Endoscopy 2015; 47(11): 972-979
DOI: 10.1055/s-0034-1392558
Original article
© Georg Thieme Verlag KG Stuttgart · New York

Learning endoscopic resection in the esophagus

Frederike G. I. van Vilsteren
1   Department of Gastroenterology, Academic Medical Center, Amsterdam, The Netherlands
,
Roos E. Pouw
1   Department of Gastroenterology, Academic Medical Center, Amsterdam, The Netherlands
,
Lorenza Alvarez Herrero
1   Department of Gastroenterology, Academic Medical Center, Amsterdam, The Netherlands
2   Department of Gastroenterology, St Antonius hospital, Nieuwegein, The Netherlands
,
Raf Bisschops
3   Department of Gastroenterology, University Hospitals Leuven, Belgium
,
Martin Houben
4   Department of Gastroenterology, Haga Teaching Hospital Den Haag, The Netherlands
,
Frans T. M. Peters
5   Department of Gastroenterology, University Medical Center Groningen, The Netherlands
,
B. E. Schenk
6   Department of Gastroenterology, Isala Klinieken, Zwolle, The Netherlands
,
Bas L. A. M. Weusten
2   Department of Gastroenterology, St Antonius hospital, Nieuwegein, The Netherlands
,
Erik J. Schoon
7   Department of Gastroenterology, Catharina Ziekenhuis Eindhoven, The Netherlands
,
Jacques J. G. H. M. Bergman
1   Department of Gastroenterology, Academic Medical Center, Amsterdam, The Netherlands
› Author Affiliations
Further Information

Corresponding author

Jacques Bergman, MD PhD
Department of Gastroenterology and Hepatology
Academic Medical Center
Meibergdreef 9
1105 AZ Amsterdam
The Netherlands   
Fax: +31-20-6917033   

Publication History

submitted: 05 February 2014

accepted after revision: 01 April 2015

Publication Date:
11 September 2015 (online)

 

Background: Endoscopic resection is the cornerstone of endoscopic management of esophageal early neoplasia. However, endoscopic resection is a complex technique requiring knowledge and expertise. Our aims were to identify the most important learning points in performing endoscopic resection in a training setting and to provide information on how to improve endoscopic resection technique.

Methods: Six gastroenterologists at centers with multidisciplinary expertise in upper gastrointestinal oncology participated in a structured endoscopic resection training program, consisting of four training days with lectures and hands-on training on live pigs, further one-to-one hands-on training days, and written feedback (by an expert) on videos of unsupervised endoscopic resection procedures. The first 20 endoscopic resections of each participant were prospectively registered. Ninety learning points were independently identified by participants using a standardized questionnaire and by an expert providing written feedback on 33 unsupervised endoscopic resection videos. Three expert endoscopists selected and ranked the most important learning points in a consensus meeting.

Results. The top 10 tips (illustrated by unique videos of three perforations) were: (1) allow time for inspection and use a high-definition endoscope; (2) create a preprocedural plan by placing electrocoagulation markings; (3) know the management of bleeding; (4) optimize the endoscopic view by repeatedly cleaning out stomach and target area; (5) use a therapeutic endoscope during resection; (6) always perform a test suction; (7) keep instruments close to the tip; (8) lift edges in piecemeal endoscopic cap resections; (9) know the management of perforation; (10) pin specimens down.

Conclusions: This study summarized the most important learning points for performing endoscopic resection encountered during a structured endoscopic resection training program.


#

Introduction

Endoscopic resection is considered the cornerstone of endoscopic management of high grade dysplasia (HGD) and early cancer in the esophagus. Endoscopic resection not only removes the neoplastic lesion, but also provides a specimen for histological assessment and staging. This allows for the selection of suitable candidates for subsequent endoscopic treatment, and defines which patients should be referred for surgery because of an increased risk of lymph node involvement. Several studies have shown that endoscopic resection is safe and effective for the removal of focal early neoplastic lesions in Barrett’s esophagus as well as for early esophageal squamous neoplasia [1] [2] [3] [4] [5] [6] [7].

Endoscopic treatment of early neoplasia in the upper gastrointestinal (GI) tract is becoming more accepted within the GI community because an increasing number of studies have shown that endoscopic treatment of early neoplastic lesions is safe and effective [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]. In addition, we expect there to be an increase in the number of early neoplastic lesions being detected because of ongoing developments in endoscopic imaging techniques and an increase in screening and surveillance programs. Furthermore, the newest generation endoscopic treatment modalities, such as endoscopic resection with multiband mucosectomy (MBM), are easy to use, making them more attractive to a wider range of endoscopists. Therefore, we anticipate that endoscopic treatment is likely to become more widely available in both expert and non-expert centers.

Endoscopic resection is, however, a technically demanding procedure that requires knowledge and expertise to ensure complete removal of the lesion, avoid complications such as perforation, and manage the intraprocedural bleeding that occurs in 20 % – 25 % of cases. Endoscopists who perform endoscopic resection should therefore be trained in the technicalities of the procedure, as well as in the prevention and management of complications [5] [6] [11].

In view of this, we developed a structured endoscopic resection training program to implement endoscopic resection in a safe and controlled manner in the Netherlands. In this training program, six endoscopists from selected centers were trained in endoscopic resection, and all procedures and complications were registered prospectively to assess the safety and efficacy of endoscopic resection in a training setting, as well as to identify potential learning-curve effects. The clinical results of this study have been previously published [12].

During the endoscopic resection training program, numerous learning points, mistakes, and difficulties of endoscopic resection were identified, which may be highly informative for endoscopists who perform endoscopic resection or who are training in endoscopic resection. The aim of the current study was to identify important learning points in endoscopic resection as pointed out by the participating endoscopists in the endoscopic resection training program and the expert endoscopists who conducted the training. Based on the learning points that were identified, we aimed to provide our “top 10” practical recommendations for the improvement of endoscopic resection skills.


#

Methods

The endoscopic resection training program

The organizational background to this study was the Dutch endoscopic resection training program, which aimed to implement endoscopic resection in the Netherlands including the proper endoscopic work-up and the histological assessment of endoscopic resection specimens in accordance with national guidelines [13]. Other objectives of the endoscopic resection training program were: (i) to improve the quality of endoscopic detection and treatment of early neoplastic lesions and the histological assessment of endoscopic resection specimens; (ii) to implement prospective registration of procedures and complications, and to agree mutual treatment protocols; (iii) to build a platform for scientific research and guideline development.

The training program was developed and managed at the Academic Medical Center, Amsterdam by a committee of two endoscopists, two pathologists, two endoscopy nurses, two research fellows, and a research nurse, all of whom had extensive experience in the field of endoscopic treatment of early esophageal neoplasia. The endoscopic resection training program was funded by the organizing center (33 %), participating centers (33 %), and sponsors (33 %).

Participants in the training program were selected to fit the following profile: they were all fulltime gastroenterologists in large regional or academic centers with established multidisciplinary expertise in oncology of the upper GI tract (high-volume center for upper GI surgery, with availability of endoscopic ultrasound [EUS], histopathological expertise, and facilities for radiotherapy and oncology care). Endoscopists participated in the training program together with an endoscopy nurse and a pathologist from their own center.

The training program consisted of four training days at 3-monthly intervals consisting of theoretical lectures, live demonstrations, and hands-on training on anaesthetized pigs under the guidance of international expert endoscopists and pathologists. Additionally, individual hands-on training days were scheduled at the participants’ centers and at the training site during which six to eight endoscopic resection procedures were performed with one-to-one training being provided for the endoscopist and endoscopy nurse.

During the training program, endoscopic resection was performed with the endoscopic resection cap technique or the MBM technique, as previously described in detail [5] [6] [14]. Participating endoscopists were encouraged to make digital video recordings of the unsupervised endoscopic resection procedures they performed at their own centers. One expert endoscopist from the training team (J. B.) reviewed all video recordings and provided a written feedback report for each recorded endoscopic resection procedure. Furthermore, these video recordings (including three during which perforations occurred) served as teaching material during the course. In addition, endoscopic resection specimens from the first 20 endoscopic resection procedures performed at the participants’ own centers were reviewed by a pathologist at the training site who had extensive experience in endoscopic resection [12].

All endoscopic resection procedures performed during the endoscopic resection training program were prospectively registered, using standardized case registration forms to document the procedure characteristics and any complications (www.endosurgery.eu).


#

Identification of learning points

Learning points in performing endoscopic resection were collected from two different sources:

(1) A questionnaire was sent to the six participating endoscopists after each had performed at least 20 endoscopic resection procedures. Participating endoscopists were asked to describe a maximum of three learning points for each of the following phases of the endoscopic resection procedure: work-up and imaging; marking of the lesion; submucosal fluid injection using the endoscopic resection cap technique and suctioning of the lesion into the cap; resection; the MBM technique; the endoscopic resection cap technique; retrieval of the specimens; complications; and miscellaneous.

(2) For 33 unsupervised endoscopic resection procedures (which were among the first 20 unsupervised endoscopic resection procedures for each of the participating endoscopists), video recordings were available with corresponding written feedback reports. In the feedback reports, an expert endoscopist from the training team (J.B.) provided technical feedback to the participating endoscopist. These feedback reports were subsequently reviewed by a research fellow, who extracted and categorized all situations that could be considered a learning point within the text.


