Background/Objectives: Contrast-enhanced harmonic imaging endoscopic ultrasound (CHI-EUS) is a valuable tool for characterizing solid pancreatic tumors. However, interobserver variability remains a significant limitation in clinical interpretation. Artificial intelligence (AI) may offer objective, reproducible assessments, potentially enhancing diagnostic performance. This study compared the diagnostic accuracy and interobserver agreement of nine physicians with varying CHI-EUS experience levels vs. a dedicated AI system and a general-purpose large language model (ChatGPT) on the same 118 histologically confirmed cases. Methods: We conducted a prospective, multicenter, observer-blinded study involving 118 CHI-EUS video cases of histologically confirmed (EUS-FNB) focal pancreatic masses from three tertiary care centers in Romania. Nine readers were stratified into three groups: trainees (<5 years CHI-EUS experience), intermediates (5-10 years), and experts (>10 years). All readers and two AI models received standardized, anonymized 2 min CHI-EUS video clips. A dedicated AI system used a convolutional neural network (CNN) for lesion segmentation and time-intensity curve (TIC) extraction, followed by a feedforward neural network (FNN) for classification. ChatGPT was separately evaluated on the same videos. Diagnostic metrics (accuracy, sensitivity, specificity, positive predictive value [PPV], negative predictive value [NPV], and AUROC) were calculated. Interobserver agreement was assessed using Fleiss' and Cohen's kappa statistics. Results: The dedicated AI system achieved an overall accuracy of 95.8% (sensitivity 96.6%; specificity 94.1%) in diagnosing pancreatic adenocarcinoma. Expert readers had a mean accuracy of 78.8% (sensitivity 86%, specificity 61%, and AUROC 0.74), intermediates 80.8% (sensitivity 83%, specificity 75%, and AUROC 0.84), and trainees had a mean accuracy of 67.2% (sensitivity 70%, specificity 60%, and AUROC 0.67). For the most-likely-diagnosis parameter, interobserver agreement was similar between intermediates (Fleiss' κ = 0.407) and experts (κ = 0.389), while trainees showed lower agreement (κ = 0.203). ChatGPT correctly classified only 14.1% of PDAC cases. Conclusions: A specialized AI model for CHI-EUS video analysis can achieve expert-level performance and reduce diagnostic variability across experience levels. Integration of dedicated AI systems into CHI-EUS interpretation may enhance accuracy and serve as a valuable decision support tool in clinical and training settings.

Background:Endoscopic ultrasound (EUS) has emerged as a valuable tool for assessing portal hypertension (PH). Our study aimed to compare endoscopic ultrasound-guided portal pressure gradient (EUS-PPG) with directly measured transjugular PPG (T-PPG) and hepatic venous pressure gradient (HVPG) during transjugular intrahepatic portosystemic shunt (TIPS). Methods:Consecutive patients scheduled for elective TIPS between March 2023 and December 2024 were included. EUS was performed using a 22-gauge FNA needle attached to an invasive pressure monitoring module. TIPS placement and HVPG were performed according to the standard methods. Results:Twenty-four patients who underwent elective TIPS for PH-related complications were enrolled (mean age 56.4 ± 8.8 years, model of end-stage liver disease [MELD] 15.0 ± 5.5). The technical success rates for EUS-PPG and HVPG were 92.3% and 100%, respectively. There was a strong correlation between EUS-PPG and T-PPG (r = 0.88, P < 0.01; intra-class correlation coefficient [ICC] = 0.930, 95% CI 0.837-0.970) and a moderate correlation between EUS-PPG and HVPG (r = 0.58, P < 0.05; ICC = 0.735, 95% CI 0.311-0.898). The presence of portal vein thrombosis (PVT) was independently associated with a ≥ 3 mmHg difference between EUS-PPG and T-PPG (P = 0.04). Child-Pugh score was independently associated with a ≥ 3 mmHg difference between EUS-PPG and HVPG (P = 0.01). Two minor adverse effects were reported. Conclusion:EUS-PPG demonstrates strong reliability when compared to direct PPG measurements obtained during TIPS insertion. Using a 22-gauge needle, EUS-PPG proved to be accurate and safe for assessing PH. The presence of PVT may be a cofounder of discrepancy between the direct methods.

