- Email: [email protected]
Su1575 Western Skill Training in Endoscopic Submucosal Dissection (ESD) - an International Remote Video Based Study - the WEST ESD Study
Abstracts endoscopists with non-expert endoscopists. Results: 95 patients (mean age: 65.3 14.1 years of age) were included in the study. Of these patients, 75.8% (72/95) were men and 87.4% (83/95) were performed after bowel preparation. 76.8% (73/95) of the stigmata were located in ascending colon. Bleeding source identified were as follows (an active bleeding: 37.9%, a nonbleeding visible vessel: 23.2%, an adherent clot: 36.8%, others: 2.1%). 48.4% (46/95) were performed by expert endosopists. 6.7% (7/95) lesions (all performed by expert endoscopists) could not be adequately suctioned into the suction sup of the endoscopic ligator, and additional treatments were needed for these lesions. There was no complications related to EBL. In comparison between expert versus non-expert groups, total colonoscopy time and endocopic banding procedure time were 34 minutes [19-101] vs 34 minutes [13-72] (PZ0.60) and 14 minutes [4-45] vs 11 minutes [4-36] (PZ0.07). Conclusions: EBL can be safely and quickly performed by non-expert endoscopists. EBL is a feasible procedure even in the institution where there are few expert endoscopists.
Su1572 A Novel Endoscopic Submucosal Dissection Technique for Proton Pump Inhibitor-Refractory Gastroesophageal Reflux Disease Kazuhiro Ota*1,2, Toshihisa Takeuchi1, Satoshi Harada1, Shoko Edogawa1, Yuichi Kojima1, Takuya Inoue1, Kazuhide Higuchi1 1 Second Department of Internal Medicine, Osaka Medical College, Habikino-city, Japan; 2Shiroyama Hospital, Habikino, Japan Objectives: Although drug treatment is the usual first-line therapy for gastroesophageal reflux disease (GERD), not all patients receive satisfactory relief from drug therapy, alone. We developed an endoscopic fundoplication technique using endoscopic submucosal dissection (ESD); the technique is referred to as ESD for GERD (ESD-G). This study investigated the safety and efficacy of this novel technique in patients with drug-refractory GERD. Patients and Methods: ESD-G narrows the hiatal opening through ESD of the esophagogastric junction (EGJ) mucosa. For safety reasons, the range of mucosal resection was limited to half (1/2 or 1/4 +1/4) of the circumference of the EGJ lumen. ESD-G was performed on 15 patients with proton pump inhibitor (PPI)-refractory GERD. GERD symptoms, PPI dose, and 24-h esophageal pH monitoring results were compared before and 6 months after the procedure. Results: In 14 cases, symptoms significantly improved after ESD-G. Five patients demonstrated improved esophagitis, three were able to discontinue PPI therapy, and three were able to reduce their PPI dosage following surgery. The esophageal pH !4 holding time ratio was also decreased after ESD-G. Conclusions: ESD-G may be useful for PPI-refractory GERD patients.