#

Selection of important learning points

A list containing all learning points from these two sources (questionnaires and video feedback reports) was created by a research fellow (F.V.) and sorted using the same subheadings as were used on the questionnaires: work-up and imaging; marking of the lesion; submucosal fluid injection (endoscopic resection cap technique) and suctioning of the lesion into the cap; tissue resection; MBM technique; endoscopic resection cap technique; retrieval of the specimens; complications; and miscellaneous.

This list was then reviewed and appraised in a consensus meeting of three endoscopists who were all experienced in endoscopic resection (qualitative evaluation). The most important learning points in performing endoscopic resection from this list were selected. “Important” was defined as having influence on the efficacy, safety, or practicality of endoscopic resection.


#

Compilation of the top 10 tips

Three endoscopists who were experts in endoscopic resection were asked to compile their personal top 10 tips from the list of important learning points with number 1 corresponding to the most important and number 10 to the least important of their top 10. All items that appeared in a personal top 10 list were scored according to the rank given by the experts. The scores for each item were then added together and divided by the number of times the item appeared in the lists to give a mean score. Items with no points were excluded and the remaining items were ranked (the lowest number corresponding to highest importance).


#

Statistics

The SPSS statistical software package (SPSS Inc.16.0.2, Chicago, Illinois, USA) was used for data analysis. For descriptive statistics, mean (± SD) was used where there was a normal distribution and median (interquartile range [IQR] or range) was used in case of a skewed distribution.


#
#

Results

Endoscopic resection training program and participants

Six teams consisting of an endoscopist, endoscopy nurse, and pathologist attended all four training days. The baseline experience in esophageal endoscopic resection of the endoscopists varied from 0 – 26 endoscopic resection procedures. None of the endoscopists had participated in a previous endoscopic resection training course. All endoscopists were experienced in interventional endoscopy including endoscopic retrograde cholangiopancreatography (ERCP). Participating endoscopists had a median of five (IQR 2 – 7) one-to-one hands-on training days, supervised by an expert endoscopist.

Of their first 20 endoscopic resection procedures performed by each of the endoscopists, 52 (43.3 %) were supervised by an expert endoscopist. A median of four (IQR 2 – 6) video recordings of unsupervised endoscopic resection procedures per endoscopist were reviewed and appraised by a single expert endoscopist from the training program committee, resulting in a total of 33 written feedback reports ranging from a third of a page to three pages in length. All six endoscopists completed the questionnaire on learning points for performing endoscopic resection [12].


#

Endoscopic resection procedures

Of 120 consecutive esophageal endoscopic resection procedures, 109 were performed in Barrett’s esophagus and 11 for squamous neoplasia, and there were 85 piecemeal endoscopic resections (median of three specimens [IQR 2 – 4]). Details of the endoscopic resection procedures have been described in detail elsewhere [12]. In summary, the endoscopic resection cap technique was used in 85 procedures and the MBM technique in 35 procedures ([Fig. 1] and [Fig. 2]).

Zoom Image
Fig. 1 Endoscopic resection using the endoscopic resection cap technique showing: a lifting of the submucosa; b suctioning of the lesion into the cap, with the resection snare placed in the rim of the cap; c closure of the snare during suctioning; d the resection wound after using coagulation to cut off the pseudopolyp and thereby remove the lesion.
Zoom Image
Fig. 2 Endoscopic resection using the multiband mucosectomy (MBM) technique showing: a suctioning of the lesion into the cap; b release of a rubber band during suctioning to create a pseudopolyp containing the lesion; c resection of the pseudopolyp with the hexagonal MBM snare; d the resection wound after using coagulation to cut off the pseudopolyp and thereby remove the lesion.

Complete endoscopic removal was achieved in 111 /120 (92.5 %) cases ([Fig. 3]). Acute complications included six perforations (5.0 %), of which five were effectively treated endoscopically (clips, covered stent). The sixth patient underwent esophagectomy the same day. Four perforations occurred during unsupervised procedures, and of these three were recorded on video ([Videos 1 – 3]). There were 11 episodes of intraprocedural bleeding (9.2 %), all of which were managed endoscopically [12].

Zoom Image
Fig. 3 Endoscopic resection of early neoplasia in Barrett’s esophagus showing: a a macroscopic type 0-IIa + IIc lesion (Paris Classification [15]) of 15 mm in size in a C0M2 Barrett’s esophagus; b the resection wound after complete endoscopic removal of the lesion; c the endoscopic resection scar 6 weeks after the procedure.

#

Identification of learning points

The six participating endoscopists described a total of 59 learning points on the study questionnaires. Additionally, 43 learning points were identified by reviewing the written feedback reports on the 33 endoscopic resection procedures performed by the participating endoscopists that were captured on video. After 12 duplicate items had been deleted, 90 learning points were identified and listed ([Table e1], available online; [Figs. 4 – 7]; [15] [16]).

Chromoendoscopy with Lugol dye in a patient with a type 0-IIb + IIa lesion of the distal squamous esophagus (later shown to be squamous cell carcinoma) showing a lesion covering the circumference from 10 o’clock to 8 o’clock. A perforation is caused by a complete “red-out” and uncontrolled suctioning while resecting a small tissue bridge between two resections that had been inadequately lifted. This perforation was subsequently managed by surgical esophagectomy because there was residual carcinoma.

Endoscopy in a patient with a type 0-IIb + IIc lesion in Barrett’s esophagus containing high grade dysplasia. A perforation is caused by: (i) impaired visibility due to fluids and foaming; and (ii) suction being applied in a position too proximal and to the left of the delineated area, where the mucosa had not been lifted. This perforation was subsequently managed endoscopically with a covered stent.

Endoscopy in a patient with a type 0-IIa lesion of the squamous epithelium in a narrowed esophagus (later shown to be squamous cell carcinoma). Blind introduction of the endoscope into the esophagus after resection results in perforation through the endoscopic resection wound into the mediastinum. This perforation was subsequently managed endoscopically by closure with clips.

Table e1

The 90 learning points for performing endoscopic resection drawn from the experience of six endoscopists participating in an endoscopic resection training program (bold entries were classed as important learning points).

Topic

Learning point

Importance

Source (number of times reported)

Work-up and imaging

Use sedation at every endoscopy for Barrett’s esophagus

Sedation promotes patient comfort and compliance, and is convenient for the endoscopist; during each endoscopy for Barrett’s esophagus, sufficient time is required for meticulous inspection and biopsies

Participant questionnaire (1)

Always use antifoam

Use antifoam either administered orally to the patient prior to the endoscopy or in the water used for flushing; the reduction in foam and bubbles improves the endoscopic view

Participant questionnaire (1)

Time the inspection of your target area with the movements of respiration and peristalsis

Continue inspection in between movements; take several (sharp) images and/or make a video to improve inspection when there is a lot of movement

Participant questionnaire (1)

Insufflate an adequate amount of air

Insufflate air until there is a good view in the esophagus; if the patient is not able to hold the air, it may be that they are too deeply sedated

Participant questionnaire (1)

Do not sedate the patient too deeply: this may lead to difficulty inflating the esophagus and result in a reduced endoscopic view

If the patient is deeply sedated, he/she may not be able to hold the air and the esophagus will not expand, making inspection of the resection area difficult

Video feedback report (2)

Optimize the endoscopic view by repeatedly cleaning out the stomach, esophagus, and the target area

At the start of the endoscopic resection procedure, the endoscopist should empty the stomach and clean the target area in the esophagus by flushing with water (and using antifoam) to remove mucus and gastric contents; during the procedure, systematic emptying of the stomach and esophagus should be repeated frequently (e. g. with each insertion of an instrument) to ensure an optimal view of the target area and prevent aspiration during the endoscopy

Participant questionnaire (5)/ video feedback report (8)

Do not compromise inspection and delineation of the lesion – allow sufficient time for inspection and use a high-definition endoscope

Allow time for meticulous inspection of the lesion; optimize the inspection by using a high-definition endoscope equipped with a (virtual) chromoendoscopy technique (most therapeutic gastroscopes have an inferior image quality compared to their diagnostic counterparts)

Participant questionnaire (1)/ video feedback report (1)

Use an endoscope with a separate water-jet channel

After imaging, switch to an endoscope with a separate water-jet channel to optimize cleaning and imaging of the target area and enable simultaneous flushing of water and passage of instruments through the working channel

Video feedback report (1)

Use advanced imaging techniques

The use of advanced imaging techniques may allow a more detailed inspection of the esophagus and the lesion (however, a critical look with high-resolution white-light endoscopy may be superior to advanced imaging techniques)

Participant questionnaire (2)

Use acetic acid 3.5 % for detection of subtle lesions

Chromoendoscopy with acetic acid may help to detect subtle lesions in patients with high grade dysplasia.

