Published papers

Background and Aims: Although lesions occupying a large circumference are associated with the risk of post–endoscopic submucosal dissection (ESD) strictures, the corresponding data for duodenal lesions are unknown. We aimed to analyze the incidence of post-ESD strictures after wide-field duodenal ESD. Methods: In this retrospective study of duodenal lesions treated with ESD between July 2010 and August 2021, we included lesions that resulted in mucosal defects occupying more than half of the circumference and excluded lesions located in bulbs and involving Vater's papilla. We analyzed the incidence rates of stricture and luminal narrowing, in addition to bleeding and perforation as the outcomes in this study. Stricture was defined as the inability of the endoscope to pass through the lumen. A single endoscopist reviewed all endoscopic images and judged the degree of luminal narrowing. Results: Eighty lesions were included, of which 2 involved mucosal defects occupying more than 90% of the circumference. The wound was closed, at least partially, in 90% in all lesions and in 86% of lesions with a mucosal defect occupying more than 75% of the circumference. None of the lesions caused delayed perforation and stricture, whereas 2 lesions caused delayed bleeding. Only 6 lesions caused luminal narrowing. When examined by the degree of closure, the rate of luminal narrowing increased with complete closure, incomplete closure, and nonclosure (4.9%, 9.1%, and 25.0%, respectively). Conclusions: Suturing may prevent post-ESD bleeding and perforation as well as stricture formation in cases of duodenal tumors, with mucosal defects occupying a large circumference.

The incidence of duodenal tumors (DTs) is increasing. However, the mechanisms underlying its development remain unclear. Environmental factors, including the microbiome and bile acids (BAs), are believed to influence tumor development. Therefore, we conducted a single-center, prospective, observational study to investigate the potential differences between patients with DTs and healthy controls (HCs) based on these factors. In addition, the BAs in the duodenal fluid were measured using liquid chromatography-tandem mass spectrometry. We recruited 41 patients and performed 16S rRNA-seq. There was no difference in the observed ASVs or PCoA plot of Bray–Curtis dissimilarity between the DTs and HCs. The lithocholic acid concentration was significantly lower in the DT group than in the control group. The ratio of CDCA to LCA was significantly higher in patients with DTs. No significant differences in microbiota were observed between DTs and HCs. In patients with DTs, the lithocholic acid concentration in duodenal was significantly lower than in HCs.

Background: A standard treatment method for pedunculated duodenal lesions has not yet been established. This study aimed to evaluate the feasibility of endoscopic resection (ER) for pedunculated duodenal lesions, especially for large lesions. Methods: This study retrospectively reviewed cases of pedunculated duodenal lesions treated with ER at our institute between July 2010 and January 2021. We collected data on the clinical characteristics and treatment outcomes. In addition, the cases were divided based on the treatment provided for the lesion, i.e., the ESD and snare resection groups, and we compared the data between the two groups. Results: Thirty-eight cases were included in this study. The overall en bloc resection rate was 95%. No perforation occurred in any of the cases. There were 10 cases treated with ESD, which were considered difficult to achieve en bloc resection using snare forceps, and 28 cases treated with snare resection. Although the median lesion size was significantly larger in the ESD group than in the snare resection group (27 [range 23–66] vs. 19 [range 6–55] mm, P = 0.0052), treatment outcomes were not significantly different between the two groups. Among the 10 cases treated using ESD, en bloc resection was achieved in all cases, along with specimen retrieval without any perforations, despite the large size of the lesions. Conclusion: ER could be performed to achieve a high en bloc resection rate without any perforations in pedunculated duodenal lesions, even in extremely large (e.g., ≧ 60 mm) lesions, suggesting that ER is feasible and may be an alternative to surgical resection for large pedunculated duodenal lesions.

