Introduction
Endoscopic submucosal dissection (ESD) has become a standard treatment for early esophageal
cancer. However, ESD may cause a series of complications, such as perforation, bleeding
and stenosis. Among them, esophageal stricture is common and probably one of the most
challenging complications. The incidence of esophageal stricture after ESD is 6 %
to 18 % and soars to 100 % after removal of circumferential or near-circumferential
neoplastic lesions [1]. Regardless of the causes, endoscopic dilation (ED) using balloons or bougies is
the most common treatment for esophageal strictures and optimal treatment by default.
Esophageal stenosis after ESD is usually complex esophageal stenosis, which requires
repeated dilation [2]. For strictures that do not respond well to dilations, other treatment methods have
been reported, such as endoscopic radial incision (ERI), steroid injection, and esophageal
stent placement [3].
ERI is traditionally performed safely and effectively in patients with simple esophageal
strictures or anastomotic strictures, and is considered an alternative therapy for
patients who have relatively short strictures [4]. For esophageal strictures after ESD, the strictures are usually longer and more
angulated than anastomotic strictures. Due to the technical challenges and risks of
ERI, it is usually reserved as a second-line treatment for refractory esophageal stenosis
after ESD. Currently, there are only a few reports evaluating the role of ERI in treating
esophageal stenosis [5]
[6]
[7].
We have noticed that, after each session of ED, only the relatively “weak” and fresh
scar tissue is torn apart, whereas the whitish, older and stiffer tissue remains intact.
These “tough” scars are ideal targets for ERI, and by combining initial ED and subsequent
ERI, the narrowed esophagus can expand completely and uniformly. Here, we report our
experience with using ED + ERI, and demonstrate that compared to ED only in previous
cases, ED + ERI can reduce the number of treatment cycles and eventually shorten the
course of treatment.
Patients and methods
Patients
This pilot study included 5 patients with post-esophageal ESD who had received ED
more than 3 times but still had recurrent dysphagia. All patients underwent ED + ERI
interspersed with ED. Some patients with anastomotic stricture also received ED + ERI
but were excluded from this study. This study was performed in compliance with the
ethical principles of the Declaration of Helsinki and approved by the Ethics Committee
of Qilu Hospital of Shandong University. Informed consent was obtained from all patients
or their authorized relatives before the procedure was performed. All patients or
their authorized relatives were informed of possible adverse events and other possible
treatment options.
Endoscopic incision and dilation procedure
Conventional ED was performed without general anesthesia; ED + ERI requires general
anesthesia with endotracheal intubation. For ED, Savary-Gilliard dilators (Cook Medical)
following guidewire placement was used. Severely narrowed, angulated strictures may
require fluoroscopy guidance or even pre-incision to allow passage of the guidewire
([Fig. 1a]). Savary-Gilliard dilations followed the “rule of three”: in each treatment session,
starting from the size when moderate resistance was felt, usually 5 mm or 7 mm, then
increased by 2 mm each time until the third dilation. Active hemorrhage after dilation,
although rare, should be managed with electrocoagulation, but not metal clips, to
avoid hindering subsequent incision. Most patients will be symptom-free if their strictures
are dilated to 11 mm. ED usually resulted in 3 reddish longitudinal mucosal lacerations,
but some old white scar tissue between the lacerations remained intact ([Fig. 1b]). It seems that ED is only capable of repeatedly tearing up relatively fresh scars,
but not the older, stiffer scars. The latter type of scars may contribute to recurrence
of strictures and, hence, the target for ERI ([Fig. 1c]). We used an IT knife to cut through the scar tissue and expose the underlying muscularis
propria. Care should be taken not to injure the smooth muscle fibers ([Fig. 1d]). Scar tissue ridges between esophageal wall indents should be cut and released
as well to prevent food retention. The ED + ERI procedure resulted in complete, uniform
expansion of the esophageal lumen ([Fig. 1e], [Video 1]).
Fig. 1 Endoscopic dilation combined with incisional therapy for refractory esophageal stricture
after endoscopic submucosal dissection. a Refractory esophageal stricture after multiple sessions of endoscopic dilation after
endoscopic submucosal dissection. b Endoscopic dilation resulted in reddish longitudinal mucosal lacerations. c White, old scar tissue remained intact between the mucosal lacerations. d Radial incision of the white, old scar tissue deep into the muscularis propria layer.
e Complete expansion of the esophageal lumen after the procedure.
Video 1 Endoscopic dilation combined with incisional therapy for refractory esophageal stricture
after endoscopic submucosal dissection.
Postoperative management and follow-up
After the procedure, patients were routinely monitored for vital signs. Alarming symptoms
such as chest pain could indicate esophageal perforation. The fasting time of patients
depended on the type of treatment. Patients who underwent ED only fasted for 2 to
4 hours before resuming a liquid diet. For those who underwent ED + ERI, fasting was
required for 24 hours, during which proton pump inhibitors and prophylactical antibiotics
were administered intravenously. Patients were regularly followed up in the outpatient
department, and endoscopy was performed when dysphagia recurred. If confirmed, we
still prefer ED as a first-line treatment because it is simple and does not require
hospitalization. If ED was not optimal due to remained fibrotic scar tissue, then
additional ERI was performed. Patients were followed up 6 months after the last treatment
if they failed to receive further treatment, and then every year thereafter.