Su1573 The Learning Curve for Endoscopic Submucosal Dissection of Early Esophageal Neoplasm Ching-Tai Lee*, Wen-Lun Wang, Chi-Yang Chang, Lein-Ray Mo Internal Medicine, E-Da hospital/I-Shou university, Kaohsiung county, Taiwan Introduction: Endoscopic submucosal dissection (ESD) is an advanced endoscopic procedure in resecting early esophageal neoplasm; however, it carries substantial risks, such as bleeding and perforation. Understanding of the procedural learning curve is important to the design of standardize training. This study aimed to clarify esophageal ESD’s learning curve for endoscopists. Materials and Methods: We retrospectively reviewed the clinical outcomes for consecutive patients with early esophageal neoplasms who underwent ESD by a single endoscopist since Dec 2007. One hundred and nineteen procedures were performed during the study period and were divided into four groups. Insulated-tip knife 2 (IT-knife 2) was used for ESD in the first (cases 1-30), second (cases 31-60), and third groups (cases 61-89). IT-knife Nano was used for ESD since Aug 2012 in the fourth group (cases 90-119). Individual factors, including demographic characteristics, social habits, co-morbidity, and initial tumor characters were collected. Area of dissected specimen divided by submucosal dissection time was defined as submucosal dissection speed. Procedure time, completed resection rate, submucosal dissection speed, and the associated complications were compared among the 4 periods. Results: Between Dec 2007 and Aug 2014, one hundred and nineteen esophageal ESD were performed, and four (3.4%) failed to be completed. From the first to the last group, completed resection rates were 93.1%, 93.1%, 100% and 100% respectively. The complication rate was higher in the first group (16.7%, 6.9%, 6.9%, and 0% respectively). The decrease in the total procedure times among these groups was statistically significant. The mean submucosal dissection speed significantly increased after the second group (pZ0.013). Conclusions: In our study, a two step learning curve was observed for esophageal ESD. First, one needs to accumulate 30 cases in order to acquire the necessary skills to avoid complications. Second, the endoscopist needs to accumulate at least 60 procedures in order to acquire the proficiency with better dissection speed and excellent completed resection rate. These results could be used as a reference for further esophageal ESD training.Keywords: early esophageal neoplasm, endoscopic submucosal dissection (ESD), learning curve.
www.giejournal.org
Su1574 A Novel Submucosal Injection Solution Is Superior to Saline in Facilitating ESD Performed by Western Endoscopists Rabindra R. Watson*1, Jason B. Klapman3, Srinadh Komanduri2, Lindsay Hosford5, Janak N. Shah4, Sachin Wani6, V. Raman Muthusamy1 1 Division of Digestive Diseases, University of California, Los Angeles, Los Angeles, CA; 2Division of Gastroenterology, Northwestern Memorial Hospital, Chicago, IL; 3Moffitt Cancer Center, Tampa, FL; 4Interventional Endoscopy Services, California Pacific Medical Center, San Francisco, CA; 5University of Colorado Denver, Denver, CO; 6Division of Gastroenterology, Anschutz Medical Center, Denver, CO Background: Endoscopic submucosal dissection (ESD) is the preferred method for en bloc resection of early neoplasia in the GI tract. ESD is technically challenging and time consuming, which may deter training in this technique in the West. A novel gel has been developed to facilitate ESD via sustained mucosal lifting and submucosal dissecting properties. We sought to assess the efficacy of the gel in facilitating ESD performed by western interventional endoscopists with minimal ESD experience. Specific Aims: 1) To compare the technical success and safety of ESD using the gel to ESD using normal saline, when performed by endoscopists with expertise in EMR but naive to ESD. 2) To assess the durability of the submucosal cushion produced by the novel gel compared to normal saline. Methods: A two phase porcine study was performed. In the first phase, the technical feasibility of gel-assisted ESD was assessed using a total of 9 submucosal lesions created in the stomach (6) and rectum (3) of two pigs. The duration of lift was also recorded at 10 minutes and 30 minute intervals for two unresected lesions created by gel and saline respectively. In the second phase performed 4 months later, 5 endoscopists each resected 2 gastric lesions (1 saline:1 gel) in 2 live pigs. The initial injectate was chosen in a randomized fashion. A 30 mm snare was used to estimate the size of simulated circular lesions, whose perimeters were marked with cautery. A hybrid knife (Erbe, Tubingen Germany) was used to perform ESD in all cases. Procedure length, volume of injectate, number of injections, lesion size, and adverse events were recorded. Results: In the first phase, successful ESD was performed in 12/12 (100%) of lesions in a mean procedure time of 29 9 minutes. The mean lesion size resected was 26.1 13.6 cm. Three perforations occurred, all successfully treated with clip closure. The cushion using the gel remained unchanged over the observation period of 126 minutes, compared with partial dissipation of saline observed at 10 minutes and total dissipation observed at the 60 minute mark. In the second phase, all 10 resections were successfully performed in both the saline and gel arms. A trend for shorter procedure time was seen in the gel group (25.2 8.4 min vs 48.2 26.2 min, p Z 0.115). Use of saline required 2.6 injections versus a single injection using the gel. No adverse events occurred in the gel group. A single perforation occurred in the saline group treated with clip closure. Conclusions: In a live porcine model, a novel injection gel facilitated successful ESD performed by a cohort of western endoscopists naïve to this technique. The gel resulted in nearly 50% reduction in procedure times and technically successful and safe resection was observed over a very short learning curve. Further study of the use of this gel for training in ESD is warranted.