Participant questionnaire (1)

Use the Paris classification [15]

Characterization of the macroscopic type of the lesion may contribute to a more thorough inspection of the lesion and helps the endoscopist to recognize early neoplastic lesions

Participant questionnaire (3)

Retroflex for inspection of lesions located at the cardia ([Fig. 4])

Perform inspection with the endoscope in the retroflexed position with adequate inflation of air as this orientation significantly improves the visualization of the distal margin of lesions and assists practically in delineation and submucosal lifting (use a flexible endoscope that can turn 180° and can easily make the U turn required)

Participant questionnaire (2)/ video feedback report (1)

Delineation and marking of the lesion

Place electrocoagulation markings to delineate the target area, creating a preprocedural plan ([Fig. 5]; [Video 2])

This preprocedural plan created under optimal imaging conditions prior to the resection provides a roadmap for an effective and safe endoscopic resection and should be adhered to during the remainder of the procedure as the view of the working area will diminish because of visualization through the endoscopic resection cap, the use of submucosal lifting, bleeding, or electrocoagulation effects, meaning the endoscopist may lose the orientation and perspective on the lesion causing incomplete or unnecessarily large resections

Participant questionnaire (1)/ video feedback report (1)

Place markings at a distance of 2 – 5 mm from the lesion

Markers summarize the preprocedural plan and serve as a guide for the endoscopist as, after lifting and resection, recognition of the lesion is difficult; markings should not be placed too close or too far away from the lesion, or too deep into the hiatal hernia; use enough markings, but do not exaggerate

Video feedback report (9)

Place markings and perform submucosal lifting (in case of the endoscopic resection cap) before attaching the cap to the endoscope

It may be easier to place markings and to lift the lesion without the cap on the tip of the endoscope because the cap decreases the view and the maneuverability

Video feedback report (2)

In a tortuous esophagus or in the cardia, lesion marking may be easier with the cap already fitted to the tip of the endoscope

Place the markings with the cap already at the tip of the endoscope, which allows the cardiac folds to be pushed away

Participant questionnaire (2)

Do not place additional markings after the first resection

After the first resection, the endoscopic view is impaired by bleeding; stick to your preprocedural plan

Video feedback report (2)

Markings are preferably placed with argon plasma coagulation (APC)

APC markings are more visible than coagulation markings made using the tip of the snare

Participant questionnaire (1)

Time the placement of the markings with the movements of respiration and peristalsis

Take your time to carefully delineate the lesion because the markings create your preprocedural plan

Participant questionnaire (1)

Retroflex to place markings where a lesion is located in the cardia or distal esophagus

If a lesion is in the cardia or distal esophagus, placing markings may be easier in the retroflexed position, as this provides a better view of the lesion

Participant questionnaire (1)

When placing markings in the retroflexed position, keep your instruments close to the tip of the endoscope and create maximal angulation

In the retroflexed position, keep the snare or injection needle at a short distance from the tip of endoscope, keep the small wheel completely turned to the right, resulting in maximal angulation, and gently pull the endoscope back into the hernia

Video feedback report (1)

Obtain a still image of the marked lesion

Evaluate the still image immediately to assess if delineation and markings are adequate (some of the markings may be more important than others, for example those close to the edge of the lesion)

Participant questionnaire (2)

In situations where there is a residual lesion or local recurrence at the site of a previous endoscopic resection, place your markings widely

Place your markings widely to achieve complete resection during this session

Participant questionnaire (1)

For squamous lesions, markings should be placed immediately after chromoendoscopy with Lugol dye

After chromoendoscopy with Lugol dye for squamous neoplastic lesions, the unstained areas correspond to the neoplastic areas, which allows for adequate delineation of the lesion – never place markings prior to chromoendoscopy

Video feedback report (1)

Always perform a test suction prior to the endoscopic resection ( [Videos 1, 2] ).

The test suction provides an estimate of how much and which part of the lesion enters the cap and is performed before placement of the endoscopic resection snare in the ridge of the cap (for endoscopic resection cap procedures) or before releasing a rubber band (for MBM procedures); it allows the endoscopist to adjust the position of the cap, the pressure of the cap on the tissue, the suction force, and the way the cap is maneuvered during suctioning of the tissue if required; a test suction is also important to judge whether the amount of overlap between two endoscopic resections is too large (increases perforation risk) or too small (results in remaining tissue bridges between adjacent resections)

Participant questionnaire (3)/ video feedback report (16)

Achieve a complete vacuum during the test suction resulting in an imprint in the tissue

If a complete vacuum is not achieved, the test suction is not an adequate test and is useless because a complete vacuum will be used during the resection

Video feedback report (2)

Apply controlled suction ( [Video 1] )

Evaluate the suction force during the test suction and prior to resection and control the suction force with the suction button or by creating an air leak between the cap and the esophageal wall during suctioning to reduce the suction force (having control over the suction force is especially helpful when a smaller sized resection is intended)

Participant questionnaire (2)

Submucosal fluid injection (endoscopic resection cap technique) and suctioning of the lesion into the cap

Use a good quality injection needle

Some injection needles result in slower submucosal injection, delaying the procedure

Participant questionnaire (1)

Start the submucosal injection distally

This way the lesion turns into the endoscopic view, not behind the lifted area

Participant questionnaire (1)

Do not apply too much forward pressure during injection

This may result in too deep an injection and does not give adequate lifting

Video feedback report (2)

Start to inject fluid through the injection needle just before the needle is pushed into the mucosa

Inject while flushing and observe if the lamina propria comes up, which shows that you are in the right (submucosal) layer; perform the fluid injection in a stepwise manner to be able to assess the effect of the lifting and avoid excessive lifting of the mucosa outside the delineated area as this only obscures the view of the working area

Participant questionnaire (2)/ video feedback report (2)

Do not use methylene blue in the lifting fluid during submucosal lifting in endoscopic resection

This reduces the visibility of the target area during the endoscopic resection procedure

Participant questionnaire (1)

Use the Kato lifting classification to carefully assess the lifting sign [16]

Describing the lifting sign promotes good inspection of the adequacy of the lifting

Participant questionnaire (2)

To discriminate between a lifting type II or III, use the biopsy forceps to gently manipulate the lifted area, to see if the area will lift [16]

Sometimes it is difficult to discriminate between a lifting type II (hard lifting) or type III (incomplete lifting), for example in the cardiac area, which has major consequences because for type III lifting, the patient should be referred for surgery

Participant questionnaire (1)

In case of piecemeal endoscopic resection, lift the edges of the previous endoscopic resection prior to subsequent resections to avoid perforation

This reduces the chance of endoscopic resection with too much overlap, which carries a risk of perforation; where residual bridges occur, lift the tissue bridge and the endoscopic resection wound to avoid perforation

Video feedback report (4)

Lift an area with fibrosis or scarring first (Kato lifting classification) [16]

This prevents the surrounding tissue lifting, which will obscure the view of the area of interest or will give a false impression of non-lifting; in case of a poor lifting sign, do not perform endoscopic resection of this area

Video feedback report (1)

Resection

Ensure that you have the right size cap

Prior to starting endoscopic resection, assess if the cap is suitable for the diameter of the endoscope; if the cap slips off easily, use a smaller size of cap

Video feedback report (1)

Fix the resection cap to the tip of the endoscope with a strip of tape

This prevents the cap slipping off when there is resistance, for instance at the upper esophageal sphincter or at the level of an esophageal stenosis

Participant questionnaire (1)

If introduction of the cap into the esophagus is difficult, use a spraying catheter as a guide

The spraying catheter serves as a guide through the pharynx and upper esophageal sphincter

Participant questionnaire (1)

Target the most suspicious area first during resection

The worst area of the lesion has to be in the center of the first endoscopic resection specimen to reduce the chance of incomplete resection, as resection at an overlap zone of two resections may be less deep, but in doing this, avoid leaving behind markings at the sides of the delineated area

Video feedback report (1)

Where no one part of the lesion appears particularly suspicious, start in the distal area

Resection of the distal area is more difficult than the proximal area; after the first resection, the view may be impaired by blood or the effect of suctioning and resection

Participant questionnaire (1)

In piecemeal resection, plan your resections in such a way that they are situated in the best way

In piecemeal resection there is an increased risk of having too much overlap or of leaving residual tissue bridges from resections that are adjacent in the longitudinal direction compared with those in the circumferential direction

Participant questionnaire (2)

Keep the vacuum on the resection cap until the snare is completely closed (for endoscopic resection cap procedures) or the rubber band is released (for MBM procedures)

If the vacuum is released too early, only a superficial resection will be performed, resulting in a small specimen and a higher risk of incomplete resection at the vertical margin

Video feedback report (2)

During resection of a small tissue bridge between two resection wounds, reduce the suction force and do not go for a complete “red-out” (endoscopic view completely filled with tissue)

Complete “red-out” during resection of a small tissue bridge may result in perforation because the tissue enters the cap more easily as a result of the previous resections; apply less suction force if you want to resect a tissue bridge (after adequate submucosal injection for the endoscopic resection cap), by letting some air escape from the vacuum

Video feedback report (1)

If the tissue lacerates during suctioning of the lesion, continue suctioning and complete the resection

There is no increased perforation risk because the laceration occurs at the level of the mucosal layer; do not stop to inspect as bleeding will impair visualization; provided that the lesion has been adequately lifted, continue suctioning and perform resection according to your preprocedural plan and test suction

Participant questionnaire (1)/ video feedback report (1)

After resection, immediately inspect the endoscopic resection wound

Inspect the endoscopic resection wound to exclude perforation and bleeding, and to assess the completeness of the resection

Video feedback report (8)

For lesions in the cardia or distal esophagus, retroflex the endoscope

Inspect, delineate and inject the distal margin with the scope in the retroflexed position (however, suctioning the lesion into the cap and the actual resection should be performed with the scope in the antegrade position)

Video feedback report (1)

If there is doubt as to whether all the markings and the complete lesion have been removed, perform an additional endoscopic resection at that site

Subsequent endoscopic resection at a later stage will be significantly more difficult because of scarring; if you do not perform an additional resection, at least take biopsies of the edges of the endoscopic resection wound ( + /- APC of the doubtful area)

