Osamu Dohi

Background: Minimally invasive treatments for superficial pharyngeal cancer including transoral videolaryngoscopic surgery (TOVS) and endoscopic submucosal dissection (ESD) have become more widespread as minimally invasive treatments; however, no reports have compared these treatment methods. Methods: We retrospectively investigated consecutive patients with superficial hypopharyngeal squamous cell carcinoma who underwent ESD or TOVS between January 2013 and December 2024. Results: A total of 60 lesions in 55 patients were included in the ESD and TOVS groups. The en bloc resection rate was 100% in both groups. The R0 resection rate in the ESD group showed a nonsignificant trend toward being higher than that in the TOVS group (58.1% vs. 34.5%; p = 0.077). No significant differences were observed in the 3‐year overall survival (OS) rates between the TOVS and ESD groups (93.8% vs. 81.3%; p = 0.330). The 3‐year cumulative incidence of local recurrence was 13.6% in the TOVS group and 5.9% in the ESD group ( p = 0.241). In an exploratory univariate analysis, tumor thickness ≥ 1000 μm was significantly associated with lymph node metastasis ( p = 0.017). Conclusions: Clinical outcomes in patients with superficial hypopharyngeal cancer were not significantly different between ESD and TOVS. Although the cumulative incidence of local recurrence was numerically lower after ESD than after TOVS, the difference was not statistically significant. Both procedures may represent feasible minimally invasive treatment options for selected patients.

Endoscopic closure of mucosal defects after colorectal endoscopic submucosal dissection (ESD) has been increasingly adopted to prevent adverse events. Various devices and techniques, including conventional clips, clips with various technical tips, clip with line, clip with device, special clips, and suturing devices, have been developed, achieving high rates of complete closure (approximately 95% on average), although procedure times and technical complexity vary considerably. Among 14 eligible comparative studies, including randomized controlled trials and retrospective studies published up to April 2026, a meta-analysis was performed to evaluate the efficacy of complete closure after colorectal ESD in preventing delayed bleeding (DB), delayed perforation (DP), and post-ESD coagulation syndrome (PECS). For DB, pooled analysis demonstrated that complete endoscopic closure was significantly associated with a reduced risk of DB (odds ratio [OR]: 0.77, 95% confidence interval [CI]: 0.60–0.97, p = 0.030, I² = 53.6%). In contrast, no significant differences were observed between the closure and non-closure groups for DP (OR: 0.60, 95% CI: 0.23–1.55, p = 0.290, I² = 0.0%) or PECS (OR: 0.94, 95% CI: 0.65–1.38, p = 0.765, I² = 56.6%). In a literature review, reported risk factors for DB include lesion size >50 mm, an ASA score of III or IV, antithrombotic therapy, age ≥75 years, and rectal location. Severe fibrosis and prolonged ESD procedure time have been reported as risk factors for DP, whereas female sex, age ≥70 years, right-sided colon, and lesion size >24 mm have been associated with PECS. Overall, various closure devices and techniques achieve high technical success rates after colorectal ESD. Complete closure appears beneficial for reducing DB, particularly in high-risk patients, whereas its efficacy for preventing DP and PECS remains uncertain. Further studies are warranted.

