Background and Aim During the Coronavirus Disease 2019 pandemic, esophagogastroduodenoscopy (EGD) has been recognized as an aerosol‐generating procedure. This study aimed to systematically compare the degree of face shield contamination between endoscopists who performed EGD on patients lying in the left lateral decubitus (LL) and prone positions. Methods This is a randomized trial in patients scheduled for EGD between April and June 2020. Eligible 212 patients were randomized with 1:1 allocation. Rapid adenosine triphosphate test was used to determine contamination level using relative light units of greater than 200 as a cutoff value. All eligible patients were randomized to lie in either the LL or prone position during EGD. The primary outcome was the rate of contamination on the endoscopist's face shield. Results The majority of patients were female (63%), with a mean age of 60 ± 13 years. Baseline characteristics were comparable between the two groups. There was no face shield contamination after EGD in either group. The number of coughs in the LL group was higher than the prone group (1.38 ± 1.8 vs 0.89 ± 1.4, P = 0.03). The mean differences in relative light units on the face shield before and after EGD in the LL and prone groups were 9.9 ± 20.9 and 4.1 ± 6 ( P = 0.008), respectively. Conclusion As measured by the adenosine triphosphate test, performing diagnostic EGD does not lead to contamination on the face shield of the endoscopist. However, placing patients in the prone position may further mitigate the risk.

Background/Aims: Previous artificial intelligence (AI) models attempting to segment gastric intestinal metaplasia (GIM) areas have failed to be deployed in real-time endoscopy due to their slow inference speeds. Here, we propose a new GIM segmentation AI model with inference speeds faster than 25 frames per second that maintains a high level of accuracy.Methods: Investigators from Chulalongkorn University obtained 802 histological-proven GIM images for AI model training. Four strategies were proposed to improve the model accuracy. First, transfer learning was employed to the public colon datasets. Second, an image preprocessing technique contrast-limited adaptive histogram equalization was employed to produce clearer GIM areas. Third, data augmentation was applied for a more robust model. Lastly, the bilateral segmentation network model was applied to segment GIM areas in real time. The results were analyzed using different validity values.Results: From the internal test, our AI model achieved an inference speed of 31.53 frames per second. GIM detection showed sensitivity, specificity, positive predictive, negative predictive, accuracy, and mean intersection over union in GIM segmentation values of 93%, 80%, 82%, 92%, 87%, and 57%, respectively.Conclusions: The bilateral segmentation network combined with transfer learning, contrast-limited adaptive histogram equalization, and data augmentation can provide high sensitivity and good accuracy for GIM detection and segmentation.

Background and study aims: Endoscopic transpapillary gallbladder stenting (ETGS) can be a bridging therapy to elective cholecystectomy or a permanent gallbladder drainage in patients with symptomatic gallbladder disease who are unfit for surgery or have cholecystectomy deferral. We evaluated the intermediate to long-term outcomes of ETGS in these groups. Patients and methods: We retrospectively reviewed 234 patients (acute cholecystitis=147), who were unfit for surgery(n=50) or had deferred cholecystectomy(n=184), and all underwent ETGS between 2012-2021. A 7-Fr,15-cm double pigtail plastic stent was placed for ETGS without scheduled stent exchange. Biliary event (cholecystitis and cholangitis)-free rates were determined at the 6-month, 1-year and 2-year follow-up periods. Results: Technical and clinical success rates were 84.6%(198/234) and 97.4%(193/198), respectively. Kaplan-Meier analysis(n=193) showed a biliary event-free rate of 99% [95% confidence interval (CI),0.95-1.00] at 6 months, 92% [95%CI,0.87-0.97] at 1 year and 76% (95%CI,0.65-0.93) after 2 years during the median follow-up period of 564 days(200-3001 days). Conclusions: Endoscopic transpapillary gallbladder stenting is an effective biliary drainage that should be considered in selected cases with common bile duct stone whose cholecystectomy could not be performed or deferred. Biliary event-free rate was 99% at 6 months and remained 92% at 1 year and 76% after 2-year follow-ups.