Endoscopic ultrasound-guided gastroenterostomy (EUS-GE) is a minimally invasive alternative to surgery and endoscopic stenting for malignant and benign gastric outlet obstruction (GOO). Fluoroscopy is traditionally used to insert an oro/naso-enteral catheter to ensure bowel distension and facilitate recognition of the target loop and final stent placement. Cautery-enhanced lumen-apposing metal stents (EC-LAMS) enable a fluorless direct freehand technique, but clinical data on its reproducibility remain limited. We assessed the technical feasibility of fluorless direct freehand EUS-GE and described its safety and clinical outcomes in patients with malignant and benign GOO.

Retrospective consecutive cohort of adults with symptomatic malignant or benign GOO who underwent an attempted fluorless EUS-GE at a tertiary interventional endoscopy centre (January 2020–October 2025). Eligibility was MDT-defined. Exclusions: planned fluoroscopy, hybrid/non-standard approaches (wire-guided/double-balloon), or incomplete data precluding outcomes assessment. All procedures were performed under deep sedation or general anaesthesia. A diagnostic forward-viewing gastroscope was first advanced to the stricture to assess traversability and identify a safe window—either passage beyond the stenosis or a pinpoint orifice suitable for tip apposition. In case of complete or near-complete obstruction, the procedure was performed under fluoroscopic guidance to ensure adequate target distension and mitigate misdeployment risk. When a fluorless approach was feasible, butylbromide 20 mg IV was administered, and the distal jejunal loop was distended by hand-instilling saline–methylene blue (plus a small amount of iodinated contrast to increase fluid density) with 50–60 mL syringes, directly via the gastroscope channel; no infusion pump was used. The gastroscope was then exchanged for a therapeutic linear echoendoscope positioned in the stomach, and an EC-LAMS (Hot-Axios, Boston Scientific, USA; 15×10 or 20×10 mm) was deployed freehand under real-time EUS guidance (a second butylbromide 20 mg dose could be given immediately before deployment). Gastrojejunostomy creation was confirmed by reflux of blue-stained fluid through the LAMS into the stomach.

Primary outcome was feasibility, defined as completion of EUS-GE entirely without fluoroscopy, with technical success achieved. Secondary outcomes were: technical success (creation of a functional gastroenteric anastomosis with correct LAMS deployment, no same-session surgical/percutaneous rescue); clinical success (symptom improvement with a GOOSS increase ≥1 within 72 h); and early safety, reported as intra-procedural adverse events (AEs) and early post-procedural AEs ≤7 days (AGREE-graded). Analyses were descriptive and included all consecutive planned fluoroscopy-free attempts.

Thirty-six consecutive patients underwent an attempted fluoroscopy-free EUS-GE. Median age 70.5 years; ASA II/III/IV 7/25/4; median Charlson Comorbidity Index 7. Etiology was malignant in 30/36 (83.3%) and benign in 6/36 (16.7%). The obstruction more often involved the second duodenal portion (22/36, 61.1%) than the pylorus/bulb (14/36, 38.9%). One patient had a prior unsuccessful duodenal stent. In two patients, the fluorless strategy could not be completed (one conversion to fluoroscopy-assisted EUS-GE without anastomosis completion; one aborted). Feasibility (primary) and technical success were both 94.4% (34/36), with a mean procedural time of 15 minutes. Clinical success was 88.9% (32/36), with rapid recovery—liquids at 24 hours and a median hospital stay of 4 days. Intraprocedural AEs occurred in 4/36 (11.1%), all AGREE IIIA, managed endoscopically (clip closure and, when required, LAMS retrieval/redeployment). Early AEs ≤7 days were 2/36 (5.6%): one minor bleeding (AGREE II) treated conservatively and one fatal misdeployment (AGREE V). Overall, early safety endpoint (intraprocedural or ≤7-day AE) was 16.7 % (6/36).

Fluorless direct freehand EUS-GE showed high feasibility and strong technical and clinical performance in a consecutive cohort. Intraprocedural and early adverse events were not uncommon, yet most were low–to–moderate in severity and readily controlled endoscopically. These findings support the adoption of a radiation-free workflow in experienced centres and motivate prospective studies to validate reproducibility.