Contrast-enhanced endoscopic ultrasound (CH-EUS) is superior to standard EUS for staging biliary duct tumors (BDTs), but its role in guiding EUS-guided fine needle aspiration (EUS-FNA) remains unclear. We compared diagnostic accuracy of CH-EUS-fine needle aspiration (CH-EUS-FNA) and standard EUS-FNA in patients with suspected malignant biliary stenosis. A parallel randomized controlled trial was conducted in a tertiary medical center and included jaundiced patients with suspected malignant biliary stenosis on computed tomography. The patients were assigned randomly to EUS-FNA or CH-EUS-FNA groups. Final diagnosis was determined based on EUS-FNA, surgical specimen results, endoscopic retrograde cholangiopancreatography (ERCP), or 12-month follow-up. Sixty-one patients were included in the study, 31 in the EUS-FNA group and 30 in the CH-EUS-FNA group. Mean age of participants was 74 ± 11.04 years and mean tumor size was 20.39 ± 9.17 mm, with 43 tumors in the distal bile duct. Final diagnoses were cholangiocarcinoma (37 cases), pancreatic ductal carcinoma (12 cases), other malignancies (3 cases), and benign lesion (9 cases). Diagnostic sensitivity, specificity, and accuracy were 83.3%, 100%, and 87.1% for EUS-FNA, and 82.1%, 100%, and 83.3% for CH-EUS-FNA. Plastic biliary stent placement and tumor location did not influence results. Hyperenhancement in the CH-EUS with rapid washout was observed in 90.9% of cholangiocarcinoma cases. Standard EUS-FNA and CH-EUS-FNA demonstrated comparable diagnostic accuracy in evaluation of extrahepatic bile duct tumors, but with better slightly efficiency and inaccuracy indices than standard EUS-FNA.

Aims Endoscopic ultrasound (EUS) has emerged as a new tool in assessing portal hypertension (PH). EUS portal pressure gradient (EUS-PPG) was described in few studies using a compact manometer, but the use of a pressure transducer connected to a pressure measurement monitor was previously described in a very small number of patients. The comparative assessment of portal pressure gradient measured by EUS and tranjugular route has not been done before. Our study aimed to compare the values of EUS-PPG and transjugular measurements (hepatic vein pressure gradient – HVPG, PPG) in patients with portal hypertension.

Contrast-enhanced endoscopic ultrasound (CH-EUS) can overcome the limitations of endoscopic ultrasound-guided acquisition by identifying microvessels inside inhomogeneous tumours and improving the characterization of these tumours. Despite the initial enthusiasm that oriented needle sampling under CH-EUS guidance could provide better diagnostic yield in pancreatic solid lesions, further studies did not confirm the supplementary values in cases of tissue acquisition guided by CH-EUS. This review details the knowledge based on the available data on contrast-guided procedures. The indications for CH-EUS tissue acquisition include isoechoic EUS lesions with poor visible delineation where CH-EUS can differentiate the lesion vascularisation from the surrounding parenchyma and also the mural nodules within biliopancreatic cystic lesions, which occur in select cases. Additionally, the roles of CH-EUS-guided therapy in patients whose pancreatic fluid collections or bile ducts that have an echogenic content have indications for drainage, and patients who have nonvisualized vessels that need to be highlighted via Doppler EUS are presented. Another indication is represented if there is a need for an immediate assessment of the post-radiofrequency ablation of pancreatic neuroendocrine tumours, in which case CH-EUS can be used to reveal the incomplete tumour destruction.

Aims Pancreatic walled-off necrosis (WON) is a severe complication of acute pancreatitis that often needs endoscopic intervention for drainage and necrosectomy. The selection of stent type for endoscopic drainage remains a topic of debate, with plastic pigtail stents (PPS) and lumen-apposing metal stents (LAMS) emerging as commonly used options. However, strict criteria for stent selection are not available yet. The aim of our study was to evaluate the utility of the QNI classification in guiding the selection of stent types for WON drainage.

Portal hypertension-related complications increase mortality in patients, irrespective of its etiology. Classically, endoscopic ultrasound (EUS) was used to assess the portal venous system and collaterals, considering size and hemodynamic parameters, which correlate with portal hypertension (PH) and related complications. Furthermore, therapeutic EUS guides treatment interventions, such as embolization of the gastric varices through coil placement and tissue adhesive injection, yielding encouraging clinical results. Recently, the direct measurement of portal pressure, emerging as an alternative to hepatic venous pressure gradient, has shown promise, and further research in this area is anticipated. In this review, we aimed to provide a detailed description of various possibilities for diagnosing vascular anatomy and hemodynamics in PH and actual knowledge on the EUS usefulness for PH vessel-related complications. Also, future promises for this field of endo-hepatology are discussed.

