Endoscopic pyloric exclusion, EOS-guided gastrojejunostomy combined with endoscopic suturing and closure of the pylorus, a novel approach to fail surgical repair of a perforated duodenal ulcer. These are our disclosures. We present an 82-year-old female with a history of metastatic breast cancer who was admitted to an outside facility with an acute abdomen. She was found to have an NSAID-associated duodenal bulb perforation for which she underwent emergent open surgical repair with the use of an omental patch. Postoperatively, the patient developed large-volume enteric output from the JP drains and was found to have a persistent duodenal leak on CT scan. She subsequently underwent an unsuccessful attempt at endoscopic closure before being referred to our center in her third postoperative week. After consultation with our surgical colleagues, considering the patient's many comorbidities including metastatic breast cancer, duration since her last operation, and high risk for surgical reintervention, it was decided to proceed with a repeat attempt at endoscopic closure. The patient was brought to the endoscopy suite and the procedure was performed under general anesthesia in the left lateral position. On endoscopy, the patient was noted to have multiple gastric ulcers, including a large ulcer at the gastric antrum. Upon intubation of the duodenal bulb, a cratered ulcer with friable margins measuring approximately 20 millimeters was visualized. The defect was not amiable to closure using an over-the-scope clip or endoscopic suturing given the size of the defect and its friable margins. Placement of a through-the-scope clip or a fully covered metal stent was also not possible due to the sharp angulation of the duodenum and the presence of a large antral ulcer. Water-soluble contrast was subsequently injected into the duodenal bulb to further delineate the anatomy. Large volume extraluminal spillage was visualized on fluoroscopy. As noted here, the contrast filled a contained cavity that appeared to be drained by the previously placed JP drain. In addition, transient spillage of small amount of contrast into the free peritoneal space was suspected. It was hence decided to divert the gastric stream away from the perforation site using the following two-step approach. Figure 1 depicts normal anatomy, where the green line represents the enteric stream and the yellow line represents the biliopancreatic stream. Figure 2 depicts our patient with a persistent duodenal bulb leak with extraluminal spillage of contrast as indicated by the red arrow. As shown here, EUS-guided gastrojejunostomy is the first step of the procedure where the gastric stream is partially diverted into the proximal jejunum and away from the duodenal bulb. In the second step of the procedure, the duodenum is completely excluded from the anterograde gastric stream by endoscopically suturing the pylorus closed. The following sequences highlight key steps of the EUS-guided gastrojejunostomy procedure. A soft-tip guide wire is carefully advanced into the proximal small bowel under fluoroscopy. A 20-millimeter extraction balloon is then railroaded over the guide wire and inflated in the proximal jejunum. A curvilinear echoendoscope is then advanced into the stomach alongside the balloon catheter and a suitable window for small bowel access is identified. A 15-millimeter by 10-millimeter coterie-enhanced luminoposing metal stent is then advanced freehand into the jejunum. This is deployed using standard technique. A gush of blue dye confirms placement into the jejunum. The lambs is then dilated using a standard dilating balloon. The echoendoscope and all accessories are subsequently withdrawn. After completing the gastroenteric anastomosis, we proceeded with endoscopic suturing and pyloric closure. A double-channeled therapeutic gastroscope mounted with an endoscopic suturing device was then advanced into the stomach and towards the pylorus. The scope is kept in a short position as much as possible to prevent excessive pressure on the recently placed lambs. As seen here, with the scope in a short position, using a proline suture, the pylorus is completely closed in a continuous fashion. While applying constant tension to the suture, water-soluble contrast is injected to fill the stomach. Complete closure of the pylorus is ensured on endoscopy and on fluoroscopy as depicted here. A cinch is subsequently deployed. After completion of the above, the stomach is insufflated with contrast once again to ensure complete diversion of the gastric stream away from the duodenal bulb and into the proximal jejunum via the lambs. As is visualized here, despite the small amount of retrograde flow of contrast towards the biliopancreatic side, the contrast does not reach the duodenal bulb. This schematic represents the final outcome of the procedure. Our patient tolerated the procedure well and was started on an oral liquid diet three days after. As seen here, there was a significant and steady decrease in the JP drain output in the days following the procedure. Our patient was discharged to a rehab facility two weeks after the procedure and died of unrelated causes five months later. In conclusion, we were able to demonstrate that endoscopic pyloric exclusion is a technically feasible and potentially reversible novel procedure. It may provide an alternative to surgical re-exploration in patients that develop duodenal leaks after undergoing primary surgical repair of duodenal perforations. Further studies are needed to evaluate the procedure's efficacy, reproducibility, and clinical use. Thank you for your attention.

perforated duodenal ulcer

elderly female patient

metastatic breast cancer

endoscopic closure

gastric stream diversion