Background and study aims: Post-endoscopic submucosal dissection (ESD) coagulation syndrome (PECS) is defined as abdominal pain with inflammatory responses, requiring careful differentiation from delayed perforation (DP). This study aimed to evaluate clinical and biochemical characteristics associated with PECS and DP following ESD, and to develop a scoring system for predicting PECS. Patients and methods: We retrospectively analyzed 318 consecutive patients who underwent colorectal ESD at our institution between January 2022 and June 2024. Patients were classified into three groups: PECS (n=32), DP (n=7), and no-PECS/no-DP (NPD) (n=279). Multivariate analysis was performed to identify predictors of PECS for clinical factors and values such as C-reactive protein (CRP), white blood cell (WBC) count, and body temperature. Based on significant predictors, we constructed a five-variable PECS-predictive scoring system (FPSS) for early post-procedure prediction of PECS. Results: Among 318 patients, the incidence of PECS and DP was 10.1% and 2.2%, respectively. DP cases demonstrated greater CRP elevation (p<0.001) and higher leukocytosis (p=0.011) compared with PECS, while PECS cases showed significantly higher WBC (p<0.001) and CRP (p<0.001) than the NPD group. Multivariate analysis identified five independent predictors of PECS: right colon (OR=3.54; p=0.004), female gender (OR=2.10; p=0.018), body temperature ≥37.6°C (OR=16.70; p8,000/µL on day 1 (OR=13.78; p24 mm (OR=3.03; p=0.003). The FPSS (range 0–9) achieved an AUC of 0.881, with high negative predictive value (97.4%). Conclusion: The characteristics of PECS were clarified in comparison with NPD and DP, and the newly developed FPSS demonstrated excellent diagnostic performance of PECS.

Background and Objectives: Bowel bubble formation during colonoscopy can impair mucosal visualization and reduce procedural efficiency. However, its clinical significance remains incompletely characterized. This study aimed to evaluate the prevalence of severe bowel bubbles in the proximal colon and their impact on colonoscopic quality in patients undergoing bowel preparation with oral sulfate solution (OSS), using either a same-day regimen (SAR) with 480 mL OSS or a split-day regimen (SPR) with 960 mL OSS. Methods: This retrospective study was conducted between October 2024 and December 2025 across two affiliated institutions. Patients who underwent colonoscopy with OSS-based bowel preparation were included. SAR was used for screening, symptomatic evaluation, and surveillance colonoscopy. SPR was applied exclusively to patients with a prior history of sodium picosulfate-related abdominal pain or inadequate bowel preparation with SAR. Sodium picosulfate was prescribed on the day before colonoscopy in the SAR group and a low-residual diet was administered in both the SAR and SPR groups. The prevalence of severe bubbles in the proximal colon was evaluated, and their impact on colonoscopic quality was examined. Results: A total of 176 SAR cases and 51 SPR cases were analyzed. The rate of severe bubbles in the proximal colon was identical between both regimens (17.6%). Compared with cases without severe bubbles, those with severe bubbles had significantly longer cecal insertion times (median [IQR]: 7.0 [5.0–10.0] vs. 9.0 [7.0–13.0] min, p = 0.018) and total procedure time (20.0 [16.0–25.0] vs. 24.0 [19.0–30.0] min, p = 0.024). Preparation-to-colonoscopy time was also longer in cases with severe bubbles (5.0 [4.0–5.5] vs. 5.0 [5.0–6.0] h, p = 0.041). Adenoma detection rates were 73.2% in cases without severe bubbles and 67.5% in those with severe bubbles (p = 0.571). Conclusions: Severe bowel bubble formation tended to be associated with longer cecal insertion and total procedure times, and was more frequently observed with a longer preparation-to-colonoscopy interval.

Objectives: Endoscopic submucosal dissection (ESD) is currently the standard treatment for early gastric cancer (EGC) without risk of lymph node metastasis (LNM). However, it is unclear whether the “eCura system,” can be applied to undifferentiated‐type EGC after non‐curative resection. Methods: This nationwide, multicenter, retrospective study included consecutive patients that resulted in non‐curative resection for undifferentiated‐type EGC between January 2011 and March 2019. Results: Overall, 1049 patients were divided into the additional surgery ( n = 716) and the follow‐up without additional surgery ( n = 333) groups. LNM occurred in 6% of the additional surgery group. Tumor size > 30 mm and lymphatic invasion were independent risk factors for LNM upon multivariate logistic regression analyses. The area under the receiver operating characteristic curve (AUC) for the exploratory scoring system (incorporating these risk factors) for LNM prediction was 0.768 (95% confidence interval [CI]: 0.687–0.848) and that for the eCura system was 0.783 (0.703–0.864; p = 0.464). The eCura system was used to stratify the follow‐up without additional surgery group according to risk. Five‐year cancer‐specific survival differed among the low‐, intermediate‐, and high‐risk groups (99.6%, 96.9%, and 60.6%, respectively; p < 0.001). The hazard ratios for cancer recurrence in the intermediate‐ and high‐risk groups were 6.54 (95% CI: 0.41–104.20) and 98.2 (12.10–798.50), respectively ( p < 0.001). Conclusions: Lymphatic invasion and a tumor size > 30 mm are associated with the risk of LNM. Our findings support using the eCura system for risk stratification in undifferentiated‐type EGC after non‐curative resection.

