Surgical gastrojejunostomy is the conventional treatment for palliating patients suffering from of inoperable malignant gastric outlet obstruction (GOO). Although, the procedure is associated with a high success and low re-intervention rates, there is a risk of morbidities (10% - 16%) and mortalities (7%). The placement of a pyloro-duodenal self-expandable metallic stent (SEMS) is an alternative to surgery for the palliating these patients. As compared to surgery, the procedure is associated with shorter hospital stay, reduced morbidities and cost. However, the clinical efficacy of SEMS is limited by several problems. In uncovered SEMS, the long-term patency is reduced by tumor ingrowth leading to subsequent re-stenosis. In covered SEMS, stent migration occurs at a frequency of 14-25% and it is a major obstacle to stent patency. As a result, randomized trials comparing uncovered stents and partially or fully covered stents in patients with malignant GOO have reported comparable stent patency between the two types of stents.

Recently, the creation of a gastrojejunostomy under EUS (EUS-GJ) guidance using lumen-apposing stents has been described. The procedure was associated with a technical success rate of around 90% and clinical success of 85% to 100%. The procedure holds the potential to create a gastrojejunostomy without surgery. Furthermore, there is a low risk of tumor ingrowth and stent migration, thus improving the stent patency and reducing the need of re-intervention. We have previously published a novel method of creating EUS-GJ with the use of a double balloon occluder (EPASS). The device provides a stable condition for performance of EUS-GJ and improves the safety of the procedure. However, there is limited data on how EPASS compares to endoscopic stenting. The aim of the current study is thus to compare EPASS and DS under a randomized setting.

In recent years, SEMSs have provided an attractive alternative to surgery for palliation of malignant gastric outlet obstruction. Several studies have assessed the clinical and technical success rates of pyloro-duodenal SEMS for malignant gastroduodenal obstruction. Technical success rates of 90% to 100% and clinical success of 80% to 95% was achieved.The procedure was associated with quicker recovery and reduced morbidities as compared to surgical gastrojeunostomies. However, the long-term patency of uncovered SEMS is limited by the risk of tumor ingrowth that would lead to subsequent re-stenosis of the stents requiring re-intervention. Thus to palliate malignant gastric outlet obstruction, surgical gastrojejunostomy is preferred in patients that are at good risk for surgery with prolonged life expectancy whilst insertion of pyloro-duodenal SEMS is preferred in patients that are associated with high-risk for surgery and short life expectancy.

In lieu of the above reasons, it would be the ideal situation if a gastrojejunostomy could be created endoscopically without surgery. The procedure could potentially avoid the risk of morbidities and mortalities associated with surgery whilst enjoying the superior patency associated with a gastrojejunostomy. However, a number of hurdles need to be overcome for safe creation of an anastomosis endoscopically. Firstly, the collapsed bowel (duodenum or jejunum) needs to be easily identified from the stomach. Secondly, a reliable method of creating an anastomosis needs to be available. Lastly, a device for maintaining the anastomosis between two non-adherent organs is required.

The use of EUS as a platform to create a gastrojejunostomy endoscopically may theoretically tackle all of the above hurdles. The device could be used in the stomach to visualize the adjacent duodenum or jejunum for anastomosis. A 19-gauge needle could be used to puncture the adjacent bowel for passage of a guide-wire to connect the two non-adherent organs. Thereafter, a SEMS could be placed between the two organs for creation and maintenance of the anastomosis.

Initial results of EUS-guided gastrojejunostomies from two studies have become available. Khashab et al reported EUS-GJ in 10 patients. Technical success was achieved in 9 patients (90%). Clinical success with resumption of solid oral intake was achieved in all 9 patients (100%), who underwent successful EUS-GE. 8 patients were able to tolerate almost a normal diet and/or full diet, and 1 patient tolerated a soft diet. There were no procedure-related adverse events. Mean procedure time was 96 minutes (range 45-152 minutes), and mean length of hospital stay was 2.2 days. In another multicenter study, 26 patients received EUS-GJ. Technical success was achieved in 24 patients (92 %). Clinical success was achieved in 22 patients (85 %). Of the 4 patients in whom clinical success was not achieved, 2 had persistent nausea and vomiting despite a patent EUS-GJ and required enteral feeding for nutrition, 1 died before the initiation of an oral diet, and 1 underwent surgery for suspected perforation. Adverse events, including peritonitis, bleeding, and surgery, occurred in 3 patients (11.5 %). In both of these studies, the AXIOS stent was used for creation of the anastomosis (Boston Scientific, Natick, MA, USA).