Participant questionnaire (1)/ video feedback report (1)

Never blindly introduce the endoscope after resection

Be aware of the vulnerability of the defect of the endoscopic resection wound and introduce the endoscope under direct vision

Participant questionnaire (1)

Choose the endoscopic resection cap technique instead of the MBM technique for a fibrosed area

It may be more difficult to suction an area with fibrosis into the cap, so use of the endoscopic resection cap with submucosal lifting may be preferable (gentle wiggling and rotating of the endoscope may promote suctioning of the lesion into the cap)

Video feedback report (1)

Multiband mucosectomy technique (MBM)

Release any bands that will not be used to optimize the endoscopic view ([Fig. 6])

If only one to three resections will be required (as in the majority of patients), release two or three rubber bands prior to introducing the endoscope as this will greatly improve visualization

Participant questionnaire (1)/ video feedback report (3)

Rotate the cap in such a way that the pulling strings are not in the endoscopic view

Align the pulling strings of the MBM barrel with the working channel to keep them out of the endoscopic view (prior to insertion of the endoscope)

Participant questionnaire (2)/ video feedback report (1)

If the rubber band slips off the pseudopolyp, do not proceed with the resection

This may indicate fixation of the tissue due to scarring or submucosal tumor infiltration and these areas are not suitable for endoscopic resection; try submucosal injection and if the area shows a good lifting sign, there may be an argument for continuing resection of the area

Participant questionnaire (1)

Be aware that where it is difficult to suction the lesion into the MBM cap, the wrong area may enter the cap, leading to placement of the rubber band at an incorrect location

When it is difficult to suction the area into the cap, this may indicate scarring or submucosal tumor infiltration and these areas are not suitable for resection; try submucosal injection and if the area shows a good lifting sign, there may be an argument for continuing resection of the area

Participant questionnaire (2)

Snare the pseudopolyp below the rubber band to ensure complete resection

If the snare is placed above the rubber band, the snare may end up too high at the back of the pseudopolyp, resulting in an incomplete resection (only if two bands are accidently fired at once is snaring above the bands advisable to ensure the resection is not too deep)

Video feedback report (2)

During MBM the endoscopy nurse needs to close the snare forcefully

During MBM more coagulation trauma is created because of a thicker snare and the absence of lifting

Video feedback report (1)

During endoscopic resection with the MBM technique, close the snare more quickly and more forcefully while cutting through the specimen

Compared with the endoscopic resection cap technique, the MBM technique may be more prone to coagulation trauma because the snare is thicker and no submucosal lifting is performed

Participant questionnaire (1)

Endoscopic resection cap technique

Place the snare in the rim of the cap in the proximal esophagus or antrum, applying a small amount of suction to the mucosa

Placing the snare in the rim of the cap is preferably done outside of the target area, because the target area needs to be naive to suctioning prior to performing the resection, except for the test suction

Participant questionnaire (1)

Position the lesion at 6 o’clock in the endoscopic view ([Fig. 7])

Most endoscopic resection cap procedures are performed with an oblique cap with the shorter part of the cap pointing towards 6 o’clock and the longer part to 12 o’clock; with this orientation, it is imperative that the endoscope is rotated in such a way that the lesion is positioned at 6 o’clock, especially during suctioning of the lesion into the cap and closure of the snare (if the lesion is located at 12 o’clock, closure of the snare will result in dislocation of the snare from the ridge of the cap and tangential cutting through the specimen, and there will be higher pressure on the esophageal wall with less effective suctioning, both of which may result in a smaller specimen and a higher chance of incomplete vertical resection)

Participant questionnaire (4)/ video feedback report (13)

Position the snare in the rim of the cap in such a way that the two wires are located at the lower half of the endoscopic view at the location of the working channel of the endoscope

This ensures that all of the tissue that is suctioned into the cap (the pseudopolyp) is captured within the snare when it is closed (if the wires are located at 12 o’clock, the snare will dislocate from the rim of the cap when it is closed or will fail to close around the base of the pseudopolyp, resulting in incomplete resection

Video feedback report (2)

Prevent dislocation of the snare from the rim of the endoscopic resection cap

Placement of the snare in the rim of the cap is not easy and the subsequent approach to the target area, correct positioning of the cap over the lesion, suctioning of the tissue, closure of the snare, and resection demand good coordination between the endoscopist and endoscopy nurse (this is a delicate process and manipulating the cap onto the lesion or changing the position of the scope may result in dislocation of the snare from the rim of the cap); a prior test suction enables the endoscopist to target the resection area in the most straightforward way

Participant questionnaire (1)

Ensure there is sufficient submucosal lifting if you want to resect a small residual tissue bridge between two resection wounds

If you want to resect a small piece, apply less suction after sufficient submucosal injection, by letting some air escape from the vacuum, as the tissue enters the cap more easily between two adjacent resection wounds, so the risk for perforation is higher

Participant questionnaire (1)

Resect the pseudopolyp outside the endoscopic resection cap

Resecting the pseudopolyp within the cap impairs the endoscopic view during the procedure

Participant questionnaire (1)

Do not wiggle the snare during electrocoagulation

Too much movement of the snare during resection may increase the perforation risk; it is better to wiggle the snare and the enclosed pseudopolyp prior to resection, not during electrocoagulation

Participant questionnaire (1)

Introduce the snare an extra 1 – 2 mm into the working channel during closure of the snare

To ensure that the snare is positioned at the distal rim of the endoscopic resection cap, introduce the snare an extra 1 – 2 mm through the working channel during closure of the snare (if the snare sheet is kept too close to the rim of the endoscopic resection cap, this may complicate subtle correction of the position of the snare)

Video feedback report (1)

A new snare should be used for every endoscopic cap resection

The crescent-shaped endoscopic resection snares are very thin and generally lose their shape during the forceful closure of the resection process so opening and repositioning of the snare in the rim of the cap for an additional resection is generally not possible (this in contrast to the MBM technique where the hexagonal snare of the MBM kit can be used for multiple resections)

Video feedback report (3)

Test if a used snare can be opened if you want to use it for coagulation, prior to introducing it and targeting the bleeding site

After a resection, a snare is frequently malformed and cannot be opened anymore; if bleeding occurs, it is therefore advisable to test if the snare can be opened prior to localization and targeting of the bleeding site for careful coagulation

Video feedback report (1)

Do not use the large flexible endoscopic resection cap for piecemeal endoscopic resection

After a resection with the flexible large endoscopic resection cap (to remove the most suspicious area), subsequent resections need to be performed with a standard cap as the greater diameter of the flexible large cap increases the perforation risk

Video feedback report (2)

Retrieval of specimens

Store the specimen in the stomach after resection

Do not immediately remove the specimen, but first inspect the resection area to exclude bleeding and perforation

Participant questionnaire (1)

Store the specimens in the stomach and use a retrieval net at the end of the procedure

Separate removal of each specimen after resection is time consuming, introduces a delay if there is bleeding, and generally does not contribute to clinical management as piecemeal resections do not allow reliable reconstruction of the resected lesion (at least not in Barrett’s lesions); use a retrieval basket to collect the specimens from the stomach (the last specimen can be sucked into the cap, which is then sealed off by pulling the basket against it)

Participant questionnaire (4)

Do not use too much irrigation

When there is a lot of fluid in the stomach, the retrieval of specimens takes more time

Participant questionnaire (1)

Use a polyp trap in the suction reservoir

This prevents small specimens being sucked into the suction waste containers

Participant questionnaire (1)

Never remove two specimens at once within the cap, because the first specimen will block the suction channel leaving the second specimen more or less free in the cap

If two specimens are simultaneously removed from the stomach with the cap, the outermost specimen may fall out into the trachea leading to a risk of aspiration

Video feedback report (1)

When emptying the stomach of fluids, keep the snare or retrieval basket in the working channel to block the working channel in case specimens enter the cap during suctioning

A tissue specimen in the cap prevents adequate removal of fluids from the stomach

Video feedback report (1)

Carefully count the number of resections and specimens

During piecemeal endoscopic resection, write down the number of resections to ensure that you know how many specimens need to be retrieved at the end of the procedure

Participant questionnaire (1)

Pin the specimens down onto cork or paraffin to optimize orientation and cutting in the pathology laboratory

Avoid inwards curling of the lateral edges because positive lateral margins (not uncommon in piecemeal resections) can give a false impression of an incomplete deep resection margin

Participant questionnaire (1)

Complications

Know the management of bleeding

Slowly move the edge of the transparent cap over the esophageal wall to localize the bleeding focus and gently compress the bleeding site to achieve hemostasis; wait to see if the bleeding stops spontaneously, while keeping the bleeding site in the endoscopic view; flush with little bursts of water using a water-jet system but avoid prolonged flushing; if the bleeding is not self-limiting, apply coagulation with the tip of the snare, which should be done with minimal pressure on the bleeding site to avoid deep damage; if this does not result in hemostasis, use a hot biopsy forceps (placing clips is not preferred as an initial measure as the clip generally complicates further piecemeal resections)

Participant questionnaire (2)

Bleeding is easier to treat with the cap on the endoscope

The cap may help to localize the bleeding focus when you slowly move it over the area in which the bleeding occurred; furthermore, the cap can be used for gentle compression to achieve hemostasis

Video feedback report (1)