Su1575 Western Skill Training in Endoscopic Submucosal Dissection (ESD) - an International Remote Video Based Study - the WEST ESD Study Amit Bhatt*1, Seiichiro Abe2, Arthi Kumaravel1, Mansour A. Parsi1, Sunguk Jang1, Tyler Stevens1, Gregory Zuccaro1, Ichiro Oda2, John J. Vargo1, Yutaka Saito2 1 Gastroenterology, Cleveland Clinic, Cleveland, OH; 2Endoscopy Division, National Cancer Center Hospital, Tokyo, Japan Background: There is increasing interest and practice of ESD in the West. ESD has a flat learning curve, and inadequate training may increase the risk of incomplete resection and perforation. In Japan, ESD is learned under the supervision of experts with hands on and verbal support. This learning structure is extremely limited in the West due to lack of experts and low case volume. While currently Western ESD training is accomplished in a porcine model, little is known on learning curves and achieving competency in this model. We analyzed the ESD learning curves of 2 Western endoscopists using a porcine model under supervision of Japanese ESD experts through a remote video based system. Methods: Two US endoscopists, one experienced advanced endoscopist who had performed O 10,000 advanced procedures and a third year gastroenterology fellow, were trained in ESD in a porcine model. Training sessions were performed on a weekly/biweekly basis. The following equipment was available; ERBE VIO 300D (ERBE, Tuebingen, Germany) electrosurgical generator, IT knife 2 (Olympus KD-611L, Tokyo Japan), Dual knife (Olympus KD-650L / KD-650U, Tokyo Japan), Hook Knife (Olympus KD-620LR / KD620UR, Tokyo Japan), and HybridKnife (ERBE, Germany). The waterjet function of the ERBE knife was not permitted as this may influence procedure time. Normal saline with indigo carmine was used as a lifting solution in all procedures. ESD was performed on 3cm lesions in the proximal and middle stomach. Each procedure was limited to 1 hour. Each training session was recorded and blindly analyzed by Japanese ESD experts. Each video was scored in marginal incision, submucosal
Volume 81, No. 5S : 2015 GASTROINTESTINAL ENDOSCOPY AB335
Abstracts
dissection and overall procedure quality using a 100mm VAS scale. Critical feedback was given after each ESD procedure. Time, en-bloc resection rate, perforation and completion rates were recorded for each procedure. Competency was defined as consistently (O3cases in a row) resecting a 3cm lesion within 30 minutes, en-bloc and without perforation. Results: Two endoscopists performed a total of 55 ESD procedures (Endoscopist 1 performed 30 and Endoscopist 2 performed 25). Table 1 presents a summary of ESD characteristics and scores by time period. Period A: 1-10 procedures, Period B: 11-20 procedures and Period C: 20-30 procedures. Figure 1 show the loess fit plots for scores. Competency was achieved by endoscopist 1 at procedure 23 and endoscopist 2 at procedure 25. Conclusion: Competency in ESD can be accomplished by porcine model training in the West. With the aid of remote based video analysis and critical feedback from ESD experts both our endoscopists achieved competency by 25 procedures. Table 1. Procedure Characteristics and Scoring Factor Time to completion (min) Procedure completed in 60 minutes Procedure completed in 30 minutes En bloc resection Perforation Completed in %30 min withen bloc resection and no perforation Scores Marginal incision Submucosal dissection technique Overall ESD technique