Background: Underwater endoscopic mucosal resection (UEMR) is a relatively new treatment method for sporadic non-ampullary duodenal epithelial tumors (SNADETs), and its usefulness has been reported for SNADETs ≤ 20 mm. However, its effectiveness and safety compared with conventional endoscopic mucosal resection (CEMR) remain controversial. This study aims to assess the treatment outcomes and evaluate the beneficial effects and safety of UEMR for SNADETs ≤ 20 mm. Methods: This is an open-label, multicenter collaborative, non-inferiority randomized controlled trial with two parallel groups conducted across 40 institutions in Japan. The study subjects will be patients with SNADETs ≤ 20 mm. A total of 320 patients will be randomized to either the CEMR or UEMR group in a 1:1 allocation ratio. The primary endpoint is the 1-year recurrence-free survival rate, defined as the number of cases with no recurrence of SNADET or death from any cause within one year of EMR. The secondary endpoints include the en bloc resection rate, histological complete resection rate, adverse events, technical success rate, total procedure time, resection time, mucosal closure time, complete mucosal closure rate, complete mucosal closure rate with standard clips, 1-year duodenum preservation survival rate, and device cost. Discussion: This multicenter, open-label, randomized controlled trial (RCT) with 320 subjects aims to determine whether the 1-year recurrence-free survival rate of underwater endoscopic resection is not inferior to that of conventional endoscopic mucosal resection for SNADETs ≤ 20 mm. If the efficacy and safety of UEMR are proven in this RCT, it is expected to be recognized as the standard treatment for SNADETs ≤ 20 mm. Owing to the absence of submucosal injection, UEMR is a simpler and more cost-effective technique compared to CEMR and is anticipated to become the primary method of EMR.

Introduction: For intermediate-sized (10–20 mm) superficial non-ampullary duodenal epithelial tumors (SNADETs), various endoscopic resection (ER) techniques are available, including conventional endoscopic mucosal resection (EMR), underwater EMR (UEMR), and endoscopic submucosal dissection (ESD). However, the optimal method remains uncertain. Methods: We conducted a systematic review of studies published from January 2013 to August 2023 using PubMed and the Japan Medical Abstracts Society database. Eligible studies reported ER outcomes for SNADETs of ≤20 mm. Data were extracted from 14 cohort studies (3 multicenter lesions 10–20 mm in size and 11 single-center lesions less than 20 mm), including en bloc and R0 resection rates, delayed bleeding, intraoperative and delayed perforations, and recurrence. Results: In the multicenter studies, the pooled en bloc resection rates for EMR, UEMR, and ESD of intermediate-sized SNADETs were 82.7%, 74.8%, and 94.6%, respectively. The corresponding R0 resection rates were 54.2%, 50.6%, and 80.9%. Delayed bleeding rates were similar across methods (3.3% for EMR, 3.2% for UEMR, and 5.4% for ESD). However, intraoperative and delayed perforations were more frequent with ESD (7.9% and 3.0%) than with EMR (1.1% and 0.3%) and UEMR (0.0% and 0.0%). Single-center studies showed consistent trends, with ESD achieving higher resection rates but also showing greater variability in adverse events. Recurrence rates were about 5% for EMR and UEMR, with no recurrences reported after ESD. Conclusion: While ESD provides superior resection quality, EMR and UEMR offer favorable outcomes with fewer adverse events. Given their safety and efficacy profiles, EMR and UEMR should be considered appropriate first-line treatment options for intermediate-sized SNADETs.

