Keywords
Endoscopy Upper GI Tract - Endoscopic resection (ESD, EMRc, ...) - Pediatric endoscopy
- Benign strictures
Introduction
Endoscopic treatment is widely used for superficial esophageal cancer. Endoscopic
treatment progressed from endoscopic mucosal resection to endoscopic submucosal dissection
(ESD), drastically improving the en bloc resection rate [1]
[2]. Intramucosal carcinoma, which has a low risk of lymph node metastasis, can be cured
via endoscopic resection of the lesion, enabling treatment while maintaining the patient’s
quality of life. Stenosis, scarring, or lesions in the cervical esophagus are the
major difficulties in esophageal ESD, among various difficulties that have been reported
[3]
[4]
[5].
Owing to their minimal invasiveness, ultrathin endoscopes are primarily used for upper
gastrointestinal endoscopy [6]. The image resolution is improved, and the lesion detection rate is comparable to
conventional endoscopy [7]. Nevertheless, ultrathin endoscopes have not been used in therapeutic endoscopy
because dedicated treatment devices are unavailable [8]
[9]. However, ultrathin endoscopes have an advantage in therapeutic endoscopy as they
can access narrow areas. This device is particularly advantageous in esophageal ESD,
where a narrow visual field is associated with treatment difficulty due to stenosis
or scarring. We collaborated with companies to develop two ultrathin endoscope devices
to enable ESD and used them in clinical practice [10]
[11]. This study retrospectively investigated the usefulness of ultrathin endoscopes
in esophageal ESD.
Patients and methods
We performed esophageal ESD on 220 patients with 303 lesions in Toranomon Hospital
from February 2021 to February 2023. In these cases, ESD with a conventional endoscope
was difficult. Ultrathin endoscopes were used for ESD in 26 lesions in 23 patients.
An ultrathin endoscope was used when a conventional endoscope could not visualize
the lesion or the site of dissection. Therefore, an ultrathin endoscope was used because
starting with a conventional endoscope approach was impossible due to stenosis or
curvature of the cervical esophagus.
When an adequate approach to the dissection area is impossible because of a condition
such as scarring, treatment is usually performed with a conventional endoscope followed
by an ultrathin endoscope. This study retrospectively investigated the safety and
efficacy of ultrathin endoscopes in esophageal ESD. The Ethics Committee of Toranomon
Hospital approved this study.
Device
A transparent hood (Nichendo) for ultrathin endoscopes was jointly developed with
Yasui Co., Ltd. (Miyazaki, Japan), and an ESD knife (Endosaber Fine) for ultrathin
endoscopes was jointly developed with Yamashina Seiki Co., Ltd. (Shiga, Japan).
The cylindrical transparent hood has an outer diameter of 5.9 mm and is made of elastomer.
The maximum outer diameter of the ESD knife was 1.95 mm. These devices were approved
and clinically introduced in January 2021.
Conventional endoscopes include an Olympus GIF H290T (maximum diameter 11.8 mm) or
a Fujifilm EG-L580RD7 (maximum diameter 11.8 mm). Ultrathin endoscopes include an
Olympus GIF 1200N (maximum diameter 5.8 mm), an Olympus GIF 290N (maximum diameter
5.8 mm), or a Fujifilm EG-L580NW7 (maximum diameter 5.9 mm). ESD was performed beyond
the stricture after pharyngeal chemoradiotherapy ([Fig. 1], [Fig. 2], and [Video 1]).
Fig. 1 Images of devices that have been developed and clinically introduced. a Transparent hood for ultrathin endoscope (Nichendo). b Endoscopic submucosal dissection knife for ultrathin endoscope (Endosaber Fine).
Fig. 2 A conventional endoscope could not pass through the stenosis after CRT.
ESD procedure using an ultrathin endoscope on a lesion with a pharyngeal stricture.Video
1
Skilled endoscopists with board certification from the Japan Gastroenterological Endoscopy
Society performed all ESD procedures in this study. Furthermore, the method of esophageal
ESD was similar to the conventional method and is described briefly. First, iodine
staining was used to diagnose the border ([Fig. 3]). Then, markings were placed approximately 2 mm outside the lesions with a soft
coagulation effect 4, 50 W with a knife tip ([Fig. 4]). Glycerol was injected outside the marking, and an incision was made with an endocut
effect of 1, duration of 1, and interval of 1 ([Fig. 5]). Afterward, dissection was performed with swift coagulation effect 3, 40 W while
observing the submucosal layer to be dissected with a transparent tip hood ([Fig. 6]), and the lesion was resected en bloc ([Fig. 7], [Fig. 8]). Intraoperative hemorrhage was coagulated with a knife tip with swift coagulation
effect 3, 40 W. Hemostatic forceps were used (Pentax or Kaneka) with soft coagulation
effect 4, 50 W when hemostasis could not be obtained. This study used a VIO 300D high-frequency
apparatus (ERBE, Tübingen, Germany).
