Background and Objectives: Bleeding from gastric varices (GV) is severe and associated with high mortality. The standard endoscopic therapy, EUS-guided coil injection with cyanoacrylate glue (EUS-COIL + glue), is highly effective but carries a risk of systemic embolization. This study investigates the safety and efficacy of PuraStat, a synthetic peptide hemostatic gel that promotes clot formation without embolization risk, as an alternative injectate in conjunction with EUS-COIL. Methods: This was a single-center, retrospective analysis of 9 patients who underwent EUS-COIL + PuraStat for GV treatment (active bleeding, recent bleeding, or primary prophylaxis). Technical success was defined as successful coil deployment and reduction of Doppler flow; clinical success as bleeding cessation and/or absence of bleeding at 30 days. Secondary outcomes included adverse events (AEs) and re-intervention rates during follow-up. Results: Technical and clinical success was achieved in 100% of the procedures. There were no intraprocedural AEs, and only 1 minor (Grade II) postprocedural AE was recorded (10%). No patients experienced systemic embolization. One patient (11%) experienced recurrent bleeding at 54 days. Over a mean follow-up of 289 days, 56% of patients required repeat EUS-COIL procedures. Conclusion: EUS-COIL with PuraStat is a safe and highly effective treatment for GV in this small cohort, demonstrating excellent immediate success without the risk of systemic embolization associated with cyanoacrylate glue. While its use may necessitate earlier and more frequent surveillance and re-intervention, its superior safety profile makes it an attractive option. Larger comparative studies are warranted.
Background and Objectives: Gastric outlet obstruction impedes gastric emptying into the duodenum. Although EUS-guided gastroenterostomy (EUS-GE) provides a minimally invasive therapeutic option, current techniques are limited by significant constraints. This study evaluated the safety and efficacy of direct EUS-GE (DGE) compared with a novel balloon-assisted EUS-GE (NBAGE). Methods: This multicenter retrospective cohort study enrolled patients who underwent EUS-GE in 4 high-volume tertiary Chinese referral centers from June 2023 to December 2025. Forty-six patients with DGE and 55 patients with NBAGE were included. Technical and clinical success rates, procedural time, adverse events, injection volume, postoperative length of stay, and mortality were compared. Results: Among 101 patients, NBAGE had significantly higher technical success as compared with DGE (100% vs. 87.0%, P = 0.007). Clinical success was similar between groups (95% vs. 96.4%, P = 1.0). The mean procedural time was similar between DGE and NBAGE (32 vs. 30.4 min, P = 0.564). Thirty-day adverse events occurred in 5 of 40 patients (12.5%) in the DGE group versus 1 of 55 patients (1.8%) in the NBAGE group ( P = 0.079). NBAGE required significantly less injection volume (101 vs. 647 mL, P < 0.001). Postoperative length of stay did not differ significantly (5 vs. 4 days, P = 0.124). The mortality rate was 0% in both groups. Conclusion: NBAGE was associated with significantly higher technical success with substantially reduced injection volume, providing a safe and effective alternative technique for EUS-GE.
Background and Objectives: Completely endophytic renal tumors, which are invisible on the renal surface, present substantial challenges during laparoscopic partial nephrectomy (LPN). These tumors frequently result in prolonged warm ischemia time, higher rates of positive surgical margins, and excessive loss of renal parenchyma. This preliminary retrospective case series evaluated the feasibility, safety, and efficacy of real-time electrode marking guided by the EG-UC5T EUS system for precise intraoperative projection of tumor contours onto the renal surface. Methods: Five patients with completely endophytic renal tumors (RENAL E score = 3 in all cases; mean RENAL score 9.4 ± 0.9; mean diameter 2.6 ± 0.7 cm) underwent retroperitoneal LPN with EG-UC5T EUS-assisted marking. The flexible echoendoscope was introduced through a laparoscopic port for systematic real-time ultrasound scanning to delineate tumor boundaries, depth, and relationships to adjacent structures. Immediate surface marking was then performed using a metal electrode loop advanced through the integrated 4.0-mm working channel under continuous EUS guidance. Perioperative outcomes, renal functional preservation, and early oncologic results were analyzed descriptively. Results: In this series of 5 patients, tumor localization and surface marking were successfully achieved in all 5 cases (100% success rate), with a mean localization time of 8.6 ± 2.1 min. Mean operative time was 141.6 ± 18.9 min, estimated blood loss 71 ± 47.8 mL, and warm ischemia time 27 ± 8.5 min. No intraoperative complications or conversions to open surgery occurred. Mean renal parenchymal preservation reached 92% ± 4%. Estimated glomerular filtration rate declined by 3.7 ± 1.8 mL/min/1.73 m 2 on postoperative day 1 and recovered to a decline of only 1.2 ± 0.8 mL/min/1.73 m 2 at 3 months. Negative surgical margins were obtained in all cases (mean margin distance 3.2 ± 1.1 mm). One Clavien–Dindo grade I perinephric hematoma occurred and resolved conservatively. At a median follow-up of 8 months (range 6–12 months), no local recurrence or distant metastasis was detected. Conclusions: EG-UC5T EUS-guided real-time electrode marking is a feasible, safe, and effective technique for localizing completely endophytic renal tumors during LPN. It enables precise surface projection, reduces blind resection, and achieves favorable perioperative and early oncologic outcomes. Larger prospective studies are warranted.