Objectives: Endoscopic papillectomy (EP) is a low-invasive treatment for duodenal tumors including papilla. The limit of lesion size and local recurrence have been issues in EP. We developed endoscopic submucosal dissection (ESD) for the duodenal tumors including papilla (ESDIP: ESD including papilla) to overcome the problems. The aim of this study was to evaluate the feasibility of ESDIP. Methods: We included the patients who underwent ESDIP from August 2010 to January 2024 in this study. We evaluated the retrospective characteristics of patients and lesions, clinical results of ESDIP and of endoscopic retrograde cholangiopancreatography (ERCP) as prevention for delayed adverse events, and pathological findings. We also calculated the cumulative recurrence rate and overall survival rate at 12 months after ESDIP. Results: Fifty-four patients were included in this study. The mean lesion size was 39 mm. The third-quarter cases revealed a less than half-circumferential lesion, and the one case with a full-circumferential lesion. Resection was accomplished in 96% (n = 52), and also the en-bloc resection rate was 96%. Of the cases in accomplished resection, 98% of patients were intubated with a pancreaticobiliary drainage tube by ERCP. Intraprocedural perforation occurred in eight cases. Delayed bleeding occurred in 10 cases. Delayed perforation was seen only in one case. The incidence of post-ERCP pancreatitis was 25%. Cumulative local recurrence rate and the overall survival rate were 15% and 96%, respectively. Conclusion: ESDIP may be feasible for duodenal tumors including papilla, and is a potential alternative option to avoid pancreaticoduodenectomy.

Background A modified double-layered suturing procedure (origami method; OGM) can achieve robust closure, even for large mucosal defects, by folding the muscle layer using through-the-scope clips. This study aimed to evaluate the feasibility of OGM in duodenal endoscopic submucosal dissection (ESD). Methods This retrospective, observational study was conducted at a tertiary care hospital. We reviewed cases of OGM attempted after duodenal ESD between June 2022 and April 2023.We excluded lesions located in the duodenal bulb or involving the major papilla. We measured the clinical characteristics and outcomes. Results OGM was attempted in 28 cases after duodenal ESD. The median mucosal defect size was 38mm (range 26 ?110). Complete closure was achieved in 27 cases (96%), including the largest lesion. The median closure time was 16 minutes. There were no perforations caused by clips. Delayed perforation and bleeding were not observed. Among 28 cases, 21 underwent follow-up endoscopy 3 5 days after ESD, and the muscle layer remained folded in all cases. Conclusion OGM achieved a high rate of complete closure even after duodenal ESD. Closure with OGM was robust enough to maintain the folded muscle layer for at least 3 days postoperatively.

Background and Aim: Multiple duodenal polyposis associated with familial adenomatous polyposis (FAP) is a high risk of duodenal cancer development. We evaluated the feasibility of intensive endoscopic resection that is a comprehensive treatment strategy combining multiple kinds of endoscopic treatments. Methods: This is a retrospective observational study. From January 2012 to July 2022, a total of 28 consecutive patients in FAP who underwent endoscopic resection for multiple duodenal polyposis more than twice were included. Various endoscopic treatments, such as cold polypectomy (CP), endoscopic mucosal resection (EMR), underwater EMR (UEMR), endoscopic submucosal dissection (ESD), and endoscopic papillectomy (EP), were applied depending on lesions size and location. We evaluated individual information from patients' medical records, including patient characteristics, lesion characteristics, details of endoscopic treatment, pathologic findings, and Spigelman index (SI). We compared the differences in the number of treatments and observation periods with and without SI decrease. Results: A total of 1040 lesions were removed by 138 sessions of endoscopic resections. The median follow-up period was 3.2 years. At the beginning of the endoscopic intervention, median SI was 9 (6–11) and the proportion of Spigelman stage (SS) IV was 61%. Repeated endoscopic treatments finally reduced SI in 26 patients (93%), and the proportion of SS IV significantly decreased to 13% with every endoscopic treatment. The mean SI change was −4.2 points per year (95% confidence interval: −0.6 to −5.9). There were no patients required surgical duodenectomy during the follow-up period. Conclusion: Intensive resection has a potential of downstaging duodenal lesions associated with FAP.

Background and Aims: Through-the-scope clips (TTSCs) are among the most common devices used to close colorectal post–endoscopic submucosal dissection (ESD) defects. Although TTSCs are easy to deliver, even to the proximal colon, simple clipping of large defects results in mucosa–mucosa closure, which could be unreliable. We developed a novel endoscopic closure technique using a modified double-layered suturing called the origami method (OGM). This method closes not only the mucosal layer but also the muscle layer with only TTSCs, which can obtain robust closure even for large defects. This study aimed to evaluate the feasibility of this new closure method for colorectal post-ESD defects. Methods: This retrospective observational study was conducted at a tertiary care hospital. We reviewed the cases of the OGM attempted after colorectal ESD at our institute between October 2021 and October 2022 and measured the clinical characteristics and outcomes of enrolled cases. Results: The OGM was attempted in 47 cases after colorectal ESD. Thirty-one cases (66%) were in the proximal colon, 5 (11%) in the distal colon, 6 (13%) in the upper rectum, and 5 (10%) in the lower rectum. The median defect size was 38 mm, with the largest being 85 mm. Complete closure was achieved in 44 cases (94%), including all lower rectum cases. No perforations were caused by clips during closure, and delayed perforation and bleeding were not observed. Conclusions: This new closure method is feasible and recommended. The OGM could achieve reliable closure of large defects in any location, including the proximal colon and thick-walled lower rectum, using only TTSCs.