If bleeding occurs, switch to a therapeutic endoscope

A therapeutic endoscope is equipped with a separate water-jet channel, which enables simultaneous flushing with water and interventions through the working channel of the endoscope to achieve an optimal view

Participant questionnaire (1)

If bleeding occurs, flush with little bursts of water using a water-jet system

Flushing with little bursts of water helps to locate the bleeding focus more easily than forceful and prolonged flushing does

Participant questionnaire (1)

During coagulation of bleeding, keep the instrument close to the tip of the endoscope

When there is too much of the snare or hot biopsy forceps outside during coagulation, targeting of the bleeding is more difficult, which increases the risk of deeper coagulation damage (too much coagulation in the endoscopic resection wound increases the risk of stenosis or perforation)

Video feedback report (2)

Know the management of perforation ( [Videos 1, 2 and 3] )

If perforation occurs during endoscopic resection, do not insufflate air, but switch to CO2 insufflation, place a sump tube at the perforation site, and start antibiotics; there are two key questions that guide further management. First, can the neoplasia still be effectively treated endoscopically? It makes little sense to manage a perforation conservatively and/or endoscopically if final management of the neoplasia will require esophagectomy anyway (for example because of a submucosal invading cancer). Second, is effective closure of the perforation feasible without compromising the success of the neoplasia treatment? New clipping devices may effectively close large perforations but may also bury residual neoplasia at the perforation site; covered stents may effectively seal the perforation site and are especially effective if there is local narrowing of the esophagus that prevents dislocation

Participant questionnaire (2)

Switch to CO2 insufflation after perforation

Beware of blowing air after perforation, as you may create a pneumomediastinum. CO2 insufflation is preferable in this situation

Participant questionnaire (1)

Be aware that there is a risk of laceration during difficult introduction of the cap and use gentle pressure or perform a dilation first

Introduction of the endoscopic resection cap may cause mucosal laceration in patients with a scarred or narrowed esophagus due to prior endoscopic resection, ulceration, or reflux

Video feedback report (2)

If closing a perforation with clips, close the defect by starting approximation at the edges; do not start approximating the center

It will be easier to achieve closure if you start closing a defect by approximating the edges

Participant questionnaire (1)

Miscellaneous

Hire and train a dedicated ‘Barrett nurse’

A Barrett nurse can assist in selecting patients, and in planning and organizing the Barrett care in your center

Participant questionnaire (1)

Act as a team together with your endoscopy nurse

It is a team effort to keep the endoscope and instruments in position and perform the resection

Participant questionnaire (1)

Use a time-out and debriefing procedure

Use a time-out before starting the procedure to inform co-workers, and a debriefing after the procedure to discuss points of importance, helping you improve as an endoscopy team

Participant questionnaire (1)

Keep instruments (snare, injection needle, APC probe) close to the tip of the endoscope

This provides optimal control of the instrument, for instance when making markings, only the tip of the snare should be used, otherwise markings will be too large and imprecise, or tissue damage and bleeding may occur

Participant questionnaire (2)/ video feedback report (15)

Do not use too much forward pressure during coagulation, snare placement, or snaring

Too much forward pressure during coagulation, snare placement, or snaring may result in the snare slipping out of the cap with a risk of tissue damage; during snaring, too much pressure may result in deep coagulation damage or perforation (the instrument should remain in the endoscope as much as possible to avoid tissue damage)

Video feedback report (4)

Zoom Image
Fig. 4 Endoscopic views in two different patients with Barrett’s esophagus and early neoplastic lesions located at the gastric cardia demonstrating the benefits of retroflexion in improving the visualization of the distal margin of the lesion and aiding delineation and submucosal lifting prior to endoscopic resection. A lesion at 10 o’clock seen in: a antegrade view; b retrograde view. A lesion at 11 o’clock seen in: c antegrade view; d retrograde view.
Zoom Image
Fig. 5 Endoscopic views showing electrocoagulation markings placed to delineate the area to be resected under optimal imaging conditions prior to the resection, thereby creating a preprocedural plan that serves as a roadmap for an effective and safe endoscopic resection.
Zoom Image
Fig. 6 Endoscopic images showing how the view can be optimized using the multiband mucosectomy (MBM) technique by: a, b aligning the pulling strings of the MBM barrel with the working channel to keep them out of the endoscopic view prior to insertion of the endoscope; and c, d releasing two to three rubber bands prior to introducing the endoscope (if only one to three resections will be required).
Zoom Image
Fig. 7 Illustration of endoscopic resection using the oblique cap showing: a a lesion in the esophagus; b incorrect positioning as a lesion positioned at 12 o’clock requires an angulated position of the endoscope to suction the lesion into the cap (in the position shown suctioning may be less effective because of higher tension on the esophageal wall, which means the snare will not close at the base of the lesion, possibly resulting in only a partial/superficial resection of the lesion; c correct positioning with the endoscope rotated so that the lesion is at 6 o’clock in the endoscopic view to facilitate injection, suctioning, and resection.

#

Selection of important learning points

Of the 90 learning points, 34 were classified as “important” by the three experienced endoscopists. These 34 points and the corresponding technical recommendations are described in [Table e2] (available online).

Table e2

The 34 most important learning points for performing endoscopic resection that were selected by three expert endoscopists from a list of 90 learning points generated following the endoscopic resection training program (bold entries indicate the top 10).

Topic

Learning point

Importance

Topic

Learning point

Importance

Work-up and imaging

Optimize the endoscopic view by repeatedly cleaning out the stomach, esophagus, and the target area ( [Videos 2 and 3] )

At the start of the endoscopic resection procedure, the endoscopist should empty the stomach and clean the target area in the esophagus by flushing with water (and using antifoam) to remove mucus and gastric contents; during the procedure, systematic emptying of the stomach and esophagus should be repeated frequently (e. g. with each insertion of an instrument) to ensure an optimal view of the target area and prevent aspiration during the endoscopy

Use an endoscope with a separate water-jet channel

After imaging, switch to an endoscope with a separate water-jet channel to optimize cleaning and imaging of the target area and enable simultaneous flushing of water and passage of instruments through the working channel

Do not compromise inspection and delineation of the lesion – allow sufficient time for inspection and use a high-definition endoscope

Allow time for meticulous inspection of the lesion; optimize the inspection by using a high-definition endoscope equipped with a (virtual) chromoendoscopy technique (most therapeutic gastroscopes have an inferior image quality compared to their diagnostic counterparts)

Retroflex for inspection of lesions located at the cardia ([Fig. 4])

Perform inspection with the endoscope in the retroflexed position with adequate inflation of air as this orientation significantly improves the visualization of the distal margin of lesions and assists practically in delineation and submucosal lifting (use a flexible endoscope that can turn 180° and can easily make the U turn required)

Insufflate an adequate amount of air

Insufflate air until there is a good view in the esophagus; if the patient is not able to hold the air, it may be that they are too deeply sedated

Use the Paris classification [15]

Characterization of the macroscopic type of the lesion may contribute to a more thorough inspection of the lesion and helps the endoscopist to recognize early neoplastic lesions

Delineation and marking of the lesion

Place electrocoagulation markings to delineate the target area, creating a preprocedural plan ([Fig. 5]; [Video 2])

This preprocedural plan created under optimal imaging conditions prior to the resection provides a roadmap for an effective and safe endoscopic resection and should be adhered to during the remainder of the procedure as the view of the working area will diminish because of visualization through the endoscopic resection cap, the use of submucosal lifting, bleeding, or electrocoagulation effects, meaning the endoscopist may lose the orientation and perspective on the lesion causing incomplete or unnecessarily large resections

Place markings at a distance of 2 – 5 mm from the lesion

Markers summarize the preprocedural plan and serve as a guide for the endoscopist as, after lifting and resection, recognition of the lesion is difficult; markings should not be placed too close or too far away from the lesion, or too deep into the hiatal hernia; use enough markings, but do not exaggerate

Obtain a still image of the marked lesion

Evaluate the still image immediately to assess if delineation and markings are adequate (some of the markings may be more important than others, for example those close to the edge of the lesion)

Submucosal fluid injection (endoscopic resection cap technique) and suctioning of the lesion into the cap

Start to inject fluid through the injection needle just before the needle is pushed into the mucosa

Inject while flushing and observe if the lamina propria comes up, which shows that you are in the right (submucosal) layer; perform the fluid injection in a stepwise manner to be able to assess the effect of the lifting and avoid excessive lifting of the mucosa outside the delineated area as this only obscures the view of the working area

In case of piecemeal endoscopic resection, lift the edges of the previous endoscopic resection prior to subsequent resections to avoid perforation

This reduces the chance of endoscopic resection with too much overlap, which carries a risk of perforation; where residual bridges occur, lift the tissue bridge and the endoscopic resection wound to avoid perforation

Lift an area with fibrosis or scarring first (Kato lifting classification) [16]

This prevents the surrounding tissue lifting, which will obscure the view of the area of interest or will give a false impression of non-lifting; in case of a poor lifting sign, do not perform endoscopic resection of this area

Always perform a test suction prior to the endoscopic resection ([Videos 1, 2])

The test suction provides an estimate of how much and which part of the lesion enters the cap and is performed before placement of the endoscopic resection snare in the ridge of the cap (for endoscopic resection cap procedures) or before releasing a rubber band (for MBM procedures); it allows the endoscopist to adjust the position of the cap, the pressure of the cap on the tissue, the suction force, and the way the cap is maneuvered during suctioning of the tissue if required; a test suction is also important to judge whether the amount of overlap between two endoscopic resections is too large (increases perforation risk) or too small (results in remaining tissue bridges between adjacent resections)

