Objective: To report the safety and efficacy of trans-arterial embolization (TAE) for upper gastrointestinal bleeding (UGIB) and lower gastrointestinal bleeding (LGIB) due to different etiologies in 91 patients for ten years. Methods: A retrospective analysis of GIB treated between January 2010 and December 2020 was performed. TAE was performed using different embolic agents (coils, particles, glue, gelatin sponge, and EVOH-based agents). Technical success, secondary technical success, clinical success, and complications were evaluated. Results: Technical success was achieved in 74/91 (81.32%) patients. Seventeen patients (18.68%) required re-intervention. Secondary technical success was achieved in all cases (100.0%). Clinical success was achieved in 81/91 patients (89.01%). No major complications were recorded; overall, minor complications occurred in 20/91 patients. Conclusions: TAE is a technically feasible and safe therapeutic option for patients with GIB from a known or blind anatomic source where endoscopic therapy has failed or is deemed unfeasible.

Background and Aims: International guidelines advise improving esophagogastroduodenoscopy (EGD) quality in Western countries, where gastric cancer is still diagnosed in advanced stages. This nationwide study investigated some indicators for the quality of EGD performed in endoscopic centers in Italy. Methods: Clinical, endoscopic, and procedural data of consecutive EGDs performed in one month in the participating centers were reviewed and collected in a specific database. Some quality indicators before and during endoscopic procedures were evaluated. Results: A total of 3,219 EGDs performed by 172 endoscopists in 28 centers were reviewed. Data found that some relevant information (family history for GI cancer, smoking habit, use of proton pump inhibitors) were not collected before endoscopy in 58.5-80.7% of patients. Pre-endoscopic preparation for gastric cleaning was routinely performed in only 2 (7.1%) centers. Regarding the procedure, sedation was not performed in 17.6% of patients, and virtual chromoendoscopy was frequently (>75%) used in only one (3.6%) center. An adequate sampling of the gastric mucosa (i.e., antral and gastric body specimens) was heterogeneously performed, and it was routinely performed only by 23% of endoscopists, and in 14.3% centers. Conclusions: Our analysis showed that the quality of EGD performed in clinical practice in Italy deserves to be urgently improved in different aspects.

Gastric leak is a serious complication of sleeve gastrectomy with a well-documented morbidity and mortality. Depending on the series the leak rate ranges between 1 and 5%. The treatment of sleeve gastrectomy leak is still challenging. Different procedures have been described in management of gastric leak, both surgical and endoscopic. The treatment of gastric leaks depends on the extent of the staple-line leak, the site of the leak and its association with stenosis. As published data are limited, there are no still standardized guidelines on best treatment. One of the most commonly used option in the treatment of gastric leak is esophageal stent. Its success rate reaches 70–80% but it is burdened by some complications. Stent migration is the most common complication in the placement of esophageal stent.We present a challenging surgical in which case the use of an esophageal stent for the treatment of a sleeve gastrectomy leak gained the resolution of the leak but was complicated by bowel obstruction due to migration of the stent.

Objectives: Linked-colour imaging/blue-laser imaging (LCI/BLI) and CAD-EYE artificial intelligence have been developed for sporadic colorectal neoplasia, but their role in ulcerative colitis (UC) surveillance remains uncertain. We evaluated CAD-EYE-assisted detection and optical characterization using conventional and disease-specific classifications in UC. Methods: In this prospective tandem study, patients with UC undergoing surveillance colonoscopy were examined sequentially using white-light imaging (WLI), LCI, and LCI with CAD-EYE. Lesions were characterized using BLI with Kudo, NICE, and Kudo-IBD classifications, followed by CAD-EYE characterization. Detection and characterization performance were assessed using miss rates and diagnostic accuracy. Results: Among 82 patients, 281 lesions, including 22 neoplastic lesions, were identified. The lesion miss rate decreased from 5.7% with WLI to 2.5% with LCI, while no lesions were missed during the CAD-EYE-assisted withdrawal. However, the fixed examination sequence and absence of prospectively recorded withdrawal times precluded isolation of CAD-EYE’s incremental contribution. The neoplasia miss rate was numerically lower with CAD-EYE than with WLI and LCI, although not significantly. Kudo-IBD showed the highest diagnostic performance, with sensitivity, specificity, and positive and negative predictive values of 90.9%, 85.3%, 34.5%, and 99.1%, respectively. Accuracy was significantly higher with KUDO-IBD than with all other methods (all p < 0.001). Conclusions: In this sequential protocol, CAD-EYE-assisted imaging was associated with complete observed lesion detection, whereas Kudo-IBD provided superior optical characterization. These findings support complementary AI-assisted detection and disease-specific optical assessment. Larger studies with independent examination sequences and prospectively recorded withdrawal times, together with external validation of Kudo-IBD, are needed.

Background: Machine learning (ML) is increasingly applied in medicine, underscoring the need for transparent and clinically relevant models. In gastrointestinal oncology, most ML studies rely on raw imaging data, which limits clinical adoption due to poor interpretability and the difficulty of collecting high-quality, large-scale video and image datasets in routine practice. Endoscopic ultrasound (EUS) plays a central role in the evaluation of pancreatic cancer; however, structured EUS features remain underused in predictive modeling. Objective: To assess the performance and interpretability of ML models for diagnosing pancreatic ductal adenocarcinoma (PDAC) using routinely collected EUS variables. Methods: We conducted a retrospective multicenter study using data from two Italian hospitals (n = 641) for model training and internal validation and from a third hospital (n = 120) for external validation, collected from 2015 to 2023. Decision trees, random forests, naïve Bayes and other classifiers were developed and evaluated. Model performance was assessed in terms of discriminative ability, calibration, and selective prediction. Results: All models demonstrated high discriminative performance (AUC ≥ 0.90). Decision trees provided the most favorable balance between interpretability and accuracy (balanced accuracy = 0.87; sensitivity = 0.89). Calibration and selective prediction analyses confirmed the robustness of the models. Conclusions: These findings demonstrate the feasibility of implementing interpretable yet high-performing ML models for PDAC diagnosis in real-life endoscopic settings.