Background and Objectives: Percutaneous biopsy is the traditional method for diagnosing splenic lesions but carries a risk of hemorrhage due to the organ’s rich vascularity. EUS-guided biopsy has emerged as a minimally invasive alternative that may reduce bleeding risk. This study compared the safety and diagnostic performance of EUS-guided and percutaneous spleen biopsies. Methods: We retrospectively analyzed 91 patients who underwent splenic biopsy at a tertiary center between 2015 and 2023 (EUS, n = 63; percutaneous, n = 28). The primary outcome was the occurrence of procedure-related adverse events graded by the Adverse Events in Gastrointestinal Endoscopy (AGREE) classification. Secondary outcomes included tissue adequacy and diagnostic performance. Statistical analyses used Fisher’s and Boschloo tests, Firth-penalized logistic regression, and propensity-score overlap weighting to adjust for confounders. Results: Clinically relevant adverse events (AGREE grade ≥2) occurred among 0% of EUS-guided and 10.7% of percutaneous biopsies (absolute risk difference: −10.7%, 95% confidence interval: −21.1% to −0.3%; P = 0.029). After adjustment, EUS-guided biopsy remained independently associated with fewer major adverse events (adjusted odds ratio: 0.09, 95% confidence interval: 0.00 to 0.78; P = 0.047). Diagnostic sensitivity was comparable between EUS and percutaneous biopsy (86.4% vs. 93.3%; P = 0.74), with 100% specificity in both. Conclusion: EUS-guided spleen biopsy was associated with a lower rate of clinically relevant adverse events, maintaining comparable diagnostic accuracy. These findings suggest that EUS-guided biopsy is a safer and equally effective alternative for splenic tissue acquisition, warranting validation in larger prospective studies.
Background and Objectives:Pancreatic neuroendocrine neoplasms (PNENs) are rare tumors with heterogeneous outcomes. Tumor grading (G), based on mitotic count and Ki-67 index, is the main prognostic factor guiding treatment. EUS-guided fine-needle aspiration/biopsy is the current standard but shows a misgrading rate up to 25%. We evaluated an artificial intelligence-based video analysis model to predict PNEN grading from contrast-enhanced EUS (CE-EUS) recordings. Methods:This retrospective study was conducted at Istituti di Ricovero e Cura a Carattere Scientifico San Raffaele Hospital, Milan, a European Neuroendocrine Tumor Society Center of Excellence. Patients were eligible if CE-EUS videos ≥1 minute (arterial and venous phases) and cyto-histological confirmation of PNEN were available. Exclusion criteria included mixed neuroendocrine-non-neuroendocrine neoplasms, missing Ki-67 grading, or poor video quality. CE-EUS videos were processed with a deep-learning video transformer model (GradAINet). The dataset was split into training (70%), validation (10%), and testing (20%) cohorts. Diagnostic performance was evaluated using sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and F1-score. Results:Between 2022 and 2024, 115 patients were included (49 female, 42.6%): 70 had G1, and 45 had G2-G3 tumors. Overall, 253,751 video frames were analyzed. GradAINet achieved a sensitivity of 0.817 (95% confidence interval [CI]: 0.556-1.000), specificity 0.806 (95% CI: 0.588-1.000), positive predictive value 0.759 (95% CI: 0.500-1.000), negative predictive value 0.856 (95% CI: 0.667-1.000), and accuracy 0.811 (95% CI: 0.654-0.962). Conclusion:This artificial intelligence-driven CE-EUS video model shows high accuracy for PNEN grading and potentially complements EUS-guided fine-needle aspiration/biopsy. As the first video-based rather than static-image model, it represents a methodological advance. Multicenter validation on larger cohorts is needed before clinical implementation.