Total (N[55)
Period A (N[20)
Period B (N[20)
Period C (N[15)
42.215.9 57.610.12,3 38.410.71,3 26.88.21,2 45(81.8) 11(55.0)2,3 19(95.0)1 15(100.0)1 15(27.3)
0(0.0)3
4(20.0)3
11(73.3)12
38(69.1) 8(14.5) 14(25.5)
5(25.0)2,3 3(15.0) 0(0.0)3
18(90.0)1 5(25.0) 3(15.0)3
15(100.0)1 0(0.0) 11(73.3)1,2
to present our results of upper GI tract ESD procedures. Methods: A total of 280 ESD procedures, which were performed in the esophagus, stomach, colorectal and duodenum between April 2012 and September 2014, were recorded prospectively before and after the procedure (figure). After 30 gastric ESD procedures had been performed, ESD of other parts of GI were performed by order of rectum, esophagus and other colon segments. Duodenal ESD was started after 224 ESD procedures. The results of 130 ESDs performed in esophagus, stomach and duodenum were analyzed retrospectively. Results: In a total of 124 patients, 109 gastric, 13 esophageal and 8 duodenal ESDs were performed (Figure). The overall en-bloc and complete resection rates were 93.8% and 93.1%, respectively.Histopathology revealed carcinoma in 34 and adenoma in 46 patients. Complete resection was not achieved in 7 patients with positive vertical border. Gastric perforation occurred in one patient and duodenal perforation in another which were treated successfully with endoscopic clip without the need for surgery. Endoscopic mucosal resection was performed to two patients with recurrence. Although cover type knife was generally preferred during the first cases noncover type knife was used for later procedures especially for esophageal and duodenal lesions. Snare was used in only 7 patients (Table). Conclusion: According to guideline recommendations and masters for ESD, if ESD training is initiated and continued, successful ESD may be performed in localized lesions in the upper GI
Demographic data and endoscopic submucosal dissection results [Case (n)[124, Lesion (N)[130] N[130 Lesion size, mm, mean (SD) (median; range) Tissue size, mm, mean (SD) (median; range)
72.320.7 53.317.62,3 68.019.3 49.514.92,3
79.314.51 76.812.91
88.39.01 81.011.41
70.818.4 52.413.22,3
78.512.91
85.07.11
Period A: 0-10 procedures performed; Period B: 11-20 procedures performed; Period C: 21-30 procedures performed. Values presented as Mean SD or N (column %). 1: Significantly different from Period A. 2: Significantly different from Period B. 3: Significantly different from Period C. A significance level of 0.017 was used for pairwise ad-hoc comparisons.
Figure 1.
Duration of procedure, min, mean (SD) (median; range) Dissection speed, mm!+O2!+O/min, mean (SD) (median; range) En-Bloc resection rate, N (%) Complete Resection, N (%) Adverse Events, N (%) -Delayed bleeding -Perforation -Delayed perforation Localization, N (%) Esophagus Gastric -Upper -Middle -Lower Duodenum First part Second part Pathology, N (%) Carcinoma Gastric adenoma GIST Neuroendocrine tumor Ectopic pancreas Hyperplastic polyp Leiomyoma Lipoma Granular Cell Tumor
23.08 (11,89) (20; 8-85) 33.08 (16.08) (30; 10-100) 1.13 (42.93) (40; 6-345) 18.99 (16.81) (13.38; 1.12-86.35) 122 (93.8) 121 (93.1) 0 (0) 2 (1.5) 0 (0) 13 (10.0) 109 (83.8) 11 (8.5) 35 (26.9) 63 (48.5) 8 (6.2) 5 (3.9) 3 (2,3) 34 (26.2) 46 (35.4) 6 (4.6) 12 (9.2) 7 (5.4) 4 (3.1) 11 (8.5) 6 (4.6) 1 (0.8)
SD; standard deviation. GIST: gastrointestinal stromal tumor.