Fig. 3 A lesion was identified as an iodine-unstained area.
Fig. 4 A marking was placed 2 mm outside the lesion.
Fig. 5 An anal side incision was performed.
Fig. 6 The submucosal layer was dissected using the pocket creation method.
Fig. 7 The lesion was resected en bloc.
Fig. 8 The pathological diagnosis was squamous cell carcinoma, pT1a-LPM, 25 × 18 mm, ly0,
v0, margin negative.
ESD was initiated with an ultrathin endoscope for lesions on the anal side of the
stenosis or the cervical esophagus when a lesion could not be approached with a conventional
endoscope. Furthermore, ESD with a conventional endoscope was combined with an ultrathin
endoscope when the dissected layer could not be visualized due to fibrosis during
dissection.
A local injection or an oral steroid was administered at operator discretion to prevent
postoperative stenosis. Fluid replacement was performed while the patient was fasting
on the day of ESD. Examination, x-ray, and blood testing were performed the day after
ESD and meals were started at operator discretion.
Evaluation items
Short-term outcomes evaluated included intraoperative perforation, muscle layer injury,
procedure time, intraoperative bleeding, postoperative bleeding, pneumonia, and fever.
Outcomes such as postoperative stenosis and recurrence were evaluated as long-term
factors. The procedure time required from the start of the oral side incision to pocket
creation in cases using the pocket creation method was measured by analyzing video
images.
Results
This study included 23 patients (19 males and four females, [Table 1]). Ultrathin endoscopes were used in 8.6% (26/303) of all esophageal ESD procedures
during the same period. The average age of the patients was 72.8 ± 8.0 years. Eight
of the 26 ESD procedures were completed with an ultrathin endoscope and 18 with an
ultrathin and a conventional endoscope. In nine of those 18 cases, ultrathin endoscopes
were used first due to the stenosis, and then conventional endoscopes were used in
combination with an ultrathin endoscope. In the remaining nine cases, conventional
endoscopes were used first, and then ultrathin endoscopes were used due to scarring.
ESD procedures were performed under general anesthesia in 18 cases and under conscious
sedation with sedatives and analgesics in eight.
Table 1 Patient characteristics.
|
SD, standard deviation.
|
|
Number of patients
|
23 cases
|
|
Age, years (± SD)
|
72.8 ± 8.0
|
|
Gender (male:female)
|
19:4
|
|
History of pharyngeal cancer, % (n/n)
|
56.5 (13/23)
|
|
History of esophageal cancer, % (n/n)
|
90.0 (21/23)
|
Lesions were found on the anal side of the stenosis in 38.6% (10/26) and with scarring
in 50% (13/26). Furthermore, 46.2% of patients (12/26) had lesions in the cervical
esophagus ([Table 2]).
Table 2 Lesion characteristics.
|
SD, standard deviation.
|
|
Number of lesions
|
26
|
|
Tumor size (mm ± SD)
|
18.8 ± 14.1
|
|
Specimen size (mm ± SD)
|
32.6 ± 13.2
|
|
Lesions on the anal side of the stenosis, % (n/n)
|
38.6 (10/26)
|
|
Lesions with scarring, % (n/n)
|
50 (13/26)
|
|
Lesions on the cervical esophagus, % (n/n)
|
46.2 (12/26)
|
En bloc resection was possible for all 26 lesions. Serious complications, such as
intraoperative perforation, pneumonia, and intraoperative bleeding, were not observed.
Muscle layer damage was observed in seven lesions, but all could be followed up conservatively.
Stenosis was prevented with corticosteroids in three cases. These patients only had
mild stenosis and balloon dilatation was not required. Balloon dilatation was required
in one patient with stenosis in whom steroids were not used.
The average size of the specimens was 32.6 ± 13.2 mm and the invasion depth was T1a-MM
in one lesion; all others were T1a-epithelial or lamina propria. Five lesions had
positive or ambiguous lateral margins and all had negative vertical margins. Total
procedure time was 64.1 ± 37.7 minutes. The pocket creation method was used in 53.8%
of cases (14/26). The average time required from the start of oral incision to pocket
creation was 121.2 ± 109.9 seconds ([Table 3]).