Apply controlled suction ([Video 1])

Evaluate the suction force during the test suction and prior to resection and control the suction force with the suction button or by creating an air leak between the cap and the esophageal wall during suctioning to reduce the suction force (having control over the suction force is especially helpful when a smaller sized resection is intended)

Resection

Ensure that you have the right size cap

Prior to starting endoscopic resection, assess if the cap is suitable for the diameter of the endoscope; if the cap slips off easily, use a smaller size of cap

Fix the resection cap to the tip of the endoscope with a strip of tape

This prevents the cap slipping off when there is resistance, for instance at the upper esophageal sphincter or at the level of an esophageal stenosis

Keep the vacuum on the resection cap until the snare is completely closed (for endoscopic resection cap procedures) or the rubber band is released (for MBM procedures)

If the vacuum is released too early, only a superficial resection will be performed, resulting in a small specimen and a higher risk of incomplete resection at the vertical margin

If the tissue lacerates during suctioning of the lesion, continue suctioning and complete the resection

There is no increased perforation risk because the laceration occurs at the level of the mucosal layer; do not stop to inspect as bleeding will impair visualization; provided that the lesion has been adequately lifted, continue suctioning and perform resection according to your preprocedural plan and test suction

After resection, immediately inspect the endoscopic resection wound

Inspect the endoscopic resection wound to exclude perforation and bleeding, and to assess the completeness of the resection

For lesions in the cardia or distal esophagus, retroflex the endoscope

Inspect, delineate and inject the distal margin with the scope in the retroflexed position (however, suctioning the lesion into the cap and the actual resection should be performed with the scope in the antegrade position)

If there is doubt as to whether all the markings and the complete lesion have been removed, perform an additional endoscopic resection at that site

Subsequent endoscopic resection at a later stage will be significantly more difficult because of scarring; if you do not perform an additional resection, at least take biopsies of the edges of the endoscopic resection wound (+/– argon plasma coagulation [APC] of the doubtful area)

Multiband mucosectomy technique

Release any bands that will not be used to optimize the endoscopic view ([Fig. 6])

If only one to three resections will be required (as in the majority of patients), release two or three rubber bands prior to introducing the endoscope as this will greatly improve visualization

Rotate the cap in such a way that the pulling strings are not in the endoscopic view

Align the pulling strings of the MBM barrel with the working channel to keep them out of the endoscopic view (prior to insertion of the endoscope)

During endoscopic resection with the MBM technique, close the snare more quickly and more forcefully while cutting through the specimen

Compared to the endoscopic resection cap technique, the MBM technique may be more prone to coagulation trauma because the snare is thicker and no submucosal lifting is performed

Endoscopic resection cap technique

Position the lesion at 6 o’clock in the endoscopic view ([Fig. 7])

Most endoscopic resection cap procedures are performed with an oblique cap with the shorter part of the cap pointing towards 6 o’clock and the longer part to 12 o’clock; with this orientation, it is imperative that the endoscope is rotated in such a way that the lesion is positioned at 6 o’clock, especially during suctioning of the lesion into the cap and closure of the snare (if the lesion is located at 12 o’clock, closure of the snare will result in dislocation of the snare from the ridge of the cap and tangential cutting through the specimen, and there will be higher pressure on the esophageal wall with less effective suctioning, both of which may result in a smaller specimen and a higher chance of incomplete vertical resection)

Position the snare in the rim of the cap in such a way that the two wires are located at the lower half of the endoscopic view at the location of the working channel of the endoscope

This ensures that all of the tissue that is suctioned into the cap (the pseudopolyp) is captured within the snare when it is closed (if the wires are located at 12 o’clock, the snare will dislocate from the rim of the cap when it is closed or will fail to close around the base of the pseudopolyp, resulting in incomplete resection

Prevent dislocation of the snare from the rim of the endoscopic resection cap

Placement of the snare in the rim of the cap is not easy and the subsequent approach to the target area, correct positioning of the cap over the lesion, suctioning of the tissue, closure of the snare, and resection demand good coordination between the endoscopist and endoscopy nurse (this is a delicate process and manipulating the cap onto the lesion or changing the position of the scope may result in dislocation of the snare from the rim of the cap); a prior test suction enables the endoscopist to target the resection area in the most straightforward way

A new snare should be used for every endoscopic cap resection

The crescent-shaped endoscopic resection snares are very thin and generally lose their shape during the forceful closure of the resection process so opening and repositioning of the snare in the rim of the cap for an additional resection is generally not possible (this in contrast to the MBM technique where the hexagonal snare of the MBM kit can be used for multiple resections)

Do not use the large flexible endoscopic resection cap for piecemeal endoscopic resection

After a resection with the flexible large endoscopic resection cap (to remove the most suspicious area), subsequent resections need to be performed with a standard cap as the greater diameter of the flexible large cap increases the perforation risk

Retrieval of specimens

Store the specimens in the stomach and use a retrieval net at the end of the procedure

Separate removal of each specimen after resection is time consuming, introduces a delay if there is bleeding, and generally does not contribute to clinical management as piecemeal resections do not allow reliable reconstruction of the resected lesion (at least not in Barrett’s lesions); use a retrieval basket to collect the specimens from the stomach (the last specimen can be sucked into the cap, which is then sealed off by pulling the basket against it)

Pin the specimens down onto cork or paraffin to optimize orientation and cutting in the pathology laboratory

Avoid inwards curling of the lateral edges because positive lateral margins (not uncommon in piecemeal resections) can give a false impression of an incomplete deep resection margin

Complications

Know the management of bleeding

Slowly move the edge of the transparent cap over the esophageal wall to localize the bleeding focus and gently compress the bleeding site to achieve hemostasis; wait to see if the bleeding stops spontaneously, while keeping the bleeding site in the endoscopic view; flush with little bursts of water using a water-jet system but avoid prolonged flushing; if the bleeding is not self-limiting, apply coagulation with the tip of the snare, which should be done with minimal pressure on the bleeding site to avoid deep damage; if this does not result in hemostasis, use a hot biopsy forceps (placing clips is not preferred as an initial measure as the clip generally complicates further piecemeal resections)

Know the management of perforation ([Videos 1 – 3])

If perforation occurs during endoscopic resection, do not insufflate air, but switch to CO2 insufflation, place a sump tube at the perforation site, and start antibiotics; there are two key questions that guide further management. First, can the neoplasia still be effectively treated endoscopically? It makes little sense to manage a perforation conservatively and/or endoscopically if final management of the neoplasia will require esophagectomy anyway (for example because of a submucosal invading cancer). Second, is effective closure of the perforation feasible without compromising the success of the neoplasia treatment? New clipping devices may effectively close large perforations but may also bury residual neoplasia at the perforation site; covered stents may effectively seal the perforation site and are especially effective if there is local narrowing of the esophagus that prevents dislocation

Miscellaneous

Keep instruments (snare, injection needle, APC probe) close to the tip of the endoscope

This provides optimal control of the instrument, for instance when making markings, only the tip of the snare should be used, otherwise markings will be too large and imprecise or tissue damage and bleeding may occur


#

Compilation of the top 10 tips

The top 10 tips compiled from the list of 34 important learning points are described in detail below and summarized in [Table 3] (they are shown as bold in [Table e2]). The individual rankings and scores for this final top 10 are shown in [Table e4] (available online).

Table 3

Top 10 tips for performing endoscopic resection based on experiences from an endoscopic resection training program.

Rank

Learning point

1

Do not compromise inspection and delineation of the lesion – allow sufficient time for inspection and use a high-definition endoscope

2

Place electrocoagulation markings to delineate the area to be resected to create a preprocedural plan ([Fig. 5]; [Video 2])

3

Know the management of bleeding

4

Optimize the endoscopic view by repeatedly cleaning out the stomach, esophagus, and the target area ([Videos 2, 3])

5

Use an endoscope with a separate water-jet channel

6

Always perform a test suction prior to the endoscopic resection ([Videos 1, 2])

7

Keep instruments (snare, injection needle, argon plasma coagulation [APC] probe) close to the tip of the endoscope

8

In case of piecemeal endoscopic resection, lift the edges of the previous endoscopic resection prior to subsequent resections ([Video 1])

9

Know the management of perforation ([Videos 1 – 3])

10

Pin the specimens down onto cork or paraffin to optimize orientation and cutting in the pathology laboratory

Table e4

Rankings and mean scores of the top 10 tips for performing endoscopic resection based on experiences from an endoscopic resection training program.