Outcome measures
Criteria for evaluating treatment methods were total treatment time, number of treatments
as well as time interval between treatments. Dysphagia score described by Knyrim,
et al. was used to grade dysphagia before surgery and during follow-up [8]. The following grading was applied: grade 0, able to eat normally; grade I, unable
to swallow some solid foods; grade II, only able to swallow semi-liquid food; grade
III, only able to swallow liquid food; grade IV, unable to swallow liquid food. Relief
of symptoms was defined as a dysphagia score < 2 or no stenosis confirmed by endoscopy.
Statistical analysis
SPSS26.0 was used for statistical analyses and for illustrations. Measurement data
were expressed as median ranges.
Results
Baseline characteristics and treatment parameters of the 5 patients with refractory
stenosis are shown in [Table 1]. The patients underwent 11 ED sessions (range, 4–28) in 322 days (range, 246–584)
before being referred to our center; the time interval between ED was typically within
1 month, except for the last patient. 2 or 3 sessions of ED + ERI were performed on
each patient interspersed with additional ED at local hospitals. After 4 treatments
(range, 2–9) in 188 days (range, 98–277), all patients were symptom-free. The first
and last patient also benefited from an elongated time interval between treatments
(21 to 69 days and 82 to 203 days, respectively). No complications such as major bleeding
or perforation occurred. The median follow-up time was 688 days (range, 188–1255),
and all 5 patients were symptom-free. Therefore, ED + ERI is a feasible method to
treat refractory esophageal stenosis after ESD.
Table 1
Baseline characters and treatment parameters of five patients with refractory esophageal
stenosis after ESD.
|
Case
|
Sex
|
Age
|
|
|
Before dilation + incision
|
|
|
After dilation + incision
|
|
Time (days)
|
Number
|
Interval (days)[1]
|
Dysphagia score[2]
|
Time (days)
|
Number
|
Interval (days)
|
Dysphagia score
|
|
1
|
F
|
64
|
584
|
28
|
21
|
3
|
277
|
5
|
69
|
0
|
|
2
|
M
|
66
|
246
|
10
|
25
|
3
|
188
|
9
|
24
|
0
|
|
3
|
M
|
62
|
267
|
12
|
22
|
4
|
98
|
4
|
33
|
0
|
|
4
|
F
|
70
|
322
|
11
|
29
|
3
|
100
|
4
|
33
|
1
|
|
5
|
F
|
65
|
329
|
4
|
82
|
3
|
203
|
2
|
203
|
0
|
1 Before dilation + incision, time interval between treatments was calculated by treatment
time/number of treatments, because patients were referred to our center when dysphagia
recurred after the last time of treatment. After dilation + incision, time interval
between treatments was calculated by treatment time/(number of treatments-1).
2 Dysphagia score originally described by Knyrim et al. in the New England Journal of Medicine in 1993. Grade 0: able to eat normally; grade I: unable to swallow some solid foods;
grade II: only able to swallow semi-liquid food; grade III: only able to swallow liquid
food; grade IV: unable to swallow liquid food.
Discussion
ERI is a second-line treatment for esophageal stenosis and has shown good safety and
efficacy, mainly in patients with simple stenosis or anastomotic strictures [4]. ERI can be conventionally performed by making radial incisions on fibrotic tissue,
or in recent years, radial incision and cutting (RIC) to remove the scar tissue [4]
[7]
[9]. However, RIC leaves a circumferential wound that resembles the post-ESD wound,
raising concerns for further strictures if applied to treat long-segment, angulated
strictures after ESD. Also, because it is difficult to inject medium beneath scar
tissue, dissection can be arduous and risky. Nevertheless, there have been a few reports
about using ERI or RIC to treat refractory esophageal stenosis after ESD, all with
favorable results, although with relatively high perforation rate, demonstrating the
feasibility of this technique [5]
[7]
[9].
Our method, for the first time, integrates ED and ERI, leveraging the advantages of
the 2 techniques. ED is used for initial dilation and ERI is used for subsequent,
targeted incision of stiff scars that remain after ED to achieve complete expansion
of the esophagus. In general, although there were only 5 cases, patients seemed to
benefit more when ED + ERI was more frequently performed. Basic science research has
demonstrated that mechanical forces play an important role in the fibrosis process:
A constant high level of tissue matrix tension promotes fibrosis progression whereas
relieving mechanical tension can reduce fibrosis and scar formation [10]. Therefore, the stiff scars left by ED may contribute to subsequent scarring by
maintaining high tension locally, and removing such scars can provide a tension-less
environment to promote normal healing. This could be a theoretical explanation for
the effect of ED + ERI as well as the established RIC. The fact that ED + ERI enabled
longer symptom-free time between treatments supports this hypothesis.
Compared with ED, ED + ERI still has certain limitations, such as the need for general
anesthesia, difficulty of the procedure, risk of perforation, need for hospitalization
after operation, and high cost. Therefore, ED + ERI may be particularly beneficial
for esophageal strictures that are at high risk of becoming refractory, such as circumferential,
long-segment lesions, injury to the muscularis propria during ESD or even perforation.
Conclusions
The current pilot study has demonstrated the feasibility and potential of ED + ERI.
Well-designed, prospective cohorts with more patients are needed to better elucidate
the benefits or disadvantages of this procedure.