INTRODUCTION: The recommendation to discontinue low-dose aspirin (LDA) during the periprocedural period of endoscopic submucosal dissection (ESD) for early gastric cancer (EGC) remains controversial in patients at low thrombotic risk. If continuation of LDA does not increase bleeding risk, gastric ESD could be safely performed under sustained antiplatelet effects, benefiting patients. This study aimed to evaluate whether continuation of LDA is noninferior to discontinuation regarding ESD-related bleeding risk in patients with low thrombotic risk. METHODS: Patients with gastric mucosal neoplasms and low thrombotic risk who were taking LDA were eligible. Participants were randomly assigned to either the discontinuation or continuation group. No prophylaxis was performed except for cauterizing exposed vessels at the ulcer base. The primary end point was the incidence of ESD-related major bleeding. RESULTS: A total of 551 patients were enrolled. In the modified intent-to-treat analysis, the incidence of ESD-related major bleeding (95%) was 7.6% (4.8–11) in the discontinuation group (n = 276) and 10% (6.9–14) in the continuation group (n = 275), yielding a risk difference of 2.5% (−2.3 to 7.4). The upper limit of the 2-sided 90% confidence interval (6.6%) exceeded the predefined noninferiority margin (Δ5%, P = 0.157). The per-protocol analysis supported this negative result, with a risk difference of 4.4% (−0.5 to 9.5). Two patients in the continuation group developed cerebral infarction. DISCUSSION: The noninferiority of LDA continuation over discontinuation for ESD-related major bleeding in gastric ESD was not demonstrated. Therefore, discontinuation of LDA during the periprocedural period is recommended for patients with low thrombotic risk (UMIN000013177).

Endoscopic submucosal dissection (ESD) for colorectal lesions has become widely accepted but remains technically demanding, with perforation risk related to luminal distension and suboptimal visualization. Gel immersion ESD (GI-ESD) may offer mechanistic advantages over conventional CO2-insufflation ESD (C-ESD), including reduced colonic distension, improved visibility/depth perception, buoyancy-assisted dissection stability, and heat-sink protection. We hypothesized that these features would improve short-term safety and procedure stability. In this single-center retrospective study, 426 consecutive colorectal ESD cases (from January 2022 to December 2024) were analyzed. C-ESD was performed from January 2022 to March 2024; GI-ESD using Viscoclear (Otsuka Pharmaceutical Factory, Tokushima, Japan) from April to December 2024. After excluding non-epithelial lesions, 392 patients were included, including interrupted procedures. The primary outcome was adverse events (perforation, delayed bleeding). Secondary outcomes were en bloc and R0 resection rates, procedure time, and postoperative inflammatory response. Of 392 patients, 293 underwent C-ESD and 99 underwent GI-ESD were analyzed. En-bloc and R0 resection rates were comparable between GI-ESD and C-ESD (97.0% vs. 99.9%, P = 0.16, 90.6% vs. 91.4%, P = 0.35). The perforation rate was significantly lower in GI-ESD (0% vs. 4.1%, P = 0.04). Postoperative fever (≥ 37.6°C; 4.2% vs. 12.7%, P = 0.02) and median CRP elevation (0.33 mg/dl vs. 0.42 mg/dl, P < 0.01) were also significantly reduced in GI-ESD; however, incidence of post-endoscopic submucosal dissection electrocoagulation syndrome (PECS) was not significantly different between groups. GI-ESD achieved curative outcomes comparable to C-ESD while reducing perforation risk; PECS remained comparable between groups.