However, a common technical problem exists in both of these series. The target duodenum or jejunum needed for creation of a GJ is collapsed and it is difficult to identify the target organ by EUS from the stomach. Furthermore, the insertion of the stent for creation of the anastomosis may be difficult with a collapsed bowel and this may result in catastrophic outcomes. Our group has overcome this problem with a novel device - the double balloon occluder. The device comprises of two balloons that could occlude a segment of bowel. Saline could then be injected in to the occluded segment resulting in distension of the bowel and making it possible to be targeted by EUS from the stomach.

Our group has published the results of using the double balloon occluder in conjunction with the AXIOS stent for creation of a GJ in 20 patients (Endoscopic ultrasonography-guided double-balloon-occluded gastrojejunostomy bypass - EPASS). The technical success rate was 90% (18/20). The median intubation time from the double-balloon tube intubation to stent placement was 25.5 min (range 10-39 min). Post-treatment gastric outlet obstruction scoring system (GOOSS) score improved in all 18 cases in which EPASS was successfully performed. The mean post-GOOSS score was significantly higher than the pre-GOOSS score (0.6±0.75 vs 2.94±0.23, p<0.001). None of the patients with successful placement of the stent suffered from adverse events or recurrent obstruction.

The outcomes of EUS-GJ were recently compared with that of conventional procedures for management of malignant GOO. In a retrospective study, EUS-GJ was compared with DS. 30 patients received EUS-GJ and 60 patients DS. The technical success rates were similar between the two groups (96.7% vs. 86.7%, P=0.07). Clinical success was also similar between the two groups (70.0% vs. 86.7%, P=0.08). The re-intervention rate was significantly higher in the ES group (43.3% vs. 3.4%, P<0.001), whilst adverse events were comparable (13.3% vs 18.3%, P = 0.549). On multivariate analysis, DS was independently associated with need for re-intervention (OR 25.7, p=0.004). In another study, EUS-GJ (25 patients) was compared with laparoscopic gastrojejunostomy (29 patients) 20. Technical success was comparable (88% vs 100%, P = 0.11). Clinical success was similar (84% vs 90%, P = 0.11). Average post-procedure length of stay was 9.4 days in the EUS-GJ group and 8.9 days in the Lap-GJ group (P = 0.75). Adverse events were significantly more the lap-GJ group (41% vs 12%, P = 0.03).

Thus, based on the above results, EUS-GJ may be associated with improved outcomes as compared with conventional procedures for management of malignant GOO. Hence, the aim of the current study is to compare the efficacies of EPASS versus uncovered pyloro-duodenal stents (DS) in unresectable malignant gastric outlet obstruction in a randomized setting.

Inclusion Criteria:

• Consecutive patients ≥ 18 years old
• Confirmed unresectable distal gastric or duodenal or pancreatico-biliary malignancies
• Suffering from gastric outlet obstruction with a gastric outlet obstruction score of ≤ 1
• Performance status ECOG ≤3

Exclusion Criteria:

• Prior metallic stent placement
• Severe comorbidities precluding the endoscopic procedure (such as cardiopulmonary disease, sepsis, or a bleeding disorder)
• Life expectancy of less than 1 month
• History of gastric surgery
• Linitus plastic
• Multiple-level bowel obstruction confirmed on radiographic studies such as small bowel series or abdominal computed tomography
• Coagulation disorders
• Pregnancy
• Unable to give informed consent

Recruited patients would not be informed of the type of treatment that was received.