Rank

Learning point

Expert endoscopist

Mean score

A

B

C

1/2

Do not compromise inspection and delineation of the lesion – allow sufficient time for inspection and use a high-definition endoscope

 5

 1

 1

2.3

1/2

Place electrocoagulation markings to delineate the area to be resected to create a preprocedural plan ([Fig. 5]; [Video 2])

 1

 2

 4

2.3

3

Know the management of bleeding

 3

 –

 5

4.0

4

Optimize the endoscopic view by repeatedly cleaning out the stomach, esophagus, and the target area ([Videos 2 and 3])

 7

 –

 2

4.5

5

Use an endoscope with a separate water-jet channel

 –

 5

 –

5.0

6/7

Always perform a test suction prior to the endoscopic resection ([Videos 1, 2])

10

 3

 3

5.3

6/7

Keep instruments (snare, injection needle, argon plasma coagulation [APC] probe) close to the tip of the endoscope

 2

 4

10

5.3

8

In case of piecemeal endoscopic resection, lift the edges of the previous endoscopic resection prior to subsequent resections ([Video 1])

 –

 6

 –

6.0

9/10

Know the management of perforation ([Videos 1 – 3])

 8

 –

 6

7.0

9/10

Pin the specimens down onto cork or paraffin to optimize orientation and cutting in the pathology laboratory

 4

10

 –

7.0

1 Do not compromise inspection and delineation of the lesion: allow sufficient time for inspection and use a high-definition endoscope

Allow time for meticulous inspection of the lesion. Optimize inspection by using a high-definition endoscope equipped with a (virtual) chromoendoscopy technique. Most therapeutic gastroscopes have an inferior image quality compared to their diagnostic counterparts.

2 Place electrocoagulation markings to delineate the target area so creating a “preprocedural plan” ([Fig. 5]; [Video 2])

This “preprocedural plan” should be adhered to during the remainder of the procedure. During the procedure the view of the working area will diminish because of visualization through the endoscopic resection cap, the use of submucosal lifting, bleeding, or electrocoagulation effects. The endoscopist may then lose the orientation and perspective of the lesion, leading to incomplete or unnecessarily large resections. A preprocedural plan created during optimal imaging conditions prior to the resection is a roadmap for an effective and safe endoscopic resection.

3 Know the management of bleeding

Slowly move the edge of the transparent cap over the esophageal wall to localize the bleeding focus and gently compress the bleeding site to achieve hemostasis. Wait to see if the bleeding stops spontaneously, while keeping the bleeding site in the endoscopic view. Flush with little bursts of water using a water-jet system but avoid prolonged flushing. If the bleeding is not self-limiting, apply coagulation with the tip of the snare. This should be done with minimal pressure on the bleeding site to avoid deep damage. If this does not result in hemostasis, use a hot biopsy forceps. Placing clips is not preferred as an initial measure because the clip generally complicates further piecemeal resections.

4 Optimize the endoscopic view by repeatedly cleaning out the stomach, esophagus, and the target area ([Videos 2, 3])

At the start of the endoscopic resection procedure, the endoscopist should empty the stomach and clean the target area in the esophagus by flushing with water (and using antifoam) to remove any mucus and gastric contents. Systematic emptying of the stomach and esophagus should be repeated frequently during the procedure, as this ensures an optimal view of the working area and prevents aspiration during the endoscopy.

5 Use an endoscope with a separate water-jet channel

After imaging, switch to an endoscope with a separate water-jet channel. This optimizes cleaning and imaging of the target area and enables simultaneous flushing with water and the passage of instruments through the working channel of the endoscope, for example if bleeding occurs.

6 Always perform a “test suction” prior to the endoscopic resection ([Videos 1, 2])

The test suction provides an estimate of how much and which part of the lesion enters the cap. The test suction is performed before placement of the endoscopic resection snare in the ridge of the cap (for endoscopic resection cap procedures) or before releasing a rubber band (for MBM procedures). The endoscopist may adjust the position of the cap, the pressure of the cap on the tissue, the amount of suction, and the way the cap is maneuvered during suctioning of the tissue. A test suction is also important to judge if the amount of overlap between two endoscopic resections is too large (increases perforation risk) or too small (results in remaining tissue bridges between adjacent resections).

7 Keep instruments (snare, injection needle, argon plasma coagulation [APC] probe) close to the tip of the endoscope

This provides optimal control of the instrument. For example, when creating markings, only the tip of the snare should be used, otherwise the markings will be too large and imprecise, or tissue damage and bleeding may be caused.

8 In case of piecemeal endoscopic resection using the endoscopic resection cap technique, lift the edges of the previous endoscopic resection prior to subsequent resections ([Video 1])

This reduces the chances of endoscopic resection with too much overlap, which carries a risk of perforation, because the relatively soft endoscopic resection wound can enter the cap more easily than the residual esophageal mucosa. If residual bridges occur, lift the tissue bridge and the endoscopic resection wound to avoid perforation.

9 Know the management of perforation ([Videos 1 – 3])

If perforation occurs during endoscopic resection, do not insufflate air, but switch to CO2 insufflation, place a sump tube at the perforation site, and start antibiotics. There are two key questions that guide further management.

First, can the neoplasia still be effectively treated endoscopically? It makes little sense to manage a perforation conservatively and/or endoscopically if the final management of the neoplasia will require esophagectomy anyway (for example, because of a submucosal invading cancer).

Second, is effective closure of the perforation feasible without compromising the success of the neoplasia treatment? New clipping devices may close large perforations effectively but may also bury residual neoplasia at the perforation site. Covered stents may effectively seal the perforation site and are especially effective if there is local narrowing of the esophagus that will prevent dislocation.

10 Pin the specimens down onto cork or paraffin to optimize orientation and cutting in the pathology laboratory

Avoid inwards curling of the lateral edges because positive lateral margins (not uncommon in piecemeal resections) could give a false impression of an incomplete deep resection margin.


#
#

Discussion

In this study, the lessons in endoscopic resection technique learned from a training program in endoscopic resection were assessed. The learning points noted by the participating endoscopists or observed by the training program committee during the evaluation of video recordings were listed. Subsequently, the most important items that were considered relevant for the safety and efficacy of endoscopic resection were selected by three expert endoscopists and were discussed in detail. This resulted in a summary of “dos and don’ts” in endoscopic resection including our top 10 tips and instructive videos, which may be valuable for endoscopists with an interest in learning endoscopic resection or in improving their technique. The information herein is also integrated in a web-based training platform on the endoscopic management of early neoplasia in Barrett’s esophagus that is freely accessible (www.Best-Academia.eu).

In our training program, participants were trained in both the endoscopic resection cap and MBM techniques. At the time the training program was initiated, the MBM technique for endoscopic resection was a relatively new technique. The program committee decided to spend the first two training days (two of the 3-monthly sessions of hands-on training and video recording) on single-piece endoscopic cap resections (step 1) and piecemeal endoscopic cap resections (step 2). The MBM procedure was subsequently taught on the third training day. As a logical consequence, the endoscopic resection cap technique was used in the majority of the first 120 endoscopic resection procedures in this program.

Studies have shown that MBM is a quicker and cheaper resection technique than the endoscopic resection cap technique for early neoplasia in Barrett’s esophagus [5] [17] [18]. We believe that the MBM technique is the preferred resection technique for piecemeal resections of most flat-type lesions in Barrett’s esophagus. However, the endoscopic resection cap technique – despite being a technically more demanding procedure – still has an important place in managing early Barrett’s esophagus neoplasia: for example, for en-bloc resection of lesions with a large-caliber endoscopic resection cap that otherwise would require piecemeal MBM or for bulky lesions with possible submucosal invasion.

A limitation of this study is that participating endoscopists in the endoscopic resection training program were strongly influenced by the expert opinion of the endoscopists on the endoscopic resection training committee who acted as trainers. This is illustrated by the overlap in the learning points derived from the questionnaire and from the feedback reports (see [Table e1]). Therefore, we have not performed a quantitative analysis of the learning points. We believe that a qualitative analysis of the learning points by three expert endoscopists suffices to select the most important items. We acknowledge that this study is partly based on expert opinion, yet we feel that it provides practical tips on the endoscopic resection technique and is based on the best available data on training in esophageal endoscopic resection.


#
#

Competing interests: J. J. Bergman received grants and medical supplies from Cook Medical, Olympus Endoscopy, and Astra Zeneca. R. Bisschops received speaker’s fees from Barrx, Olympus, Fujifilm and Pentax, and is supported by a grant from FWO Vlaanderen.

Acknowledgement

The authors thanks www.Best-Academia.eu for the support.