Su1576 Endoscopic Submucosal Dissection in Upper Gastrointestinal Lesions: Experience of 130 Cases From a Tertiary Reference Center in Turkey Fatih Aslan1, Zehra Akpinar1, Cem Cekic1, Mehmet Camci1, Mustafa Kartal1, Hakan Camyar1, Emrah Alper*1, Nese Ekinci2, Mahmut Arabul1, Belkis Unsal1 1 Gastroenterology, Katip Celebi University Ataturk Training and Research Hospital, Izmir, Turkey; 2Pathology, Katip Celebi University Ataturk Training and Research Hospital, Izmir, Turkey Background: Endoscopic submucosal dissection (ESD) is minimally invasive technique, providing en-bloc resection of premalignant and malignant lesions in early stage gastrointestinal (GI) cancers. This procedure is commonly used in Far East countries with high gastric cancer prevalence like Japan and Korea; however, it has come into use in western countries in recent years. In the present study, we aimed
AB336 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 5S : 2015
www.giejournal.org
Su1572 A Novel Endoscopic Submucosal Dissection Technique for Proton Pump Inhibitor-Refractory Gastroesophageal Reflux Disease Kazuhiro Ota*1,2, Toshihisa Takeuchi1, Satoshi Harada1, Shoko Edogawa1, Yuichi Kojima1, Takuya Inoue1, Kazuhide Higuchi1 1 Second Department of Internal Medicine, Osaka Medical College, Habikino-city, Japan; 2Shiroyama Hospital, Habikino, Japan Objectives: Although drug treatment is the usual first-line therapy for gastroesophageal reflux disease (GERD), not all patients receive satisfactory relief from drug therapy, alone. We developed an endoscopic fundoplication technique using endoscopic submucosal dissection (ESD); the technique is referred to as ESD for GERD (ESD-G). This study investigated the safety and efficacy of this novel technique in patients with drug-refractory GERD. Patients and Methods: ESD-G narrows the hiatal opening through ESD of the esophagogastric junction (EGJ) mucosa. For safety reasons, the range of mucosal resection was limited to half (1/2 or 1/4 +1/4) of the circumference of the EGJ lumen. ESD-G was performed on 15 patients with proton pump inhibitor (PPI)-refractory GERD. GERD symptoms, PPI dose, and 24-h esophageal pH monitoring results were compared before and 6 months after the procedure. Results: In 14 cases, symptoms significantly improved after ESD-G. Five patients demonstrated improved esophagitis, three were able to discontinue PPI therapy, and three were able to reduce their PPI dosage following surgery. The esophageal pH !4 holding time ratio was also decreased after ESD-G. Conclusions: ESD-G may be useful for PPI-refractory GERD patients.
Su1573 The Learning Curve for Endoscopic Submucosal Dissection of Early Esophageal Neoplasm Ching-Tai Lee*, Wen-Lun Wang, Chi-Yang Chang, Lein-Ray Mo Internal Medicine, E-Da hospital/I-Shou university, Kaohsiung county, Taiwan Introduction: Endoscopic submucosal dissection (ESD) is an advanced endoscopic procedure in resecting early esophageal neoplasm; however, it carries substantial risks, such as bleeding and perforation. Understanding of the procedural learning curve is important to the design of standardize training. This study aimed to clarify esophageal ESD’s learning curve for endoscopists. Materials and Methods: We retrospectively reviewed the clinical outcomes for consecutive patients with early esophageal neoplasms who underwent ESD by a single endoscopist since Dec 2007. One hundred and nineteen procedures were performed during the study period and were divided into four groups. Insulated-tip knife 2 (IT-knife 2) was used for ESD in the first (cases 1-30), second (cases 31-60), and third groups (cases 61-89). IT-knife Nano was used for ESD since Aug 2012 in the fourth group (cases 90-119). Individual factors, including demographic characteristics, social habits, co-morbidity, and initial tumor characters were collected. Area of dissected specimen divided by submucosal dissection time was defined as submucosal dissection speed. Procedure time, completed resection rate, submucosal dissection speed, and the associated complications were compared among the 4 periods. Results: Between Dec 2007 and Aug 2014, one hundred and nineteen esophageal ESD were performed, and four (3.4%) failed to be completed. From the first to the last group, completed resection rates were 93.1%, 93.1%, 100% and 100% respectively. The complication rate was higher in the first group (16.7%, 6.9%, 6.9%, and 0% respectively). The decrease in the total procedure times among these groups was statistically significant. The mean submucosal dissection speed significantly increased after the second group (pZ0.013). Conclusions: In our study, a two step learning curve was observed for esophageal ESD. First, one needs to accumulate 30 cases in order to acquire the necessary skills to avoid complications. Second, the endoscopist needs to accumulate at least 60 procedures in order to acquire the proficiency with better dissection speed and excellent completed resection rate. These results could be used as a reference for further esophageal ESD training.Keywords: early esophageal neoplasm, endoscopic submucosal dissection (ESD), learning curve.