Duodenal endoscopic submucosal dissection (ESD) is a high-risk technique; however, prophylactic closure of mucosal defects reduces the risk. Unfortunately, we have encountered cases where closure is difficult, especially in large lesions. Therefore, we developed a novel closure technique, a string clip suturing method with an anchor (SCSM-A). This study aimed to elucidate the feasibility of this method. Five patients underwent this method for the closure of mucosal defects after duodenal ESD. The initial string clip was deployed at the anal end of the mucosal defects and the second clip was deployed at the other end of the mucosal defect. A third clip was deployed on the muscular layer in the middle of the mucosal defect. The free end of the string was pulled, and additional clips were deployed around the first to the third clips for complete closure. Because of grasping the muscle layer, SCSM-A can be employed for secure closure without creating a pocket. We reviewed the background and clinical course of hospitalization of patients who underwent this method. The resected specimens ranged from 52 to 103 mm in diameter. Complete closure of the mucosal defects was possible in all the cases. There were no adverse events, and no cases required additional treatment. All the patients were discharged within 7 days. The new method achieved secure closure even for large mucosal defects after duodenal ESD. This is a technique that can be applied to other organs, e.g., the colon.

Water pressure method (WPM) is useful for colorectal endoscopic submucosal dissection (ESD), characterized not only by underwater conditions but also by active water pressure via the waterjet function. However, the extension of the colorectum by injecting excess water and contaminating the operative field by stool and bleeding have been issues. This study aimed to evaluate the feasibility of a novel perfusion system using a continuous liquid-suction catheter attachment (CLCA) in colorectal ESD with WPM. We retrospectively reviewed cases in which the perfusion system was used in colorectal ESD with WPM between August 2022 and September 2023. We evaluated clinical characteristics, treatment outcomes, volume of injection by the waterjet function, volume of suction by the endoscope and CLCA, and concentration of floating matter in the operative field over time. Thirty-one cases were enrolled. The median lesion size was 30 (range, 15–100) mm. In all cases, en bloc resection was achieved without perforation. The median injection volume was 2312 (range, 1234–13,866) g. The median suction volumes by the endoscope and CLCA were 918 (range, 141–3162) and 1147 (range, 254–11,222) g, respectively. The median concentration of floating matter in the operative field (measured in 15 cases) was 15.3 (range, 7.3–112) mg/mL when the endoscope arrived at the lesion and 8.0 (range, 3.2–16) mg/mL after endoscopically washing at the beginning of the ESD. It ranged from 7.6 to 13.4 mg/dL every 20 min during ESD. This perfusion system could prevent the extension of the lumen and maintain a good field of view in colorectal ESD with WPM.

Objectives: Although endoscopic submucosal dissection (ESD) training is important, quantitative assessments have not been established. This study aimed to explore a novel quantitative assessment system by analyzing an electrical surgical unit (ESU). Methods: This was an ex vivo study. Step one: to identify the novel efficiency indicators, 20 endoscopists performed one ESD each, and we analyzed correlations between their resection speed and electrical status. Step two: to identify the novel precision indicators, three experts and three novices performed one ESD each, and we compared the stability of the electrical status. Step three: three novices in step two performed 19 additional ESDs, and we analyzed the learning curve using novel indicators. Results: Step one: the percentage of total activation time (AT) of ESU in the procedure time (β coefficient, 0.80; P < 0.01) and AT required for submucosal dissection (β coefficient, −0.57; P < 0.01) were significantly correlated with the resection speed. Step two: coefficient of variation of the AT per one pulse (0.16 [range, 0.13–0.17] vs. 0.26 [range, 0.20–0.41], P = 0.049) and coefficient of variation of the peak electric power per pulse during mucosal incision (0.14 [range, 0.080–0.15] vs. 0.25 [range, 0.24–0.28], P = 0.049) were significantly lower in the experts than in the novices. Regarding the learning curve, the percentage of total AT of ESU in the procedure time and AT required for submucosal dissection had a trend of improvement. Conclusion: Novel indicators identified by analyzing ESU enable quantitative assessment for endoscopist's skill.

