bs_bs_banner

Asian J Endosc Surg ISSN 1758-5902

ORIGINAL ARTICLE

Learning curve for the thoracoscopic repair of esophageal atresia

with tracheoesophageal fistula
Hiroomi Okuyama,1 Yuko Tazuke,1 Takehisa Ueno,1 Hiroaki Yamanaka,1 Yuichi Takama,1 Ryuta Saka,1
Noriaki Usui,2 Hideki Soh2 & Takeo Yonekura3
1 Department of Pediatric Surgery, Osaka University Graduate School of Medicine, Osaka, Japan
2 Department of Pediatric Surgery, Osaka Women’s and Children’s Hospital, Osaka, Japan
3 Department of Pediatric Surgery, Kindai University Nara Hospital, Nara, Japan

Keywords:
Esophageal atresia; learning
thoracoscopic repair
Correspondence
Hiroomi Okuyama, Department of
Pediatric Surgery, Osaka University
Graduate School of Medicine, 2-2
Yamadaoka, Suita, Osaka 565-0871 Japan.
Tel: +81 6 6879 3753
Email: okuyama@pedsurg.med.osaka-u.
ac.jp
Received 5 April 2017; revised 25 May
2017; accepted 11 June 2017
DOI: 10.1111/ases.12411
Abstract
curve;
Aim: Thoracoscopic repair (TR) of esophageal atresia with tracheoesophageal
fistula (EA/TEF) remains a considerable challenge, even for the most experienced
pediatric surgeons. The aim of this study is to report the outcomes of our
experience with TR of EA/TEF and to determine the learning curve for this
procedure.
Methods: Eleven consecutive cases that had undergone TR of EA/TEF at our
institutes were included in this study. The medical charts were reviewed
retrospectively. To determine the learning curve for TR of EA/TEF, a logarithmic
curve-fitting analysis was performed. The data were expressed as medians with
ranges.
Results: The median age and birth weight were 1 day (range, 1–3 days) and
2.8 kg (range, 2.5–3.7 kg), respectively. TR was completed in all cases without
any complications. The median operative time was 230 min (range, 164–
383 min). There were no cases of anastomotic leakage. One patient with a long
gap required repeated balloon dilatation for refractory anastomotic stricture. No
mortality or recurrence of tracheoesophageal fistula occurred. The operative time
was significantly longer in patients with a long gap (>20 mm) than in those with
a shorter gap. Once the three cases with a long gap had been excluded, the oper-
ative time decreased as the number of treated cases increased. The relationship
between the operative time and case number fit a logarithmic function curve well
(operative time in minutes = 300 – 62 × log (case number), R2 = 0.8359,
P = 0.0015).
Conclusions: Our results suggest that TR of EA/TEF is a safe procedure. It has a
considerable learning curve, but requires advanced endoscopic surgical skills.

Introduction individual practitioners is limited. Given that an endoscopic

procedure of such complexity requires a great deal of trial
Thoracoscopic repair (TR) of esophageal atresia with
and error before it can be performed without issue, reports
tracheoesophageal fistula (EA/TEF) was first performed in
on early trials may provide crucial knowledge to surgeons
2000 (1). Since then, there have been many reports
who are going to start performing the procedure. The aim
describing the operative techniques and outcomes of TR of
of this study is to report the outcomes of our experience
EA/TEF. It was recently reported by several large series
and to determine the learning curve for TR of EA/TEF.
and a meta-analysis that the outcomes are similar or supe-
rior to those of open thoracotomy, particularly with regard
to avoiding the musculoskeletal morbidity associated with Materials and Methods
that technique (2–4). TR of EA/TEF, however, remains At our institutes, we performed the first case of TR of
rather challenging, even for the most experienced pediatric EA/TEF in 2003. We have since used the thoracoscopic
surgeons because the number of procedures performed by approach in select cases. Our exclusion criteria include

