent years, further improvements in surgical outcomes will
require progress in overcoming postoperative complications.
Various complications are seen with pulmonary lobectomy
for NSCLC [1], pulmonary fistula is one of the common
complications. Previous reports have shown a 90% incidence
of air leakage at staples during the operation [2], and have
indicated that air leakage might increase the risk of serious
postoperative complications, such as postoperative pneumo-
thorax [3]. Pulmonary fistula encountered during surgery
is generally covered with a reinforcing material, such as
age. Several reinforcing materials have been proposed [4–6].
Recently, a linear staple device has been used with PGA for
pulmonary wedge resection in fragile or fibrotic pulmonary
parenchyma to prevent air leakage from the needle hole of
the staple line.
Pulmonary fistula is likely to occur at the site of dividing
the incomplete interlobular fissure in pulmonary lobectomy,
and an automatic suturing device is often used for dividing
the incomplete interlobular fissure to prevent postoperative
air leakage. However, the efficacy of using a stapler with
PGA for dividing an incomplete interlobular fissure in pul-
monary lobectomy has yet to be elucidated.

* Hajime Saito
hasaito@iwate‑med.ac.jp
1
Department of Thoracic Surgery, School of Medicine,
Iwate Medical University, 2‑1‑1 Idai‑dori, Yahaba, Shiwa,
Iwate 028‑3695, Japan