Objectives: Fever and increased inflammatory responses sometimes occur following endoscopic resection (ER). However, the differences in causes according to the organ are scarcely understood, and several modified ER techniques have been proposed. Therefore, we conducted a comprehensive prospective study to investigate the cause of fever and increased inflammatory response across multiple organs after ER. Methods: We included patients who underwent gastrointestinal endoscopic submucosal dissection (ESD) and duodenal endoscopic mucosal resection at our hospital between January 2020 and April 2022. Primary endpoints were fever and increased C-reactive protein (CRP) levels following ER. The secondary endpoints were risk factors for aspiration pneumonia. Blood tests and radiography were performed on the day after ER, and computed tomography was performed if the cause was unknown. Results: Among the 822 patients included, aspiration pneumonia was the most common cause of fever and increased CRP levels after ER of the upper gastrointestinal tract (esophagus, 53%; stomach, 48%; and duodenum, 71%). Post-ER coagulation syndrome was most common after colorectal ESD (38%). On multivariate logistic regression analysis, lesions located in the esophagus (odds ratio [OR] 3.57; P < 0.001) and an amount of irrigation liquid of ≥1 L (OR 3.71; P = 0.003) were independent risk factors for aspiration pneumonia. Conclusions: Aspiration pneumonia was the most common cause of fever after upper gastrointestinal ER and post-ER coagulation syndrome following colorectal ESD. Lesions in the esophagus and an amount of irrigation liquid of ≥1 L were independent risk factors for aspiration pneumonia.

Background: Bleeding and hematoma formation during submucosal injection in esophageal endoscopic submucosal dissection (ESD) reduce the visibility of the submucosa. Red dichromatic imaging (RDI) is an endoscopic technique that provides better visualization of the deep submucosal blood vessels. We speculated that blood vessel injury could be avoided with RDI. This pilot study evaluated the role of RDI in preventing bleeding and hematoma formation during esophageal ESD. Methods: This was a single-center retrospective observational study. We examined 60 patients who underwent ESD with white light imaging (WLI) and RDI. A single endoscopist reviewed all of the surgical videos to document the incidence and severity of bleeding episodes. Eighteen videos provided adequate quality and detail, and the number of blood vessels traversing the mucosal incision lines of the lesions in these videos was evaluated under WLI and RDI. Results: The WLI group had a significantly higher incidence of hematomas per unit area compared to the RDI group (0.18/cm2 [range 0–0.38] vs. 0 [0–0.18]/cm2, p = 0.024). The WLI group also had a significantly higher incidence of total bleeding episodes compared to the RDI group (42.9% [range 21.7–60.4] vs 16.7% [range 13.8–22.9], p < 0.001). Significantly more blood vessels were visible under RDI compared to WLI (5 [range 4–8] vs. 2 [range 1–5], p = 0.0020). Conclusion: RDI reduced the incidence of bleeding and hematoma formation during submucosal injection in esophageal ESD. It was assumed that the improvement of blood vessel visibility by RDI might have contributed to the result.

INTRODUCTION:Underwater endoscopic mucosal resection (UEMR) and cold snare polypectomy (CSP) are novel endoscopic procedures for superficial nonampullary duodenal epithelial tumors (SNADET). However, consensus on how to use both procedures appropriately has not been established. In this study, we evaluated treatment outcomes of both procedures, including resectability.METHODS:In this single-center randomized controlled study conducted between January 2020 and June 2022, patients with SNADET ≤12 mm were randomly allocated to UEMR and CSP groups. The primary end point was sufficient vertical R0 resection (SVR0), which was defined as R0 resection including a sufficient submucosal layer. We compared treatment outcomes including SVR0 rate between groups.RESULTS:The SVR0 rate was significantly higher in the UEMR group than in the CSP group (65.6% vs 41.5%, P = 0.01). By contrast, the R0 resection rate was not significantly different between study groups (70.3% vs 61.5%, P = 0.29). The submucosal layer thickness was significantly greater in the UEMR group than in the CSP group (median 546 [range, 309-833] m vs 69 [0-295] m, P < 0.01). CSP had a shorter total procedure time (median 12 [range, 8-16] min vs 1 [1-3] min, P < 0.01) and fewer total bleeding events (9.4% vs 1.5%, P = 0.06).DISCUSSION:UEMR has superior vertical resectability compared with CSP, but CSP has a shorter procedure time and fewer bleeding events. Although CSP is preferable for most small SNADET, UEMR should be selected for lesions that cannot be definitively diagnosed as mucosal low-grade neoplasias.