Asian J Endosc Surg •• (2017) ••– ••

© 2017 Japan Society for Endoscopic Surgery, Asia Endosurgery Task Force and John Wiley & Sons Australia, Ltd 1
Thoracoscopic esophageal atresia repair
major coexisting anomalies, low birth weight (<2 kg), and
an unstable general condition. Because of the wide range
of operative techniques applied and patients’ clinical
courses, infants who had esophageal atresia (EA) without
tracheoesophageal fistula (TEF) were also excluded from
this study. Medical charts were reviewed retrospectively.
Outcomes examined included associated anomalies, birth
weight, age at operation, operation method (number of
ports, use of single-lung ventilation), gap between the
proximal and distal esophagus, and operative time. The
gap was measured using intraoperative videos. Complica-
tions included mortality, open conversion, anastomotic
leakage, anastomotic stricture requiring dilatation, gastro-
esophageal reflux requiring fundoplication, tracheomalacia
requiring aortopexy, and recurrence of TEF. To determine
the learning curve of TR of EA/TEF, a logarithmic
curve-fitting analysis was performed using the DeltaGraph
ver. 5.5.5ia J software program (Red Rock Software, Salt
Lake City, USA). The data were expressed as medians with
ranges.
This retrospective study was approved by the institutional
review board of Osaka University Hospital (#17004).
Operative method
The operation was uniformly performed via an intrapleural
approach with the patient in the 30° to 45° prone position.
Preoperative bronchoscopy was routinely performed in our
Figure 1 Operative methods. (a) Three or four ports were used with the patient in the 30° to 45° prone position. (b) The tracheoesophageal fistula
was occluded with two surgical clips. (c) Anastomosis was performed with 5-0 monofilament absorbable sutures. (d) A transanastomotic tube was placed
routinely.
2 © 2017 Japan Society for Endoscopic Surgery, Asia Endosurgery Task Force and John Wiley & Sons Australia, Ltd
H Okuyama et al.
institutes. Single-lung ventilation using a bronchial blocker
was attempted in the first seven cases, but it was not in the
last four cases. Three or four ports were used in each case.
After the first 3-mm trocar was inserted just below the
tip of the right scapula, artificial pneumothorax was
established at 4–6 mmHg by CO2 insufflation to collapse
the right lung. A 3-mm 30° telescope was inserted, and
two additional instrument ports were placed to achieve a
90° angle at the presumed site of the anastomosis. The
upper port was 5 mm to allow for the introduction of a clip
applier and suture needles. The lower port was 3 mm in size
and placed one or two intercostal spaces below and slightly
posterior to the camera port. A fourth port was occasionally
placed either higher or lower in the thoracic cavity to help
retract the lung (Figure 1a). Once pneumothorax was
established, the visceral pleura of the posterior mediasti-
num was incised. The azygos vein was divided using a
vessel-sealing system. The TEF was occluded with two sur-
gical clips (Figure 1b). The proximal esophageal pouch was
identified and dissected from the trachea up to the thoracic
inlet. Anastomosis was performed using the extracorporeal
knot-tying technique with 5-0 monofilament absorbable
sutures (Figure 1c). A transanastomotic tube was placed
routinely after completion of the posterior wall anastomosis
(Figure 1d). In cases of a long gap, more extensive dissec-
tion was performed up to the neck region to enable primary
anastomosis. After anastomosis was completed, a chest tube
was placed routinely, but gastrostomy was not placed
Asian J Endosc Surg •• (2017) •• –••
H Okuyama et al. Thoracoscopic esophageal atresia repair

routinely. Patients were left intubated and paralyzed for

2–4 days postoperatively depending on the tension of the
anastomotic site.

Results
A total of 11 neonates who had undergone TR of EA/TEF
were included in this study. In all cases, TR was performed
or supervised by a single surgeon. The major associated
anomalies were Treacher Collins syndrome in one, coarcta-
tion of the aorta with ventricular septal defect in two,
congenital duodenal stenosis in one, and congenital esoph-
ageal stenosis in one. Table 1 shows the patients’ records in
chronological order. The median age at operation was 1 day
(range, 1–3 days). The median birth weight was 2.8 kg
(range, 2.5–3.7 kg). While four ports were used in the first
six cases, TR was completed using three ports in the last five
cases. Single-lung ventilation was attempted in the first
seven cases with limited success, but it was not attempted
in the last four cases. The median operative time was
230 min (range, 164–383 min). The median gap between
the proximal and distal esophagus was 5 mm (range,
0–30 mm). The operative time was significantly longer in
the three patients with a long gap (>20 mm) than in those
with a shorter gap (long gap vs. shorter gap; 315 min
[range, 280–383 min] vs 215 min [range, 164–294 min],
P = 0.024). After the three cases of long gap were
excluded, the operative time decreased as the number of
consecutive cases increased. The relationship between
the operative time and the consecutive case number fit
a logarithmic function curve well (operative time in
minutes = 300 62 × log (case number), R2 = 0.8359,
P = 0.0015) (Figure 2).
TR was completed in all cases without any intraoperative
complications. Table 2 shows the incidence of postoperative
complications. There were no mortalities or no anastomotic
leakage. One patient with a long gap required repeated
balloon dilatation for refractory anastomotic stricture.
Figure 2 After the exclusion of the cases with a long gap, the relationship
between the operative time (op. time) and the consecutive case number fit
a logarithmic function curve well (operative time in minutes = 300 62 × log
2
(case number), R = 0.8359, P = 0.0015).
There was no chylothorax. Two patients required
fundoplication for gastroesophageal reflux. No patient
required aortopexy for tracheomalacia. No recurrent TEF
occurred.
Discussion
For all types of surgery, a learning curve exists before the
surgeon is able to acquire the adequate skills and experi-
ence. This learning curve is more evident for highly ad-
vanced endoscopic procedures, such as TR of EA/TEF,
even for the most experienced endoscopic surgeons (5,6).
However, there have been few reports regarding the learn-
ing curve for TR of EA/TEF. Lee et al. reported that the mean
operative time was significantly longer in the first 13 cases
in their series than in the later 9 cases; they also found that
leakage and stenosis occurred less frequently in the later
cases than in the earlier cases (7). Hiradfar et al. reported
that the open conversion rate decreased from 58.3% to
35.7% after the first 10 cases during the learning curve
period (8). Van der Zee et al. reported the results of