www.giejournal.org
Su1574 A Novel Submucosal Injection Solution Is Superior to Saline in Facilitating ESD Performed by Western Endoscopists Rabindra R. Watson*1, Jason B. Klapman3, Srinadh Komanduri2, Lindsay Hosford5, Janak N. Shah4, Sachin Wani6, V. Raman Muthusamy1 1 Division of Digestive Diseases, University of California, Los Angeles, Los Angeles, CA; 2Division of Gastroenterology, Northwestern Memorial Hospital, Chicago, IL; 3Moffitt Cancer Center, Tampa, FL; 4Interventional Endoscopy Services, California Pacific Medical Center, San Francisco, CA; 5University of Colorado Denver, Denver, CO; 6Division of Gastroenterology, Anschutz Medical Center, Denver, CO Background: Endoscopic submucosal dissection (ESD) is the preferred method for en bloc resection of early neoplasia in the GI tract. ESD is technically challenging and time consuming, which may deter training in this technique in the West. A novel gel has been developed to facilitate ESD via sustained mucosal lifting and submucosal dissecting properties. We sought to assess the efficacy of the gel in facilitating ESD performed by western interventional endoscopists with minimal ESD experience. Specific Aims: 1) To compare the technical success and safety of ESD using the gel to ESD using normal saline, when performed by endoscopists with expertise in EMR but naive to ESD. 2) To assess the durability of the submucosal cushion produced by the novel gel compared to normal saline. Methods: A two phase porcine study was performed. In the first phase, the technical feasibility of gel-assisted ESD was assessed using a total of 9 submucosal lesions created in the stomach (6) and rectum (3) of two pigs. The duration of lift was also recorded at 10 minutes and 30 minute intervals for two unresected lesions created by gel and saline respectively. In the second phase performed 4 months later, 5 endoscopists each resected 2 gastric lesions (1 saline:1 gel) in 2 live pigs. The initial injectate was chosen in a randomized fashion. A 30 mm snare was used to estimate the size of simulated circular lesions, whose perimeters were marked with cautery. A hybrid knife (Erbe, Tubingen Germany) was used to perform ESD in all cases. Procedure length, volume of injectate, number of injections, lesion size, and adverse events were recorded. Results: In the first phase, successful ESD was performed in 12/12 (100%) of lesions in a mean procedure time of 29 9 minutes. The mean lesion size resected was 26.1 13.6 cm. Three perforations occurred, all successfully treated with clip closure. The cushion using the gel remained unchanged over the observation period of 126 minutes, compared with partial dissipation of saline observed at 10 minutes and total dissipation observed at the 60 minute mark. In the second phase, all 10 resections were successfully performed in both the saline and gel arms. A trend for shorter procedure time was seen in the gel group (25.2 8.4 min vs 48.2 26.2 min, p Z 0.115). Use of saline required 2.6 injections versus a single injection using the gel. No adverse events occurred in the gel group. A single perforation occurred in the saline group treated with clip closure. Conclusions: In a live porcine model, a novel injection gel facilitated successful ESD performed by a cohort of western endoscopists naïve to this technique. The gel resulted in nearly 50% reduction in procedure times and technically successful and safe resection was observed over a very short learning curve. Further study of the use of this gel for training in ESD is warranted.