13
Vol.:(0123456789)

Patients and methods
Patient selection
We reviewed medical records for this retrospective cohort
study. All patients had NSCLC and underwent radical
lobectomy between January 2014 and October 2018 at
Iwate Medical University Department of Thoracic Surgery.
This study was approved by the institutional review board
at Iwate Medical University, and the need for informed
consent was waived based on the retrospective design (per-
mit number: H28-194).
Surgical indications for complete video-assisted thora-
coscopic surgery (c-VATS) lobectomy comprise the
concept of a thoracoscopically resectable lesion, cover-
ing almost 95% of surgical patients with NSCLC in our
institute. Cases of completed interlobular fissure in which
an automatic linear stapler was not required, pneumonec-
tomy, bilobectomy, and sublobar resection were excluded.
All patients underwent complete preoperative pulmonary
evaluation. Pulmonary function was tested at our insti-
tute using a spirometer (CHESTAC 8800; CHEST M.I.,
Tokyo, Japan) according to the standards of the Ameri-
can Thoracic Society [7]. Vital capacity (VC) and forced
expiratory volume in 1 s (­ FEV1) were measured in patients
preoperatively within 1 month before surgery. Patients
with chronic obstructed pulmonary disease (COPD) were
treated by bronchodilator therapy for at least 4 weeks pre-
operatively, and smoking was stopped for at least 8 weeks
before surgery. Postoperative complications within 30 days
after surgery were defined as Clavien–Dindo classifica-
tion grade II or higher [8]. Postoperative pulmonary com-
plications [9] included pneumonia, prolonged air leak,
interstitial pneumonitis (IP), atelectasis, bronchopleural
fistula, bronchial asthma, hypoxemia and acute respiratory
distress syndrome. Cases showing multiple complications
were categorized according to the complication that most
affected the postoperative outcome. Prolonged air leak was
defined as air leakage lasting ≥ 7 days [10]. A final total of
463 patients met the selection criteria.
Surgical procedures
Pulmonary lobectomy was performed under general anes-
thesia with a double-lumen endotracheal tube for single-
lung ventilation. The affected lung was deflated as soon
as the pleural space was opened, and deflation was main-
tained throughout most of the operative period. The frac-
tion of inspired oxygen ­(FiO2) during surgery ranged from
0.3 to 1.0 based on intraoperative blood gas analysis evalu-
ation. The open method was performed via posterolateral
13
General Thoracic and Cardiovascular Surgery
thoracotomy dividing part of the latissimus dorsi and ante-
rior serratus muscles. Complete VATS lobectomy was per-
formed via a three-port method with monitor vision only.
To divide the incomplete interlobular fissure in pulmo-
nary lobectomy, automatic linear stapler devices were used.
Up to January 2015, simple staplers (without PGA) were
used for dividing the incomplete interlobular fissure (stapler
group, n = 135) in our institute, and staplers with PGA were
used from February 2015 (stapler with PGA group, n = 328).
In particular, Endo GIA Reinforced Reload with Tri-Staple
­technology® (purple, 60 mm; Medtronic, Minneapolis, MN)
were used as staplers with PGA, and Endo GIA Reload with
Tri-Staple ­Technology® (purple, 60 mm; Medtronic) were
mainly used as a simple stapler.
Systematic complete hilar and mediastinal lymph node
dissection was performed in all cases. After completing
the procedure, a sealing test was performed before wound
closure. The sealing test was confirmed with reinflation of
the lung on the affected side, additional management with
PGA sheet (­ NEOVEIL® sheet; GUNZE, Osaka, Japan) and/
or fibrin glue (­ Beriplast® P combi-set tissue adhesion; CSL
Behring, Tokyo, Japan, or B ­ OLHEAL®, TEIJIN Pharma,
Tokyo, Japan) was performed in the case of air leakage.
Finally, a chest tube (­ Blake®, 19-Fr; Ethicon, Somerville,
NJ) was placed from the fifth intercostal trocar to the apex.
Postoperative management
Chest X-rays were obtained daily. The chest tube was suc-
tioned at -5 ­cmH2O on the morning of postoperative day 1.
Chest tube withdrawal criteria were: absence of air leak-
age through the chest tube at the time of evaluation; pleural
fluid drainage < 200 mL/24 h; and postoperative chest X-ray
showing no pneumothorax. On the morning after chest tube
withdrawal, chest X-ray was performed to exclude the pres-
ence of pneumothorax. Routine postoperative pain manage-
ment was performed in all patients of both groups. Briefly,
oral analgesia was started a day after surgery and typically
included loxoprofen at 60 mg, three times per day, some-
times with 25-mg diclofenac suppositories one or two times
per day, as needed. Patients were discharged when conveni-
ent, if no complications occurred during this perioperative
period. Our institutional standard protocol is to follow-up