Background and Objectives:Limited prospective data exist comparing long-term outcomes of endoscopic placement of a self-expandable metallic stent (ES) and EUS-guided gastroenterostomy (EUS-GE) for unresectable malignant gastric outlet obstruction (mGOO). This large-scale study aimed to prospectively compare these 2 procedures with long-term follow-up. Methods:This single-center, prospective study enrolled 152 consecutive patients with unresectable mGOO who underwent ES (n = 76) or EUS-GE (n = 76) between August 2021 and January 2024, with follow-up until death or administrative censoring on January 31, 2025. The primary outcome was the reintervention rate, while secondary outcomes included technical and clinical success, changes in the gastric outlet obstruction score (GOOS), adverse events (AEs), and survival. A Cox regression model identified factors associated with reintervention and survival. Results:Both groups had comparable technical and clinical success rates and AEs. After a median follow-up of 121 days (interquartile range, 55-285), the EUS-GE group had a significantly lower reintervention (adjusted hazard ratio, 0.22, P < 0.001), longer stent patency (median, 566 days vs. 195 days; P < 0.001), and greater GOOS improvement (median, 3 vs. 2; P < 0.001) compared with the ES group. Poor performance status, the presence of ascites, and lack of chemotherapy were independent mortality risk factors. Conclusion:Compared with ES, EUS-GE provides longer stent patency, fewer reinterventions, and better relief of GOO symptoms while maintaining a similar safety profile. In an expert setting, EUS-GE should be considered the optimal approach for managing unresectable mGOO, as it offers superior short- and long-term outcomes.
Background and Objectives:Large bile duct stone removal in patients with surgically altered anatomy (SAA) is technically challenging. Although balloon-enteroscopy-assisted endoscopic retrograde cholangiopancreatography (BE-ERCP) has been widely adopted, EUS-guided treatment (EUS-T) may achieve higher stone removal rates owing to improved device accessibility after anastomosis. In this multicenter retrospective study, we aimed to compare the outcomes of EUS-T and BE-ERCP for large bile duct stones in patients with SAA. Methods:Patients with SAA who underwent EUS-T or BE-ERCP for large bile duct stones (≥12 mm) between January 2016 and February 2025 were included. The primary outcome was the complete stone removal rate. Secondary outcomes included the number of sessions, procedure time, and adverse events. Propensity score matching was applied to adjust for baseline differences between groups. Results:In total, 22 EUS-T and 87 BE-ERCP procedures were analyzed. After matching, the complete stone removal rate was significantly higher in the EUS-T group than in the BE-ERCP group (90.5% vs. 61.9%, P = 0.030), despite a greater number of sessions (2.1 vs. 1.2, P < 0.001). Mean procedure times and adverse event rates were comparable. Cumulative success significantly improved up to the third session in EUS-T (first session, 27%; third, 86%). Conclusion:EUS-T achieved a higher complete stone removal rate than BE-ERCP without a significant increase in adverse event rates, and may be a first-line treatment option for large bile duct stones in patients with SAA.
Background and Objectives:Pancreatic steatosis (PS) is increasingly recognized as a clinically significant condition. Although EUS is commonly used to evaluate PS, existing diagnostic criteria lack histologic validation. This study aimed to establish histology-validated EUS criteria for diagnosing PS. Methods:Patients undergoing EUS before pancreatic surgery were prospectively enrolled. Two experienced endosonographers independently assessed predefined EUS features, with interobserver agreement evaluated. Histologic confirmation of PS was obtained from surgical specimens. Features significantly associated with histologic PS were used to develop the pancreatic steatosis endosonographic criteria (PSEC). Results:Of 132 enrolled patients, 96 completed the study. Histologic PS was confirmed in 27.10% of cases. EUS features independently associated with PS included hyperechoic parenchyma (odds ratio [OR]: 51.96), obscured main pancreatic duct margin (OR: 10.88), obscured "salt-and-pepper" appearance (OR: 8.25), and absence of hyperechoic foci or strands (OR: 5.11). Based on these findings, the PSEC were established, consisting of 1 major criterion (hyperechoic parenchyma) and 3 minor criteria (obscured salt-and-pepper appearance, obscured main pancreatic duct margin, and absence of hyperechoic foci or strands). The optimal diagnostic threshold was defined as either 1 major plus 1 minor criterion or 3 minor criteria. This model achieved an area under the receiver operating characteristic curve of 0.84 (95% confidence interval [CI]: 0.8-0.9), sensitivity of 84.60%, specificity of 87.10%, positive predictive value of 71%, and negative predictive value of 93.80%, with substantial interobserver agreement. Conclusion:PSEC represents the first histology-validated EUS criteria for PS, demonstrating robust diagnostic performance and practical applicability in clinical practice.