Table e1, e2, e4

  • References

  • 1 Katada C, Muto M, Momma K et al. Clinical outcome after endoscopic mucosal resection for esophageal squamous cell carcinoma invading the muscularis mucosae – a multicenter retrospective cohort study. Endoscopy 2007; 39: 779-783
  • 2 Pech O, May A, Gossner L et al. Curative endoscopic therapy in patients with early esophageal squamous-cell carcinoma or high-grade intraepithelial neoplasia. Endoscopy 2007; 39: 30-35
  • 3 Ell C, May A, Pech O et al. Curative endoscopic resection of early esophageal adenocarcinomas (Barrett’s cancer). Gastrointest Endosc 2007; 65: 3-10
  • 4 Inoue H. Endoscopic mucosal resection for esophageal and gastric mucosal cancers. Can J Gastroenterol 1998; 12: 355-359
  • 5 Peters FP, Kara MA, Curvers WL et al. Multiband mucosectomy for endoscopic resection of Barrett’s esophagus: feasibility study with matched historical controls. Eur J Gastroenterol Hepatol 2007; 19: 311-315
  • 6 Peters FP, Brakenhoff KP, Curvers WL et al. Endoscopic cap resection for treatment of early Barrett’s neoplasia is safe: a prospective analysis of acute and early complications in 216 procedures. Dis Esophagus 2007; 20: 510-515
  • 7 May A, Gossner L, Pech O et al. Local endoscopic therapy for intraepithelial high-grade neoplasia and early adenocarcinoma in Barrett’s oesophagus: acute-phase and intermediate results of a new treatment approach. Eur J Gastroenterol Hepatol 2002; 14: 1085-1091
  • 8 Pouw RE, Wirths K, Eisendrath P et al. Efficacy of radiofrequency ablation combined with endoscopic resection for Barrett’s esophagus with early neoplasia. Clin Gastroenterol Hepatol 2010; 8: 23-29
  • 9 Shaheen NJ, Sharma P, Overholt BF et al. Radiofrequency ablation in Barrett’s esophagus with dysplasia. NEJM 2009; 360: 2277-2288
  • 10 Van Vilsteren FG, Pouw RE, Seewald S et al. Stepwise radical endoscopic resection versus radiofrequency ablation for Barrett’s oesophagus with high-grade dysplasia or early cancer: a multicentre randomised trial. Gut 2011; 60: 765-773
  • 11 Ell C, May A, Gossner L et al. Endoscopic mucosal resection of early cancer and high-grade dysplasia in Barrett’s esophagus. Gastroenterology 2000; 118: 670-677
  • 12 Van Vilsteren FG, Pouw RE, Herrero LA et al. Learning to perform endoscopic resection of esophageal neoplasia is associated with significant complications even within a structured training program. Endoscopy 2012; 44: 4-12
  • 13 Kwaliteitsinstituut voor de Gezondheidszorg CBO. Richtlijn diagnostiek en behandeling oesofaguscarcinoom. Alphen aan den Rijn Van Zuiden Communications; 2005
  • 14 Pouw RE, Bergman JJ. Endoscopic resection of early oesophageal and gastric neoplasia. Best Pract Res Clin Gastroenterol 2008; 22: 929-943
  • 15 The Paris endoscopic classification of superficial neoplastic lesions. Esophagus, stomach, and colon: November 30 to December 1, 2002. Gastrointest Endosc 2003; 58: 3-43
  • 16 Kato H, Haga S, Endo S et al. Lifting of lesions during endoscopic mucosal resection (EMR) of early colorectal cancer: implications for the assessment of resectability. Endoscopy 2001; 33: 568-573
  • 17 Alvarez HL, Pouw RE, Van Vilsteren FG et al. Safety and efficacy of multiband mucosectomy in 1060 resections in Barrett’s esophagus. Endoscopy 2011; 43: 177-183
  • 18 Pouw RE, Van Vilsteren FG, Peters FP et al. Randomized trial on endoscopic resection-cap versus multiband mucosectomy for piecemeal endoscopic resection of early Barrett’s neoplasia. Gastrointest Endosc 2011; 74: 35-43

Corresponding author

Jacques Bergman, MD PhD
Department of Gastroenterology and Hepatology
Academic Medical Center
Meibergdreef 9
1105 AZ Amsterdam
The Netherlands   
Fax: +31-20-6917033   

  • References

  • 1 Katada C, Muto M, Momma K et al. Clinical outcome after endoscopic mucosal resection for esophageal squamous cell carcinoma invading the muscularis mucosae – a multicenter retrospective cohort study. Endoscopy 2007; 39: 779-783
  • 2 Pech O, May A, Gossner L et al. Curative endoscopic therapy in patients with early esophageal squamous-cell carcinoma or high-grade intraepithelial neoplasia. Endoscopy 2007; 39: 30-35
  • 3 Ell C, May A, Pech O et al. Curative endoscopic resection of early esophageal adenocarcinomas (Barrett’s cancer). Gastrointest Endosc 2007; 65: 3-10
  • 4 Inoue H. Endoscopic mucosal resection for esophageal and gastric mucosal cancers. Can J Gastroenterol 1998; 12: 355-359
  • 5 Peters FP, Kara MA, Curvers WL et al. Multiband mucosectomy for endoscopic resection of Barrett’s esophagus: feasibility study with matched historical controls. Eur J Gastroenterol Hepatol 2007; 19: 311-315
  • 6 Peters FP, Brakenhoff KP, Curvers WL et al. Endoscopic cap resection for treatment of early Barrett’s neoplasia is safe: a prospective analysis of acute and early complications in 216 procedures. Dis Esophagus 2007; 20: 510-515
  • 7 May A, Gossner L, Pech O et al. Local endoscopic therapy for intraepithelial high-grade neoplasia and early adenocarcinoma in Barrett’s oesophagus: acute-phase and intermediate results of a new treatment approach. Eur J Gastroenterol Hepatol 2002; 14: 1085-1091
  • 8 Pouw RE, Wirths K, Eisendrath P et al. Efficacy of radiofrequency ablation combined with endoscopic resection for Barrett’s esophagus with early neoplasia. Clin Gastroenterol Hepatol 2010; 8: 23-29
  • 9 Shaheen NJ, Sharma P, Overholt BF et al. Radiofrequency ablation in Barrett’s esophagus with dysplasia. NEJM 2009; 360: 2277-2288
  • 10 Van Vilsteren FG, Pouw RE, Seewald S et al. Stepwise radical endoscopic resection versus radiofrequency ablation for Barrett’s oesophagus with high-grade dysplasia or early cancer: a multicentre randomised trial. Gut 2011; 60: 765-773
  • 11 Ell C, May A, Gossner L et al. Endoscopic mucosal resection of early cancer and high-grade dysplasia in Barrett’s esophagus. Gastroenterology 2000; 118: 670-677
  • 12 Van Vilsteren FG, Pouw RE, Herrero LA et al. Learning to perform endoscopic resection of esophageal neoplasia is associated with significant complications even within a structured training program. Endoscopy 2012; 44: 4-12
  • 13 Kwaliteitsinstituut voor de Gezondheidszorg CBO. Richtlijn diagnostiek en behandeling oesofaguscarcinoom. Alphen aan den Rijn Van Zuiden Communications; 2005
  • 14 Pouw RE, Bergman JJ. Endoscopic resection of early oesophageal and gastric neoplasia. Best Pract Res Clin Gastroenterol 2008; 22: 929-943
  • 15 The Paris endoscopic classification of superficial neoplastic lesions. Esophagus, stomach, and colon: November 30 to December 1, 2002. Gastrointest Endosc 2003; 58: 3-43
  • 16 Kato H, Haga S, Endo S et al. Lifting of lesions during endoscopic mucosal resection (EMR) of early colorectal cancer: implications for the assessment of resectability. Endoscopy 2001; 33: 568-573
  • 17 Alvarez HL, Pouw RE, Van Vilsteren FG et al. Safety and efficacy of multiband mucosectomy in 1060 resections in Barrett’s esophagus. Endoscopy 2011; 43: 177-183
  • 18 Pouw RE, Van Vilsteren FG, Peters FP et al. Randomized trial on endoscopic resection-cap versus multiband mucosectomy for piecemeal endoscopic resection of early Barrett’s neoplasia. Gastrointest Endosc 2011; 74: 35-43

Zoom Image
Fig. 1 Endoscopic resection using the endoscopic resection cap technique showing: a lifting of the submucosa; b suctioning of the lesion into the cap, with the resection snare placed in the rim of the cap; c closure of the snare during suctioning; d the resection wound after using coagulation to cut off the pseudopolyp and thereby remove the lesion.
Zoom Image
Fig. 2 Endoscopic resection using the multiband mucosectomy (MBM) technique showing: a suctioning of the lesion into the cap; b release of a rubber band during suctioning to create a pseudopolyp containing the lesion; c resection of the pseudopolyp with the hexagonal MBM snare; d the resection wound after using coagulation to cut off the pseudopolyp and thereby remove the lesion.
Zoom Image
Fig. 3 Endoscopic resection of early neoplasia in Barrett’s esophagus showing: a a macroscopic type 0-IIa + IIc lesion (Paris Classification [15]) of 15 mm in size in a C0M2 Barrett’s esophagus; b the resection wound after complete endoscopic removal of the lesion; c the endoscopic resection scar 6 weeks after the procedure.
Zoom Image
Fig. 4 Endoscopic views in two different patients with Barrett’s esophagus and early neoplastic lesions located at the gastric cardia demonstrating the benefits of retroflexion in improving the visualization of the distal margin of the lesion and aiding delineation and submucosal lifting prior to endoscopic resection. A lesion at 10 o’clock seen in: a antegrade view; b retrograde view. A lesion at 11 o’clock seen in: c antegrade view; d retrograde view.
Zoom Image
Fig. 5 Endoscopic views showing electrocoagulation markings placed to delineate the area to be resected under optimal imaging conditions prior to the resection, thereby creating a preprocedural plan that serves as a roadmap for an effective and safe endoscopic resection.
Zoom Image
Fig. 6 Endoscopic images showing how the view can be optimized using the multiband mucosectomy (MBM) technique by: a, b aligning the pulling strings of the MBM barrel with the working channel to keep them out of the endoscopic view prior to insertion of the endoscope; and c, d releasing two to three rubber bands prior to introducing the endoscope (if only one to three resections will be required).
Zoom Image
Fig. 7 Illustration of endoscopic resection using the oblique cap showing: a a lesion in the esophagus; b incorrect positioning as a lesion positioned at 12 o’clock requires an angulated position of the endoscope to suction the lesion into the cap (in the position shown suctioning may be less effective because of higher tension on the esophageal wall, which means the snare will not close at the base of the lesion, possibly resulting in only a partial/superficial resection of the lesion; c correct positioning with the endoscope rotated so that the lesion is at 6 o’clock in the endoscopic view to facilitate injection, suctioning, and resection.