Su1575 Western Skill Training in Endoscopic Submucosal Dissection (ESD) - an International Remote Video Based Study - the WEST ESD Study Amit Bhatt*1, Seiichiro Abe2, Arthi Kumaravel1, Mansour A. Parsi1, Sunguk Jang1, Tyler Stevens1, Gregory Zuccaro1, Ichiro Oda2, John J. Vargo1, Yutaka Saito2 1 Gastroenterology, Cleveland Clinic, Cleveland, OH; 2Endoscopy Division, National Cancer Center Hospital, Tokyo, Japan Background: There is increasing interest and practice of ESD in the West. ESD has a flat learning curve, and inadequate training may increase the risk of incomplete resection and perforation. In Japan, ESD is learned under the supervision of experts with hands on and verbal support. This learning structure is extremely limited in the West due to lack of experts and low case volume. While currently Western ESD training is accomplished in a porcine model, little is known on learning curves and achieving competency in this model. We analyzed the ESD learning curves of 2 Western endoscopists using a porcine model under supervision of Japanese ESD experts through a remote video based system. Methods: Two US endoscopists, one experienced advanced endoscopist who had performed O 10,000 advanced procedures and a third year gastroenterology fellow, were trained in ESD in a porcine model. Training sessions were performed on a weekly/biweekly basis. The following equipment was available; ERBE VIO 300D (ERBE, Tuebingen, Germany) electrosurgical generator, IT knife 2 (Olympus KD-611L, Tokyo Japan), Dual knife (Olympus KD-650L / KD-650U, Tokyo Japan), Hook Knife (Olympus KD-620LR / KD620UR, Tokyo Japan), and HybridKnife (ERBE, Germany). The waterjet function of the ERBE knife was not permitted as this may influence procedure time. Normal saline with indigo carmine was used as a lifting solution in all procedures. ESD was performed on 3cm lesions in the proximal and middle stomach. Each procedure was limited to 1 hour. Each training session was recorded and blindly analyzed by Japanese ESD experts. Each video was scored in marginal incision, submucosal
Volume 81, No. 5S : 2015 GASTROINTESTINAL ENDOSCOPY AB335
Abstracts
dissection and overall procedure quality using a 100mm VAS scale. Critical feedback was given after each ESD procedure. Time, en-bloc resection rate, perforation and completion rates were recorded for each procedure. Competency was defined as consistently (O3cases in a row) resecting a 3cm lesion within 30 minutes, en-bloc and without perforation. Results: Two endoscopists performed a total of 55 ESD procedures (Endoscopist 1 performed 30 and Endoscopist 2 performed 25). Table 1 presents a summary of ESD characteristics and scores by time period. Period A: 1-10 procedures, Period B: 11-20 procedures and Period C: 20-30 procedures. Figure 1 show the loess fit plots for scores. Competency was achieved by endoscopist 1 at procedure 23 and endoscopist 2 at procedure 25. Conclusion: Competency in ESD can be accomplished by porcine model training in the West. With the aid of remote based video analysis and critical feedback from ESD experts both our endoscopists achieved competency by 25 procedures. Table 1. Procedure Characteristics and Scoring Factor Time to completion (min) Procedure completed in 60 minutes Procedure completed in 30 minutes En bloc resection Perforation Completed in %30 min withen bloc resection and no perforation Scores Marginal incision Submucosal dissection technique Overall ESD technique
Total (N[55)
Period A (N[20)
Period B (N[20)
Period C (N[15)
42.215.9 57.610.12,3 38.410.71,3 26.88.21,2 45(81.8) 11(55.0)2,3 19(95.0)1 15(100.0)1 15(27.3)
0(0.0)3
4(20.0)3
11(73.3)12
38(69.1) 8(14.5) 14(25.5)
5(25.0)2,3 3(15.0) 0(0.0)3
18(90.0)1 5(25.0) 3(15.0)3
15(100.0)1 0(0.0) 11(73.3)1,2