Table 1 Consecutive patients’ records

Patient # Age (days) Birth weight (kg) Number of ports Single-lung ventilation Gap (mm) Operation time (min)
1 1 2.8 4 Attempted 25 315
2 3 2.6 4 Attempted 5 294
3 1 2.5 4 Attempted 0 275
4 3 3.5 4 Attempted 10 201
5 1 2.6 4 Attempted 0 230
6 1 3.0 4 Attempted 10 230
7 2 3.0 3 Attempted 30 280
8 2 3.7 3 Not attempted 5 180
9 1 3.2 3 Not attempted 25 383
10 1 2.8 3 Not attempted 0 174
11 1 2.8 3 Not attempted 5 164

Long gap >20 mm.

Asian J Endosc Surg •• (2017) ••– ••

© 2017 Japan Society for Endoscopic Surgery, Asia Endosurgery Task Force and John Wiley & Sons Australia, Ltd 3
Thoracoscopic esophageal atresia repair
Table 2 Incidence of complications
Complications
Mortality
Open conversion
Anastomotic leakage
Stenosis requiring dilatation
Chylothorax
GER requiring fundoplication
Tracheomalacia requiring aortopexy
Recurrent TEF
GER, gastroesophageal reflux; TEF, tracheoesophageal fistula.
thoracoscopic esophageal atresia repair and compared 41
earlier cases with 31 later cases (5). The mean operative
time remained relatively unchanged, although they
claimed that this was because cases in the later part of their
series were performed by junior staff members who were
less experienced in the procedure.
Upon a retrospective review of our experience, the
operative time was found to be significantly longer in
patients with a long gap than in those without a long gap,
although there was no conversion to open surgery. Kanojia
et al. stated that long-gap atresia with a gap greater than the
vertebral length was difficult to handle when creating an
anastomosis (9). Nguyen et al. also reported that long gap
defects require more extensive dissection and difficult anas-
tomosis, and are therefore associated with longer operative
times (10). Our results were comparable with these reports.
In the three cases with a long gap, the combination of an
extensive dissection and difficult anastomosis resulted in a
longer operative time. Specific esophageal elongation pro-
cedures, such as Livaditis and Foker procedures, were not
performed in this series. When the three cases of long gap
were excluded, we were able to observe a steady decline
in operative time, although there was no change in postop-
erative complications. Our analysis clearly showed a learn-
ing curve that fit a logarithmic function curve well, which
seems to be reasonable for the learning curve analysis for
all other types of surgical skills. Nachulewicz et al. reported
that the learning curve was obviously visible in eight
successive cases, although the duration of the operation
did not diminish markedly (11). Lee et al. also showed that
the initial period of the learning curve was evident in their
earlier 13 cases of thoracoscopic EA/TEF repair (7).
Although the number of patients in the present study did
not seem sufficient for a learning curve analysis, our results
were comparable with their previous reports.
The longer operative time in the early stage may be
due to earlier problems encountered because of a lack of
experience in the optimal placement of trocars, the
surgeon’s lack of familiarity with the procedure, and
difficulties in anesthetic management. Therefore, sharing
the initial experience through a single surgeon’s
4 © 2017 Japan Society for Endoscopic Surgery, Asia Endosurgery Task Force and John Wiley & Sons Australia, Ltd
H Okuyama et al.
supervision may help rapidly reduce the operative time
Incidence for subsequent procedures. Overall, there were no mortal-
0% (0/11)
ities and no conversions to open thoracotomy in our series.
0% (0/11) The incidence of anastomosis stricture requiring dilatation