Table 3 Study results.
|
ESD, endoscopic submucosal dissection; SD, standard deviation.
|
|
En bloc resection rate, % (n/n)
|
100 (26/26)
|
|
R0 resection rate, % (n/n)
|
80.8 (21/26)
|
|
Procedure time (min ± SD)
|
62.2 ± 36.7
|
|
Perforation, % (n/n)
|
0 (0/26)
|
|
Muscle injury, % (n/n)
|
26.9 (7/26)
|
|
Massive bleeding, % (n/n)
|
0 (0/23)
|
|
Delayed bleeding, % (n/n)
|
0 (0/23)
|
|
Pneumonia, % (n/n)
|
0 (0/23)
|
|
Stenosis after ESD, % (n/n)
|
4.3 (1/23)
|
Discussion
Squamous cell carcinoma (SCC) has been identified as cancer with a high metachronous
carcinogenesis rate. The incidence of metachronous carcinogenesis 5 years after ESD
for pharyngeal and esophageal SCC is approximately 20% [12]
[13]. Careful endoscopic follow-up is important, but patient prognosis is excellent if
the disease is detected early. We usually perform follow-up endoscopys twice a year
in patients with a history of SCC. All of the patients in this study had a history
of pharyngeal or esophageal SCC that was discovered during follow-up endoscopy.
Esophageal stenosis and narrowing of the lumen occasionally is seen in patients and
the causes are various. Stenosis is a major clinical problem, especially after extensive
ESD, and various methods to prevent stenosis have been reported [14]
[15]
[16]
[17]
[18]. Among them, steroids are currently the most widely used. Nevertheless, we have
often seen patients for whom balloon dilatation was required. Stenosis also can occur
after radiation therapy for esophageal and pharyngeal cancer. Radiation-induced stenosis
is more severe than post-ESD stenosis, and dilation may not resolve the stenosis.
Esophageal ESD can be complicated by various factors, one of which is stenosis, which
is the major factor impacting procedure success. Furthermore, lesions from post-ESD
scarring and cervical esophageal lesions are sometimes difficult to treat with normal
ESD. All of our patients had at least one of these complications. An ultrathin endoscope
may not be necessary when balloon dilatation is performed, and treatment with a conventional
endoscope is possible. However, the lesion will be damaged if balloon dilation is
performed on the anal side of the stenosis with a normal diameter device. An ultrathin
endoscope has great advantages if a conventional endoscope cannot pass through the
balloon dilation because of severe stenosis.
Moreover, dissecting the scar under direct vision using a conventional endoscope is
sometimes difficult. Even in such a case, an ultrathin endoscope can be used in a
narrow space in the scar to perform dissection under direct vision. Furthermore, lesions
in the cervical esophagus are sometimes difficult to approach using a conventional
endoscope. An ultrathin endoscope approach is possible in such cases. Subsequent treatment
is relatively easy if the submucosal space can be entered using an ultrathin endoscope
using the pocket creation method [19]. Approximately half of the patients in this study underwent ESD of the cervical
esophagus. Ultrathin endoscopes are useful for ESD in the narrow and curved cervical
esophagus.
This study has some limitations, as does ESD with an ultrathin endoscope. First, this
study was not a comparison study with a conventional endoscope. ESD with conventional
endoscopes was performed as much as possible and ultrathin endoscopes were used only
when treatment with conventional endoscopes was challenging. In the future, we will
further clarify the criteria for using ultrathin endoscopes and conduct randomized
controlled trials to verify the usefulness of ultrathin endoscopes. An ultrathin endoscope
is more unstable than a conventional endoscope because of its thinness and softness.
However, the pocket creation method was used as much as possible because the endoscope
provides stability when entering the narrow submucosal space. A pocket can be created
quickly using an ultrathin endoscope, which is considered a great advantage.
Furthermore, an ultrathin endoscope in an ESD procedure is difficult to aspirate and
waterjets cannot be used. ESD using an ultrathin endoscope is more difficult than
with a conventional endoscope. Massive bleeding during an ESD procedure can be a difficult
complicatoin. Hence, we attempted to perform ESD with a conventional endoscope as
much as possible. Operators who perform ESD with an ultrathin endoscope mst have a
higher level of skill. Using an ultrathin endoscope does not eliminate all complications.
Ultrathin endoscopes are considered suitable for ESD because the procedure causes
relatively little bleeding. An ultrathin endoscope with a large forceps diameter and
a waterjet function should be developed.
Conclusions
We developed two devices that enable ESD with an ultrathin endoscope and introduced
them for clinical use. They were used for esophageal ESD 2 years after their introduction.
ESD is usually difficult to perform with a conventional endoscope. The ultrathin endoscope
was used for 8.6% of all esophageal ESDs. The procedure took a long time, but en bloc
resection was safely performed in all cases. Ultrathin endoscopes may be useful for
difficult cases of esophageal ESD.