Pancreatic fluid collection often occurs as a local complication of acute pancreatitis, and drainage is indicated in symptomatic patients. The drainage may be surgical, percutaneous, or endoscopic ultrasound (EUS) guided. In symptomatic collections older than 4 weeks and localized in the upper abdomen, EUS-guided drainage is the first choice of treatment. Lumen-apposing metal stents are useful in cases of walled-off necrosis, facilitating access to the cavity; however, they do not reduce the number of necrosectomy sessions required. In most pancreatic pseudocysts requiring drainage, plastic stents remain the first choice of treatment. This review aimed to summarize the principles and techniques of step-up therapy of pancreatic fluid collections, including preprocedural and postprocedural assessment and practical approaches of drainage and necrosectomy, making available evidence more accessible to endoscopists aiming to train for this procedure. Successful and safe EUS drainage connotes early recognition and treatment of complications and the presence of a multidisciplinary team for optimal patient management. However, the best time for necrosectomy, modality of drainage method (lumen-apposing metal stents or plastic stents), and duration of antibiotherapy are still under evaluation.

Aims Contrast enhanced endoscopic ultrasound (CH-EUS) is superior to standard endoscopic ultrasound (EUS) for T staging of biliary duct tumors (BDT) but its role in guiding EUS-fine needle aspiration (EUS-FNA) is unknown. We compared the diagnostic performance of CH-EUS-fine needle aspiration (CH-EUS-FNA) and standard EUS-FNA in BTD and aimed to determine the factors influencing the results.

Background: Transcatheter aortic valve replacement (TAVR) became the leading therapeutic strategy for aortic valve replacement in older patients with severe symptomatic aortic stenosis. Echocardiographic parameters that mark the left ventricle and right ventricle reverse remodeling after the TAVR are not well established. The aim of the current study is to describe the dynamics of both left ventricle (LV) and right ventricle (RV) strain derived from speckle tracking echocardiography in elderly patients at 3-months after the TAVR procedure.Methods: We enrolled 52 consecutive patients (77 +/- 4.9 years old, median STS score of 3.1) who underwent transfemoral TAVR at our tertiary care center. All patients were evaluated at baseline and 3 months following TAVR.Results: The LV global longitudinal strain (GLS) 3-month following TAVR was significantly improved compared with baseline values (-16 +/- 4.2% vs -16 +/- 4.2%; p < 0.001) but no significant changes in the RV GLS 3 and 6 segments model following TAVR were registered. The LV ejection fraction was significantly improved 3-months after the TAVR procedure. LV-GLS at baseline demonstrated a strong positive correlation with LV-GLS at 3 months (r = 0.69) and a moderate correlation with RV strain parameters (r = 0.38 and r = 0.56), but also a negative correlation with LVEF at follow-up (r=-0.61). Interestingly, in contrast to LVEF, none of the strain parameters correlated with age. NT-proBNP values were correlated with both LV-GLS (r = 0.37) and LVEF (r=-0.5) at baseline. However, at follow-up, baseline NTproBNP values remained correlated only to LV-GLS at 3-months (r = 0.24), but the correlation was weak.

Carcinoid heart disease is a rare presentation of the carcinoid syndrome, which is caused by excessive tumoral hormone production and the abundant release of vasoactive substances with systemic expressions. A 62-year-old woman presented with flushing, diarrhea, weight loss, and right-sided heart failure symptoms. Specific carcinoid heart disease features were identified using transthoracic and transesophageal echocardiography at the tricuspid and pulmonic valves. Biomarkers, 99mTc-Tektrotyd scintigraphy, SPECT-CT, and a biopsy later confirmed the diagnosis, and the patient began treatment for the underlying condition.

Aim: Percutaneous radiofrequency (RFA) and microwave ablation (MWA) are currently the best treatment options forpatients with liver metastases (LM) who cannot undergo a liver resection procedure. Presently, few studies have evaluated theefficacy of tumor ablation in beginner’s hands but none at all in hepatic metastasis. Our aim was to report the initial experiencewith ultrasound as a tool to guide tumor ablation in a low volume center with no experience in tumor ablation.Material and methods: We conducted a retrospective cohort study, on a series of 61 patients who had undergone percutaneous US-guided ablations for 82 LM between 2010 and 2015. Long term outcome predictors were assessed using univariate and multivariate analysis.Results: Complete ablation was achieved in 86.9% of cases (53/61). All MWA sessions (20/20) attained ablation margins >5mm, compared to 79% (49/62) for RFA sessions (p=0.031). Ablation time was significantly shorter for MWA, with a median duration of 10 minutes (range: 6-12) vs. 14 minutes (range: 10-19.5, p=0.003). There was no statistically significant difference in local tumor progression (LTP)-free survival rates between MWA and RFA (p=0.154). On univariate analysis, significant predictors for local recurrence were multiple metastases (p=0.013) and ablation margins <5 mm (p<.001), both retaining significance on multivariate analysis. Significant predictors for distant recurrence on both univariate and multivariate analysis were multiple metastases (p<0.001) and non-colorectal cancer metastases (p<0.05).Conclusion: A larger than 5 mm ablation size is critical for local tumor control. We favor the use of MWA due to its ability to achieve ablation in significantlyshorter times with less incomplete ablations.