almost patients every 3–6 months after surgery for 5 years.
Cost analysis
The cost of the stapler cartridge was 36,000 JPY and the
stapler with PGA cartridge was 58,200 JPY for each device.
The cost of each additional PGA sheet was 9,650 JPY, fibrin
glue cost 36,625 JPY for 3 ml and 59,259 JPY for 5 ml. This
study calculated costs based on these prices.
General Thoracic and Cardiovascular Surgery
Statistical analysis
JMP version 12.2.0 statistical software (SAS Institute,
Cary, NC) was used for all statistical analyses. Groups were
compared using the Pearson Chi-square test or Wilcoxon’s
rank-sum test. To control for potential differences in the
preoperative characteristics of patients in the two groups,
a propensity score matching method was used. Propensity
scores were generated using logistic regression based on
clinically relevant preoperative variables such as age, gen-
der, body mass index (BMI), pack-years smoked (Brinkman
index), presence of COPD (defined as percentage in F ­ EV1:
­FEV1% < 70%), presence of IP findings from preoperative
computed tomography (CT) of the chest, history of steroid
therapy, diabetes mellitus (DM), tumor size, and surgical
approach, which were considered as possible confounders
for potential associations with the outcome of interest based
on clinical knowledge. Patients were matched 1:1 by nearest
neighbor matching (caliper width, 0.2) without replacement.
Comparisons between matched groups were performed with
McNemar’s test for categorical variables and the paired t
test or Wilcoxon’s signed rank test for continuous variables.
The standardized difference was used to measure covariate
balance, whereby an absolute standardized difference > 0.1
represents a meaningful imbalance.
Multivariate predictors were evaluated using logistic
regression analysis, and odds ratios (ORs) and 95% confi-
dence intervals (CIs) were estimated. On logistic regression
analysis, the conventional receiver-operating characteristic
(ROC) curve was used to determine the cut-off value for
each variable that offered maximal sensitivity and specificity
with respect to predicting postoperative air leakage in this
study population. Differences between groups were consid-
ered significant at the level of p < 0.05. Continuous data are
expressed as mean ± standard deviation. Categorical data are
expressed as counts and proportions.
Results
The 546 patients with resected NSCLC who underwent
radical surgery were retrospectively reviewed. Reasons for
excluding patients from this study are shown in Fig. 1. A
total of 463 patients (stapler group, 135 patients; stapler with
PGA group, 328 patients) were enrolled in this study.
The clinical characteristics of the study population are
summarized in Table 1. For eligible patients, no signifi-
cant differences in age, gender, BMI, or Brinkman index
were evident between groups. Compared with the stapler
group, the stapler with PGA group showed a significantly
higher frequency of patients with COPD (10.4% vs. 3.4%,
p < 0.002), IP (15.2% vs. 8.2%, p < 0.040), and steroid
treatment history (15.2% vs. 3.0%, p < 0.001), respectively.
Patients with radical lobectomy for lung cancer
between January 2014 and October 2018 (N=546)
Exclusion
Pneumonectomy (n=17)
Bilobectomy (n=19)
Segmentectomy (n=24)
Completed fissure (n=23)
Eligible patients underwent lobectomy (n=463)
Stapler (n=135)
Stapler with PGA (n=328)
Propensity score matching
Un-matched (n=213)
Stapler (n=10)
Stapler with PGA (n=203)
Matched (n=250)
Stapler (n=125)
Stapler with PGA (n=125)
Fig. 1  Diagram for patient selection
Cases with postoperative air leakage were less frequent in
the stapler with PGA group (12.5% vs. 25.2%, p < 0.001),
and the frequencies of additional management of PGA
and/or fibrin glue were also lower in the stapler with PGA
group (55.2% vs. 71.1%, p = 0.002, 54.3% vs. 69.6%,
p = 0.002, respectively).
Propensity score analysis generated two sets of 125
well-matched pairs (Table 2); briefly, 92.6% (125/135)
in the stapler group and 38.1% (125/328) in the stapler
with PGA group were matched. No significant differences
were seen in observed any preoperative variables such
as age, gender, BMI, Brinkman index, history of under-
lying disease, or perioperative variables such as tumor
size, operation time, and postoperative stay between both
groups after matching. The balance of each sample size
was assessed by standardized differences, and values for
pre- and perioperative variables were almost all under 0.1.
The stapler with PGA group still showed a significantly
lower incidence of postoperative pulmonary complications
(8.0% vs.16.8%, p = 0.035), postoperative air leakage com-
pared to the stapler group (9.6% vs. 22.4%, p = 0.006),
and additional management with PGA and/or fibrin glue
(56.0% vs. 70.4%, p = 0.018; 54.4% vs. 69.6%, p = 0.013),
respectively.
13
 General Thoracic and Cardiovascular Surgery