to present our results of upper GI tract ESD procedures. Methods: A total of 280 ESD procedures, which were performed in the esophagus, stomach, colorectal and duodenum between April 2012 and September 2014, were recorded prospectively before and after the procedure (figure). After 30 gastric ESD procedures had been performed, ESD of other parts of GI were performed by order of rectum, esophagus and other colon segments. Duodenal ESD was started after 224 ESD procedures. The results of 130 ESDs performed in esophagus, stomach and duodenum were analyzed retrospectively. Results: In a total of 124 patients, 109 gastric, 13 esophageal and 8 duodenal ESDs were performed (Figure). The overall en-bloc and complete resection rates were 93.8% and 93.1%, respectively.Histopathology revealed carcinoma in 34 and adenoma in 46 patients. Complete resection was not achieved in 7 patients with positive vertical border. Gastric perforation occurred in one patient and duodenal perforation in another which were treated successfully with endoscopic clip without the need for surgery. Endoscopic mucosal resection was performed to two patients with recurrence. Although cover type knife was generally preferred during the first cases noncover type knife was used for later procedures especially for esophageal and duodenal lesions. Snare was used in only 7 patients (Table). Conclusion: According to guideline recommendations and masters for ESD, if ESD training is initiated and continued, successful ESD may be performed in localized lesions in the upper GI
Demographic data and endoscopic submucosal dissection results [Case (n)[124, Lesion (N)[130] N[130 Lesion size, mm, mean (SD) (median; range) Tissue size, mm, mean (SD) (median; range)
72.320.7 53.317.62,3 68.019.3 49.514.92,3
79.314.51 76.812.91
88.39.01 81.011.41
70.818.4 52.413.22,3
78.512.91
85.07.11
Period A: 0-10 procedures performed; Period B: 11-20 procedures performed; Period C: 21-30 procedures performed. Values presented as Mean SD or N (column %). 1: Significantly different from Period A. 2: Significantly different from Period B. 3: Significantly different from Period C. A significance level of 0.017 was used for pairwise ad-hoc comparisons.
Figure 1.
Duration of procedure, min, mean (SD) (median; range) Dissection speed, mm!+O2!+O/min, mean (SD) (median; range) En-Bloc resection rate, N (%) Complete Resection, N (%) Adverse Events, N (%) -Delayed bleeding -Perforation -Delayed perforation Localization, N (%) Esophagus Gastric -Upper -Middle -Lower Duodenum First part Second part Pathology, N (%) Carcinoma Gastric adenoma GIST Neuroendocrine tumor Ectopic pancreas Hyperplastic polyp Leiomyoma Lipoma Granular Cell Tumor
23.08 (11,89) (20; 8-85) 33.08 (16.08) (30; 10-100) 1.13 (42.93) (40; 6-345) 18.99 (16.81) (13.38; 1.12-86.35) 122 (93.8) 121 (93.1) 0 (0) 2 (1.5) 0 (0) 13 (10.0) 109 (83.8) 11 (8.5) 35 (26.9) 63 (48.5) 8 (6.2) 5 (3.9) 3 (2,3) 34 (26.2) 46 (35.4) 6 (4.6) 12 (9.2) 7 (5.4) 4 (3.1) 11 (8.5) 6 (4.6) 1 (0.8)
SD; standard deviation. GIST: gastrointestinal stromal tumor.
Su1576 Endoscopic Submucosal Dissection in Upper Gastrointestinal Lesions: Experience of 130 Cases From a Tertiary Reference Center in Turkey Fatih Aslan1, Zehra Akpinar1, Cem Cekic1, Mehmet Camci1, Mustafa Kartal1, Hakan Camyar1, Emrah Alper*1, Nese Ekinci2, Mahmut Arabul1, Belkis Unsal1 1 Gastroenterology, Katip Celebi University Ataturk Training and Research Hospital, Izmir, Turkey; 2Pathology, Katip Celebi University Ataturk Training and Research Hospital, Izmir, Turkey Background: Endoscopic submucosal dissection (ESD) is minimally invasive technique, providing en-bloc resection of premalignant and malignant lesions in early stage gastrointestinal (GI) cancers. This procedure is commonly used in Far East countries with high gastric cancer prevalence like Japan and Korea; however, it has come into use in western countries in recent years. In the present study, we aimed
AB336 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 5S : 2015
www.giejournal.org