0% (0/11) in our series was 9% (1/11), which is less frequent than in
9% (1/11) previous reports in larger series (3,12,13). None of our 11
0% (0/11) cases experienced any recurrence of TEF. As we routinely
18% (2/11) used surgical clips in all of our cases, the method of fistula
0% (0/11)
closure may not influence the incidence of TEF recurrence.
0% (0/11)
Our results suggest that TR of EA/TEF is feasible and safe
with considerable superiority to conventional open thora-
cotomy. However, there is a learning curve and achieving
acceptable results requires advanced endoscopic surgical
skill.
In conclusion, there is a considerable learning curve
within initial experiences performing TR of EA/TEF.
Because of this learning curve, less experienced surgeons
should perform the procedure under the guidance of an
experienced endoscopic surgeon.
Acknowledgments
The authors have no conflict of interest to disclose.
References
1. Rothenberg SS Thoracoscopic repair of a tracheoesophageal in
a neonate. Pediatr Endosurg Innovative Tech 2000; 4: 150–156.
2. Yang YF, Dong R, Zheng C et al. Outcomes of thoracoscopy
versus thoracotomy for esophageal atresia with tra-
cheoesophageal fistula repair: A PRISMA-compliant systematic
review and meta-analysis. Medicine (Baltimore) 2016; 95: e4428.
3. Holcomb GW III, Rothenberg SS, Bax KM et al. Thoracoscopic
repair of esophageal atresia and tracheoesophageal fistula: A
multi-institutional analysis. Ann Surg 2005; 242: 422–428.
4. Okuyama H, Koga H, Ishimaru T et al. Current practice and
outcomes of Thoracoscopic esophageal atresia and
tracheoesophageal fistula repair: A multi-institutional analysis
in Japan. J Laparoendosc Adv Surg Tech A 2015; 25: 441–445.
5. van der Zee DC, Tytgat SH, Zwaveling S et al. Learning curve of
thoracoscopic repair of esophageal atresia. World J Surg 2012;
36: 2093–2097.
6. Szavay PO, Zundel S, Blumenstock G et al. Perioperative
outcome of patients with esophageal atresia and tracheo-
esophageal fistula undergoing open versus thoracoscopic
surgery. J Laparoendosc Adv Surg Tech A 2011; 21: 439–443.
7. Lee S, Lee SK, Seo JM Thoracoscopic repair of esophageal atre-
sia with tracheoesophageal fistula: Overcoming the learning
curve. J Pediatr Surg 2014; 49: 1570–1572.
8. Hiradfar M, Gharavifard M, Shojaeian R et al. Thoracoscopic
esophageal atresia with tracheoesophageal fistula repair: The
first Iranian group report, passing the learning curve. J Neonatal
Surg 2016; 5: 29.
Asian J Endosc Surg •• (2017) •• –••
H Okuyama et al.
9. Kanojia RP, Bhardwaj N, Dwivedi D et al. Thoracoscopic repair
of esophageal atresia with tracheoesophageal fistula: Basics of
technique and its nuances. J Indian Assoc Pediatr Surg 2016;
21: 120–124.
10. Nguyen T, Zainabadi K, Bui T et al. Thoracoscopic repair of
esophageal atresia and tracheoesophageal fistula: Lessons
learnedJ Laparoendosc Adv Surg Tech A 2006; 16: 174–178.
11. Nachulewicz P, Zaborowska K, Rogowski B et al. Thoracoscopic
repair of esophageal atresia with a distal fistula – lessons from
Asian J Endosc Surg •• (2017) ••– ••
© 2017 Japan Society for Endoscopic Surgery, Asia Endosurgery Task Force and John Wiley & Sons Australia, Ltd
Thoracoscopic esophageal atresia repair
the first 10 operations. Wideochir Inne Tech Maloinwazyjne
2015; 10: 57–61.
12. Ron O, De Coppi P, Pierro A The surgical approach to
esophageal atresia repair and the management of long-gap
atresia: Results of a survey. Semin Pediatr Surg 2009; 18:
44–49.
13. Lugo B, Malhotra A, Guner Y et al. Thoracoscopic versus open
repair of tracheoesophageal fistula and esophageal atresia. J
Laparoendosc Adv Surg Tech A 2008; 18: 753–756.
5