Table 1  Clinical background All patients

of all patients who underwent
pulmonary lobectomy Stapler Stapler with PGA p value
(n = 135) (n = 328)

Age (years) 68.8 ± 9.3 70.1 ± 9.0 0.125

Gender
 Male 87 (64.4) 188 (57.3) 0.156
 Female 48 (35.6) 140 (42.7)
BMI (kg/m2) 23.7 ± 3.4 23.1 ± 3.3 0.210
Brinkman index 500.4 ± 517.2 546.8 ± 615.1 0.720
Underlying disease
 COPD 11 (3.4) 14 (10.4) 0.002*
 Interstitial pneumonia 11 (8.2) 50 (15.2) 0.040*
 Steroid therapy 4 (3.0) 50 (15.2)  < 0.001*
 Diabetes mellitus 24 (17.8) 84 (25.6) 0.070
Preoperative pulmonary function
 VC (ml) 3264.0 ± 756.6 3158.3 ± 749.5 0.126
 FEV1 (ml) 2378.3 ± 587.3 2276.6 ± 527.4 0.083
 FEV1% (%) 74.2 ± 8.9 73.9 ± 8.8 0.302
Procedure
 RUL 36 (26.7) 108 (32.9) 0.157
 RML 8 (5.9) 20 (6.1)
 RLL 26 (19.3) 90 (27.4)
 LUL 38 (28.2) 58 (17.7)
 LLL 27 (20.0) 52 (15.9)
Approach
 c-VATS 123 (91.1) 291 (88.7) 0.447
Histology
 Adenocarcinoma 98 (72.6) 244 (74.4) 0.537
 Squamous cell carcinoma 22 (16.3) 58 (17.7)
 Other 15 (11.1) 26 (7.9)
Tumor size (mm) 29.7 ± 17.2 30.9 ± 19.1 0.486
Operation time (min) 250.4 ± 66.4 263.3 ± 62.5 0.016*
Blood loss (ml) 52.6 ± 59.5 48.9 ± 62.8 0.154
Additional management
 PGA 96 (71.1) 181 (55.2) 0.002*
 Fibrin glue 94 (69.6) 178 (54.3) 0.002*
Period of chest tube placement (days) 4.9 ± 2.0 5.2 ± 3.3 0.201
Postoperative stay (days) 11.5 ± 12.6 10.6 ± 8.5 0.386
Postoperative complications
 Total 39 (28.9) 86 (26.2) 0.557
 Pulmonary 24 (17.8) 31 (9.5) 0.029*
Postoperative air leakage 34 (25.2) 41 (12.5)  < 0.001*
Prolonged air leakage (>7 days) 4 (3.0) 10 (3.0) 0.961

*p < 0.05 vs. non-reinforced group

The results of multivariate analysis of the independent
risk predictors for postoperative air leakage are shown in
Table 3. Using the stapler with PGA was an independent fac-
tor related to reducing postoperative air leakage in matched
patients (OR, 0.38; p = 0.015).
Table 4 shows the mean cost comparison between stapler
