No abstract

Background Disconnected pancreatic duct (DPD) after development of walled-off necrosis (WON) predisposes to recurrent (peri)pancreatic fluid collection (PFC). In this randomized controlled trial, we compared plastic stents with no plastic stent after removal of a large-caliber metal stent (LCMS) on incidence of recurrent PFCs in DPD. Methods Consecutive patients with WON who underwent endoscopic ultrasound (EUS)-guided drainage with LCMS between September 2017 and March 2020 were screened for eligibility. At LCMS removal (4 weeks after drainage), patients with DPD were randomized to plastic stent or no stent groups. The primary outcome was incidence of recurrent PFC at 3 months. Secondary outcomes were technical success of plastic stent deployment, adverse events, stent migration, and recurrence of PFC at 6 and 12 months. Results 236 patients with WON underwent EUS-guided drainage using LCMS, and 104 (males 94, median age 34 years (interquartile range [IQR] 26–44.7) with DPD were randomized into stenting (n = 52) and no-stenting (n = 52) groups. Plastic stent deployment was successful in 88.5 %. Migration occurred in 19.2 % at median follow-up of 8 months (IQR 2.5–12). Recurrent PFCs occurred in six patients at 3 months (stent n = 3, no stent n = 3). There was no significant difference in PFC recurrence between the two groups at 3, 6, and 12 months. Reintervention was required in seven patients with recurrent PFCs, with no significant difference between the two groups. Conclusion In patients with WON and DPD, deployment of plastic stents after LCMS removal did not reduce recurrence of PFC.

No abstract

Background & Aims Lack of clinical validation and inter-observer variability are two limitations of endoscopic assessment and scoring of disease severity in patients with Ulcerative Colitis. We developed a deep learning (DL) model to improve, accelerate and automate UC detection, and predict the Mayo Endoscopic Subscore (MES) and the Ulcerative Colitis Endoscopic Index of Severity (UCEIS). Methods A total of 134 prospective videos (1,550,030 frames) were collected and those with poor quality were excluded. The frames were labeled by experts based on MES and UCEIS scores. The scored frames were used to create a preprocessing pipeline and train multiple convolutional neural networks (CNNs) with proprietary algorithms in order to filter, detect and assess all frames. These frames served as the input for the DL model, with the output being continuous scores for MES and UCEIS (and its components). A graphical user interface was developed to support both labeling video sections and displaying the predicted disease severity assessment by the AI from endoscopic recordings. Results Mean absolute error (MAE) and mean bias were used to evaluate the distance of continuous model's predictions from ground truth and its possible tendency to over/under-predict were excellent for MES and UCEIS. The quadratic weighted kappa used to compare the inter-rater agreement between experts’ labels and the model's predictions showed strong agreement (0.87, 0.88 frame-level, 0.88, 0.90 section-level and 0.90, 0.78 at video-level, for MES and UCEIS, respectively). Conclusions We present the first fully automated tool that improves the accuracy of the MES and UCEIS, reduces the time between video collection and review, and improves subsequent quality assurance and scoring.

Think you can outsmart VideoGIE? Try your luck with our new Quiz series. Using cases published in VideoGIE, we post questions designed to improve the education of our fellows. Go to http://www.videogie.org to submit your answer, and keep checking back for more quizzes!