Background and study aims  Self-expandable metallic stents (SEMS) are now widely used even for patients with borderline resectable (BR) pancreatic cancer (PC), as neoadjuvant therapy has become common. Therefore, we conducted this study to evaluate safety of SEMS placement in the population including BR PC and to explore risk factors for recurrent biliary obstruction (RBO), pancreatitis, and cholecystitis. Tamoxifen Patients and methods  We retrospectively investigated consecutive patients with PC who received initial SEMS between January 2015 and March 2019. We compared time to RBO (TRBO), causes of RBO, and stent-related adverse events (AEs) according to resectability status. Univariate and multivariate analyses were performed to explore risk factors for TRBO, pancreatitis, and cholecystitis. Results  A total of 135 patients were included (BR 31 and unresectable [UR] 104). Stent-related AEs occurred in 39 patients pancreatitis 14 (mild/moderate/severe 1/6/7), cholecystitis 12, and non-occluding cholangitis 13. TRBO, causes of RBO, and stent-related AEs were not significantly different according to resectability status. Overall rate of RBO was higher in UR PC due to the longer follow-up period. Sharp common bile duct (CBD) angulation was an independent risk factor for short duration of TRBO. High pancreatic volume index and SEMS of high axial force were independent risk factors for pancreatitis, whereas tumor involvement to orifice of cystic duct was the only risk factor for cholecystitis. Conclusions  We demonstrated that SEMS can be safely deployed even in patients with BR PC. Sharp CBD angulation and high pancreatic volume index were identified as novel risk factors for RBO and pancreatitis, respectively, after SEMS placement.Background and study aims  Use of cold endoscopic resection has increased due to excellent results with it and the ability to avoid electrosurgery related complications. The aim of this study was to evaluate the feasibility and safety of cold-endoscopic submucosal dissection (C-ESD) in an in vivo porcine model. Patients and methods  C-ESD with circumferential incision and submucosal dissection with a predominantly cold technique was tested in the esophagus and colorectum. Incision and dissection were attempted with a cold technique with a biopsy forceps and an endoscopic Maryland dissector. Large vessels were pre-coagulated with the latter device. Different traction methods were applied. Results  Twelve dissections were performed four esophageal, four colonic, and four rectal. Tunnel and pocket methods were applied successfully. Full C-ESD was possible in the colorectum. In the esophagus, an initial incision had to be done with electrocautery. No major bleeding occurred. Two perforations occurred in the colon, one was endoscopically treated. Conclusions  Full C-ESD is feasible in the colorectum, whereas a small hot incision is needed in the esophagus. However, in 50 % of the colonic cases, there were perforations caused by the biopsy forceps making the circumferential incision. Therefore, potential benefits of endoscopic resection without cautery would warrant futures studies in humans initially in esophagus and rectal locations.Background and study aims  Recently, a growing body of evidence has been amassed on evaluation of artificial intelligence (AI) known as deep learning in computer-aided diagnosis of gastrointestinal lesions by means of convolutional neural networks (CNN). We conducted this meta-analysis to study pooled rates of performance for CNN-based AI in diagnosis of gastrointestinal neoplasia from endoscopic images. Methods  Multiple databases were searched (from inception to November 2019) and studies that reported on the performance of AI by means of CNN in the diagnosis of gastrointestinal tumors were selected. A random effects model was used and pooled accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated. Pooled rates were categorized based on the gastrointestinal location of lesion (esophagus, stomach and colorectum). Results  Nineteen studies were included in our final analysis. The pooled accuracy of CNN in esophageal neoplasia was 87.2 % (76-93.6) and NPV was 92.1 % (85.9-95.7); the accuracy in lesions of stomach was 85.8 % (79.8-90.3) and NPV was 92.1 % (85.9-95.7); and in colorectal neoplasia the accuracy was 89.9 % (82-94.7) and NPV was 94.3 % (86.4-97.7). Conclusions  Based on our meta-analysis, CNN-based AI achieved high accuracy in diagnosis of lesions in esophagus, stomach, and colorectum.Background and study aims  The aim of thi systematic review and consensus report is to standardize the practice of endoscopic ultrasound (EUS-guided needle-based confocal laser endomicroscopy (nCLE) for pancreatic cystic lesion (PCL) evaluation. Methods  We performed an international, systematic, evidence-based review of the applications, outcomes, procedural processes, indications, training, and credentialing of EUS-nCLE in management of PCLs. Based on available clinical evidence, preliminary nCLE consensus statements (nCLE-CS) were developed by an international panel of 15 experts in pancreatic diseases. These statements were then voted and edited by using a modified Delphi approach. An a priori threshold of 80 % agreement was used to establish consensus for each statement. Results  Sixteen nCLE-CS were discussed. Thirteen (81 %) nCLE-CS reached consensus addressing indications (non-communication PCL meeting criteria for EUS-FNA or with prior non-diagnostic EUS-FNA), diagnostic outcomes (improved accuracy for mucinous PCLs and serous cystadenomas with substantial interobserver agreement of image patterns), low incidence of adverse events (fluorescein-associated and pancreatitis), procedural processes (nCLE duration, manipulation of needle with probe), and training (physician knowledge and competence). Conclusion  Based on a high level of agreement pertaining to expert consensus statements, this report standardizes the practice of EUS-nCLE. EUS-nCLE should be systematically considered when EUS-FNA is indicated for PCL evaluation.