and stapler with PGA groups. The mean cost of additional
PGA and fibrin glue was lower in the stapler with PGA

13
General Thoracic and Cardiovascular Surgery

Table 2  Clinical characteristics after propensity score matching

of patients who underwent
pulmonary lobectomy after Stapler Stapler with Standardized p value
propensity score matching (n = 125) PGA (n = 125) difference

Age (years) 68.7 ± 9.5 69.4 ± 9.3 0.074 0.523

Gender
 Male 77 (61.6) 70 (56.0) 0.093 0.368
 Female 48 (38.4) 55 (44.0)
BMI (kg/m2) 23.6 ± 3.5 24.1 ± 3.1 0.151 0.211
Brinkman index 475.0 ± 514.5 472.2 ± 560.9 0.005 0.870
Underlying disease
 COPD 6 (4.8) 8 (6.4) 0.063 0.582
 Interstitial pneumonia 6 (4.8) 5 (4.0) 0.035 0.758
 Steroid therapy 4 (3.2) 4 (3.2) 1.000 0.022*
 Diabetes mellitus 21 (16.8) 19 (15.2) 0.039 0.730
Preoperative pulmonary function
 VC (ml) 3216.5 ± 757.4 3291.2 ± 777.0 0.097 0.510
 FEV1 (ml) 2354.0 ± 597.0 2322.3 ± 555.9 0.055 0.746
 FEV1% (%) 74.3 ± 9.0 73.2 ± 7.6 0.132 0.061
Procedure
 RUL 33 (26.4) 31(24.8) 0.032 0.772
 RML 8 (6.4) 9 (7.2) 0.029
 RLL 23 (18.4) 36 (28.8) 0.218
 LUL 37 (29.6) 26 (20.8) 0.180
 LLL 24 (19.2) 23 (18.4) 0.018
Approach
 c-VATS 113 (90.4) 110 (88.0) 0.045 0.541
Histology
 Adenocarcinoma 94 (75.2) 98 (78.4) 0.055 0.550
 Squamous cell carcinoma 18 (14.4) 18 (14.4) 0.000
 Other 13 (10.4) 9 (7.2) 0.102
Tumor size (mm) 29.9 ± 17.6 29.9 ± 20.0 0.000 0.919
Operation time (min) 248.9 ± 66.4 257.3 ± 53.8 0.139 0.055
Blood loss (ml) 50.4 ± 59.1 51.1 ± 62.5 0.012 0.901
Additional management
 PGA 88 (70.4) 70 (56.0) 0.240 0.018*
 Fibrin glue 87 (69.6) 68 (54.4) 0.254 0.013*
Period of chest tube placement (days) 4.8 ± 2.1 5.2 ± 3.5 0.139 0.110
Postoperative stay (days) 10.7 ± 7.2 11.5 ± 12.4 0.079 0.327
Postoperative complications
 Total 35 (28.0) 31 (24.8) 0.064 0.566
 Pulmonary 21 (16.8) 10 (8.0) 0.242 0.035*
Postoperative air leakage 28 (22.4) 12 (9.6) 0.316 0.006*
Prolonged air leakage (>7 days) 3 (2.4) 1 (0.8) 0.116 0.314