Malignant gastric outlet obstruction is a frequent complication of advanced pancreatic and gastric antral malignancies requiring palliative intervention. EUS-guided gastroenterostomy (EUS-GE) has emerged as an alternative to enteral stent placement (ESP), but comparative evidence remains limited. We conducted a pairwise and reconstructed individual patient data meta-analysis of randomized controlled trials and cohort studies comparing EUS-GE with duodenal stent placement in adults with malignant gastric outlet obstruction. Systematic searches of PubMed, Web of Science, Scopus, and the Cochrane Library were performed from inception through November 15, 2025. Stent-related failure was the primary outcome and was analyzed as a time-to-event endpoint using reconstructed individual patient data from published Kaplan-Meier curves. Secondary outcomes included technical success, clinical success, overall survival, length of hospital stay, and adverse events. Random-effects models were applied using hazard ratios and risk ratios (RRs) with 95% confidence intervals (CIs). Ten studies comprising 1359 patients were included. EUS-GE was associated with a significantly lower rate of stent-related failure compared with duodenal stenting (hazard ratio, 0.57; 95% CI, 0.45-0.73; P < 0.001). Technical success was slightly lower with EUS-GE (RR, 0.98; 95% CI, 0.97-1.00; P = 0.028), whereas clinical success was significantly higher than ESP (RR, 1.13; 95% CI, 1.08-1.19; P < 0.001). Overall survival, length of hospital stay, and rates of adverse events were comparable between groups. EUS-GE was associated with lower stent-related failure rates and higher clinical success compared with enteral stenting, with similar survival and safety outcomes. Despite slightly lower technical success rates, EUS-GE remains a durable palliative option in experienced centers, while ESP remains appropriate for short-term palliation.
Background and Objectives:Pancreatic ductal adenocarcinoma (PDAC) is a lethal malignancy with limited treatment options. Comprehensive molecular profiling with next-generation sequencing (NGS) may enable personalized therapies, but its feasibility using EUS-guided fine-needle biopsy (EUS-guided FNB) and broad panels in routine practice remains unclear. This study aimed to assess the feasibility of NGS using residual diagnostic tissue from EUS-guided FNB in PDAC, without dedicated sampling. Methods:We performed a retrospective single-center study of patients with PDAC who underwent EUS-guided FNB, analyzing residual paraffin-embedded tissue blocks for NGS using a 63-gene panel. Samples required a minimum tumor cellularity of 20% and tumor area ≥10 mm2. Baseline, procedural, and genomic data were compared between NGS-feasible and NGS-unfeasible groups. The primary endpoint was technical success. Results:Thirty-five patients were included. NGS was successful in 74.3% of cases (26/35) using residual diagnostic material without additional biopsy passes. No significant baseline or procedural differences were observed between groups, except for metastatic disease, which was more frequent in the NGS-unfeasible group (55.6% vs. 11.5%, P = 0.029). KRAS mutations were identified in 73.1% of sequenced cases, predominantly at codon 12. Conclusion:EUS-guided FNB samples can support NGS from residual diagnostic tissue with high feasibility, avoiding dedicated sampling. This pragmatic approach may streamline molecular profiling and expand access to precision oncology in PDAC.
While direct endoscopic cyanoacrylate injection (DEI-CYA) is the standard hemostatic therapy, it is associated with risks of ectopic embolization and ulceration. EUS-guided injection (EUS-CYA) allows for precise targeting of feeder vessels, potentially improving outcomes. A comprehensive search was conducted across PubMed, Web of Science, CENTRAL, and Scopus for RCTs up to January 2026. Primary outcomes included early and late rebleeding rates and variceal eradication. Risk ratios (RRs) and standardized mean differences (SMD) were pooled using a random-effects model. Five RCTs involving 440 patients were included. EUS-CYA was associated with a significant reduction in late rebleeding compared with DEI-CYA (RR: 0.40; 95% confidence intervals [CI]: 0.27-0.60; P < 0.001). However, there was no significant difference in early rebleeding (P = 0.53) or overall variceal eradication rates (RR: 1.13; 95% CI: 0.97-1.31; P = 0.11). Regarding safety, EUS-CYA significantly reduced the risk of postinjection ulcers (RR: 0.32; 95% CI: 0.11-0.93; P = 0.04), while rates of ectopic embolism (P = 0.75) and mortality (P = 0.65) were comparable. Procedurally, EUS-CYA was more efficient, requiring significantly less cyanoacrylate (SMD: -0.35; 95% CI: -0.61 to -0.09; P = 0.01) and fewer sessions for eradication (SMD: -1.28; 95% CI: -2.50 to -0.05; P = 0.04). Current evidence suggests that EUS-CYA likely reduces the risk of late rebleeding compared with DEI-CYA. While the data indicate potential benefits regarding procedural efficiency (reduced cyanoacrylate volume and fewer sessions) and a lower incidence of postinjection ulcers, the certainty of evidence for these outcomes remains very low.