We aimed to compare contrast-enhanced-guided liver biopsy (CEUSLB) and ultrasound-guided liver biopsy (USLB) in the diagnosis of focal liver lesions (FLLs) developed on a background of advanced chronic liver disease (ACLD). Between 2011 and 2019, patients diagnosed with liver tumors on a background of ACLD were evaluated for inclusion in the study. Patients were randomly assigned to the CEUSLB or USLB group. In total, 144 patients were randomly assigned to either CEUSLB (n = 79) or USLB (n = 65). Overall, in the CEUSLB group, the sensitivity was significantly better (94.74% vs. 74.6%, respectively; p = 0.001). Both the fragment length of the biopsy specimen and the single puncture success rate were statistically higher in the CEUSLB group (p = 0.022 and p = 0.0006, respectively). There was no difference in terms of major or minor complications (p = 0.682). CEUSLB is a feasible technique that increases the diagnostic sensitivity for liver tumors developed in ACLD. (C) 2019 World Federation for Ultrasound in Medicine & Biology. All rights reserved.

BACKGROUND & AIMS Gallbladder carcinoma is a rare yet very aggressive cancer. In this study we evaluate the presentation, staging, procedures, complications and survival of patients with gallbladder carcinoma. MATERIAL AND METHODS Data at presentation, operative findings, postoperative evolution, complications and survival data were analyzed for 37 patients with gallbladder carcinoma (as cohort study) confirmed at histopathology between January 2005 and December 2011 in Surgical Department of Regional Institute of Gastroenterology And Hepatology "Octavian Fodor" Cluj-Napoca, Romania. RESULTS In 12 cases we had the suspicion of GBC (gallbladder carcinoma) before surgery, in 6 cases GBC was suspected intraoperatory and in 19 cases only after the histopathology exam. Radical cholecystectomy was considered in 9 cases (24.32%): 4 cases with cholecystectomy alone (patients with Tis-T1) and in 5 cases liver resection was associated. CONCLUSION The GBC has a low incidence (0.35% out of all cholecystectomies), the females being more affected (F:B=4.3:1). GBC was associated with low resecability rate (24.32%) and having a bad prognosis (survival under a year in stages T3 and T4). In most cases the diagnosis was hidden by an acute inflammatory process (acute cholecystitis) and the diagnosis was made after surgical intervention, therefore, the histopathology is crucial in these situations. KEY WORDS Gallbladder carcinoma, Jaundice, Palliative treatment, Resection, Survival.

BACKGROUND AND AIM:Novel biological therapies in Crohn's disease (CD) or Ulcerative colitis (UC) require a proper follow-up for the assessment of bowel inflammation. While endoscopy is the standard method, the imaging techniques using contrast, particularly contrast enhanced ultrasonography (CEUS), are better tolerated by the patients and can be used more frequently. Our aim was to find the usefulness of dynamic CEUS quantification as compared to endoscopy in the assessment of disease activity and in the follow-up under therapy of the patients suffering from either CD or UC.METHOD:We have prospectively evaluated 67 patients with UC and 46 with CD, diagnosed by ileo-colonoscopy and biopsy, comparing the endoscopic scores with clinical scores, C reactive protein (CRP), intestinal wall thickness, layer scores after CEUS and TIC parameters (using SonoLiver® software - Imax, RT, TTP, mTT and AUC). For 25 patients with UC and 13 with CD we performed comparisons of the parameters before and after 3 months of treatment and correlated them with the changes in the endoscopic scores.RESULTS:For UC, time-intensity curves (TIC) volume parameters (AUC) correlated better with endoscopy (ρ=0.64) than the clinical score (ρ =0.62). Other parameters such as CRP and thickness showed significant but less strong correlation, while TIC flow parameters (RT, TTP and mTT) did not show a significant correlation. Results were similar for CD (ρ=0.64 for Imax vs ρ=0.58 for CDAI). The best predictor for endoscopic improvement in both UC and CD was ln(AUC), with a Wilcoxon Z score of 3.76 and 2.61, respectively. There was also a good correlation between the difference of its values and the difference in endoscopic scores before and after the treatment (rho is 0.68 in UC and 0.73 in CD).