*p < 0.05 vs. non-reinforced group

group compared to the stapler group (p = 0.019, p < 0.014, Discussion

respectively), with no significant difference in total mean
costs, comprising the stapler devices plus additional PGA
and/or fibrin glue for surgery in both groups (p = 0.059).
This study was undertaken to evaluate the usefulness of the
stapler with PGA for dividing the incomplete interlobular
fissure to reduce the incidence of postoperative air leakage
in patients with NSCLC. Propensity score analysis generated

13
 General Thoracic and Cardiovascular Surgery

Table 3  Logistic regression analysis of independent risk predictors postoperative air leakage even if no problem is evident mac-
for postoperative air leakage on pulmonary lobectomy roscopically during surgery. In other words, the macroscopic
Variable After propensity score matching findings may not necessarily agree with the actual vulner-
ability. Prediction of postoperative air leakage before surgery
OR 95% CI p value
has proven extremely difficult so far, and preoperative factors
Age (≧ 70 years) 1.47 0.67–3.32 0.338 such as COPD, history of DM or steroid therapy did not cor-
Gender (male) 2.61 0.90–7.95 0.077 relate with postoperative air leakage in this study.
BMI (≧ 25 kg/m2) 0.64 0.24–1.55 0.326 The principle of preventing pulmonary fistula, of course,
Brinkman index (≧ 400) 1.49 0.54–4.17 0.438 involves not damaging the pleura of the remaining pulmo-
COPD (presence) 1.68 0.34–7.72 0.512 nary lobe when dividing the incomplete interlobular fissure
Interstitial pneumonia (presence) 8.08 0.33–2.97 0.167 in pulmonary lobectomy. Moreover, pulmonary fistula is
Steroid therapy (presence) 0.88 0.01–3.15 1.000 also a complication that can be prevented by intraoperative
Diabetes mellitus (presence) 1.22 0.41–3.31 0.711 surgical techniques [13]. The following points are consid-
Tumor size (> 30 mm) 1.63 0.71–3.72 0.242 ered important when using an automatic linear staple device.
Histology (adenocarcinoma) 1.06 0.35–3.53 0.147 First, separate the interlobular fissure to a depth that does
Approach (c-VATS) 1.44 0.41–6.04 0.575 not damage the pleura of the remaining pulmonary lobe.
Additional PGA 3.18 0.24–36.82 0.362 This prevents excessive tension on the pulmonary tissue that
Additional fibrin glue 1.16 0.12–14.00 0.900 could lacerate the lung tissue. Second, select a stapler device
Stapler with PGA 0.38 0.17–0.83 0.015* of appropriate height. Third, in cases with damage to the
residual pulmonary pleura, division of the interlobular fis-
*p < 0.05
sure is performed on the remaining pulmonary lobe side so
as to avoid leaving a damaged lesion. By paying attention
matched pairs of 125 patients from the stapler and stapler to these above points, pulmonary fistula can be prevented
with PGA groups, and the incidence of postoperative air to some extent. Although pulmonary fistula caused by sta-
leakage was significantly lower in the stapler with PGA pler needle holes in fragile tissues such as pulmonary tis-
group than in the stapler group. On logistic regression analy- sue can be prevented, pulmonary fistula caused by cracks
sis, stapler with PGA was demonstrated to be an independent in the vicinity of the staple line (e.g., fibrotic pulmonary
factor for reducing postoperative air leakage on pulmonary parenchyma) cannot be prevented even when using the
lobectomy. This result indicates that using the stapler with stapler with PGA. No significant difference in the popula-
PGA for dividing incomplete interlobular fissure on pulmo- tion of COPD and IP were seen between both groups after
nary lobectomy in patients with NSCLC reduces postopera- propensity score matching, suggesting that the bias in the
tive air leakage. fragility of lung tissue that causes leakage near the staple
Several reports have examined the usefulness of stapler line would be comparable between with and without PGA
with PGA in the area of gastroenterology [11, 12]. However, groups. However, our study actually demonstrated decreased
while the stapler with PGA is used for pulmonary wedge the incidence of pulmonary fistula in the stapler with PGA
resection with fragile or fibrotic pulmonary parenchyma group. This result indicated the utility of using a stapler with
to prevent air leakage, the effect of stapler with PGA for PGA in preventing leakage, especially air leakage from the
dividing the incomplete interlobular fissure in pulmonary hole of stapler needle. For these reasons, the occurrence
lobectomy has not been clarified. If fragility of the lung tis- of pulmonary fistula cannot be completely controlled, but
sue is predicted before or during surgery, a PGA automatic using a stapler with PGA decreased the incidence of pul-
suturing device can be used alone in the adapted cases. monary fistula in this study. However, pulmonary fistula is
Although pulmonary fistula is considered less likely to occur thought to be influenced by surgeon factors [13], and only
based on intraoperative findings, several cases have shown two expert surgeons currently perform pulmonary lobectomy

Table 4  The mean cost of Stapler (n = 125) Stapler with PGA (n = 125) p value
stapler and additional PGA
and fibrin glue in both stapler Additional management
and stapler with PGA group on
 PGA 6793.6 ± 4422.9 5404.0 ± 4809.4 0.019*
c-VATS lobectomy in patient
with lung cancer  Fibrin glue 25,491.0 ± 16,914.7 19,924.0 ± 18,314.9 0.014*
Stapler 252,576.0 ± 98,800.2 297,859.2 ± 138,818.5 0.012*
Total 284,860.6 ± 103,539.8 323,187.2 ± 136,085.9 0.059

The unit is Japanese Yen (JPY). *p < 0.05

13
General Thoracic and Cardiovascular Surgery
at our institute. Each is an experienced surgeon, so the learn-
ing curve for this short-term study period was not a problem.
In addition, no difference in postoperative management was
seen during these two periods.
The stapler with PGA offers a useful option for reduc-
ing postoperative air leakage, but potential cost problems
need to be considered. In this study, using the stapler with
PGA did not decrease the postoperative hospital stay, but
decreased the use of additional treatments such as PGA sheet
and fibrin glue. As a result, no significant differences were
seen in the total cost for pulmonary lobectomy. It is true
that the cost of staples with PGA tends to be higher, even
if not significantly so, but the advantage of the stapler with
PGA significantly reducing the use of blood products such
as fibrin glue (Table 4, p < 0.014). Reducing postoperative
complications is extremely important, and reducing the use
of blood products is also considered important.
Our study had several limitations, including the retro-
spective nature and small number of patients from a single
institution. To reach strong conclusions, a randomized study
is essential. Propensity score matching is a useful method
that is effective when prediction is already attached to results
to some extent and forming a random comparison test is
difficult. However, not all preoperative differences between
groups were compensated for by propensity score-matching,
despite real efforts to control for possible biases. Neverthe-
less, we believe that our findings will contribute to improv-
ing the quality of pulmonary lobectomy in patients with
NSCLC. A multicenter prospective study is warranted to
validate our results in the future.
Conclusion
In conclusion, using a stapler with PGA for dividing incom-
plete interlobular fissure on pulmonary lobectomy in patients
with NSCLC significantly reduced postoperative air leakage.
Using the stapler with PGA also reduced the use of blood
products such as fibrin glue.
Acknowledgements  The authors thank Dr. Eri Maeda for her sugges-
tions and support with statistical analyses.
Compliance with ethical standards 
Conflict of interest  None of the authors have any conflicts of interest
to declare.
References
1. Committee for Scientific Affairs, The Japanese Association for
Thoracic Surgery, Masuda M, Endo S, Natsugoe S, Shimizu H,
Doki Y, Hirata Y, et al. Thoracic and cardiovascular surgery in
Japan during 2015: Annual report by The Japanese Associa-
tion for Thoracic Surgery. Gen Thorac Cardiovasc Surg. 2018;
66:581–615
2. Droghetti A, Schiavini A, Muriana P, Folloni A, Picarone M,
Bonadiman C, et al. A prospective randomized trial comparing
completion technique of fissures for lobectomy: stapler versus
precision dissection and sealant. J Thorac Cardiovasc Surg.
2008;136:383–91.
3. Ferguson MK, Gaissert HA, Grab JD, Sheng S. Pulmonary com-
plications after lung resection in the absence of chronic obstruc-
tive pulmonary disease: the predictive role of diffusing capacity.
J Thorac Cardiovasc Surg. 2009;138:1297–302.
4. Venuta F, Rendina EA, De Giacomo T, Flaishman I, Guarino E,
Ciccone AM, et al. Technique to reduce air leaks after pulmonary
lobectomy. Eur J Cardiothorac Surg. 1998;13:361–4.
5. Downey DM, Michel M, Harre JG, Pratt JW. Functional assess-
ment of a new staple line reinforcement in lung resection. J Surg
Res. 2006;131:49–52.
6. Hashimoto A, Kuwabara M, Hirasaki Y, Tsujimoto H, Torii T,
Nakamura T, et al. Reduction of air leaks in a canine model of
pulmonary resection with a new staple-line buttress. J Thorac Car-
diovasc Surg. 2011;142:366–71.
7. American Thoracic Society (1987) Standardization of spirom-
etry—1987 update. Statement of the American Thoracic Society.
Am Rev Respir Dis 1987; 136:1285–98
8. Dindo D, Demartines N, Clavien PA. Classification of surgical
complications: a new proposal with evaluation in a cohort of 6336
patients and results of a survey. Ann Surg. 2004;240:205–13.
9. Miskovic A, Lumb AB. Postoperative pulmonary complications.
Br J Anaesth. 2017;118:317–34.
10. Rivera C, Bernard A, Falcoz PE, Thomas P, Schmidt A, Bénard S,
et al. Characterization and prediction of prolonged air leak after
pulmonary resection: a nationwide study setting up the index of
prolonged air leak. Ann Thorac Surg. 2011;92:1062–8.
11. Naito M, Yamanashi T, Nakamura T, Miura H, Tsutsui A, Sato T,
et al. Safety and efficacy of a novel linear staple device with bioab-
sorbable polyglicolic acid felt in laparoscopic colorectal surgery.
Asian J Endosc Surg. 2017;10:35–9.
12. Naito M, Sato T, Nakamura T, Yamanashi T, Miura H, Tsutsui A,
et al. Secure overlap stapling using a linear stapler with bioabsorb-
able polyglycolic acid felt. Asian J Endosc Surg. 2017;10:308–12.
13. Okereke I, Murthy SC, Alster JM, Blackstone EH, Rice TW.
Characterization and importance of air leak after lobectomy. Ann
Thorac Surg. 2005;79:1167–73.
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
13