Society: ASGE

Background
EUS-guided gastroenterostomy (EUS-GE) is a novel method for palliating gastric outlet obstructon due to unresectable malignancies. The procedure is associated with high technical and success rates and potentially lower recurrent obstructions, reinterventions and better gastric outlet obstruction scores. However, results from randomized controlled studies are lacking. Hence, the aim of the current study is thus to compare EUS-GE with the use of a double balloon occluder (EPASS) and duodenal stent (DS) under a randomized setting. We hypothesize that EPASS could reduce the re-intervention rates when compared to DS.
Methods
The was a prospective multi-centered randomized controlled study conducted between December 2020 to February 2022 in 7 institutions. We included consecutive patients suffering from malignant gastric outlet obstruction due to unresectable primary gastro-duodenal or pancreatico-biliary malignancies. Included patients were randomized to receive EPASS or DS. The primary outcome measurement was the 6 month re-intervention rate. Secondary outcomes include technical and clinical success, adverse events rate, mortality, pre and post stenting gastric outlet obstruction scores (GOOS), causes of stent dysfunction, the duration of stent patency, cost and quality of life assessment scores.
Results
During the study period, 97 patients were included (EPASS 48, DS 49). There were no significant differences in background demographics. The EPASS group was associated with significantly better 6 month stent patency rate (97.9% vs 71.4%, P < 0.001) and mean stent patency days [176.7 (22.7) vs 152.9 (59.1), P = 0.013]. The reintervention rate was significantly lower in the EPASS group (2.1% vs 28.6%, P < 0.001). The 1 month gastric outlet obstruction score was significantly better in the EPASS group [2.3 (0.7) vs 1.9 (0.95), P = 0.031]. There were no significant differences in technical and clinical success rates (95.8% vs 100%, P = 0.242 and 100% vs 91.7%, P = 0.117 respectively), procedural time [40.3 (16.6) vs 37.8 (91.7), P = 0.865], hospital stay [7.1 (9.4) vs 9.0 (7.2), P = 0.289], 30-day mortality [5 (10.4) vs 1 (2), P = 0.097] and 30-day adverse events [7 (14.6) vs 7 (14.6), P = 1].
Conclusions
EPASS significantly improved the stent patency and eating functions in patients suffering from malignant gastric outlet obstruction. This should be the procedure of choice when the expertise is available.

Comparison of outcomes between the 2 groups.

Comparison of outcomes between the 2 groups.

The Kaplan-Meier curve showing the cumulative stent patency rate of the 2 groups of patients. The stent patency rate is significantly higher in the EPASS group at 6 months (P < 0.001).

The Kaplan-Meier curve showing the cumulative stent patency rate of the 2 groups of patients. The stent patency rate is significantly higher in the EPASS group at 6 months (P < 0.001).

Background and Aims:
Gastric outlet and biliary obstruction are common manifestations of gastrointestinal malignancies and some benign diseases, for which a standard treatment would be surgical gastroenterostomy and hepaticojejunostomy (i.e. ‘double bypass’). Therapeutic endoscopic ultrasound (EUS) has allowed for the creation of an EUS-guided double bypass using EUS-guided gastroenterostomy (EUS-GE) and biliary drainage (EUS-BD). However, same-session double EUS-bypass has only been described in small proof-of-concept series and lack comparison with surgical double bypass.

Methods:
A retrospective multicenter analysis was performed of all consecutive same-session double EUS-bypass procedures performed in 5 academic centres. EUS-GE was performed using either WEST or EPASS-technique, whereas EUS-guided hepaticogastrostomy, choledochobulbostomy, rendez-vous and antegrade stenting were allowed for EUS-BD. Surgical comparators were extracted from these centers’ databases from the same time interval. Efficacy, safety, hospital stay, nutrition and chemotherapy resumption, long-term patency and survival were compared.

Results:
In total, 154 patients were identified, of which 53 (34.4%) received treatment with EUS and 101 patients with surgery (65.6%). At baseline, patients undergoing EUS exhibited higher ASA scores and a higher median Charlson Comorbidity Index (9.0 [IQR 7.0-10.0] vs. 7.0 [IQR 5.0-9.0], p<0.001).
Technical success (96.2% vs. 100%, p=0.117) and clinical success rates (90.6% vs. 82.2%, p=0.234) were similar when comparing EUS and surgery. Overall (11.3% vs. 34.7%, p=0.002) and severe adverse events (3.8% vs. 19.8%, p=0.007) occurred more frequently in the surgical group. In the EUS group, median time to oral intake (0 [IQR 0-1] vs. 6 [IQR 3-7] day(s), p<0.001) and hospital stay (4.0 [IQR 3-9] vs 13 [IQR 9-22] days, p<0.001) were significantly shorter.

Conclusion:
Despite being used in a patient population with more comorbid conditions and potentially more advanced disease stage, same-session double EUS bypass achieved similar efficacy with lower overall and severe adverse events when compared to surgery. The current study furthermore suggests that EUS-GE and EUS-BD can be performed safely and effectively in a single session. Although future efforts should be directed towards further optimizing outcomes of double EUS-bypass, and adequate long-term biliary drainage in particular, the current study underlines the usefulness of EUS in patients with both GOO and biliary obstruction.

Background: Enteral stents (ES) and the more recent, Endoscopic ultrasound guided Gastrojejunostomy (EUS-GJ) are endoscopic techniques for palliation of malignant gastric outlet obstruction with benefit in terms of early resumption of oral intake and relatively non-invasive nature of procedure. Previous studies have shown EUS-GJ to be superior to ES in terms of clinical success in short and long term. We hypothesized that the benefit of EUS-GJ was driven by better gastric emptying as compared to ES. The aim of this pilot study was to analyze the difference in gastric emptying times between EUS-GJ and ES.
Methods: Consecutive patients who underwent ES and EUS-GJ for palliation of gastric outlet obstruction at a tertiary care oncology institute were screened over 6 months between April and October 2022. Patients should have either undergone ES with WallFlex Enteral Stents (Boston Scientific Ltd., USA; 22 mm diameter) or EUS-GJ using 20 X 10 mm Hot Axios stents (Boston Scientific Ltd., USA). Exclusion criteria were poor performance status (ECOG >2) or other critical illness for which patient was hospitalised. On follow-up at 2 weeks after ES or EUS-GJ, patients underwent Gastric Emptying Study after consumption of a semisolid test meal (porridge 100-150 gm) labelled with 0.8-1.2 mCi of Technetium-99m Sulphur colloid using a dual-head gamma camera. Patients were nil-per-orally prior to procedure. Readings of gastric emptying were taken at 0, 1, 2 and 4 hours and t1/2 of semisolid emptying was calculated. In addition, Gastric Outlet Obstruction symptom score (GOOSS) at baseline and at 2 weeks were also compared. Primary outcome measure was Gastric emptying time with t1/2 of semisolid test meal. Normal value of t1/2 was 120 mins for semisolids as per previous Indian normative data. Secondary outcome was clinical success as measured by 2-point increase in GOOSS at baseline and at 2 weeks.
Results: Twenty-Six patients were enrolled in this pilot study (12 EUS-GJ and 14 ES). There were no differences with respect to age, sex, site of obstruction, Charlson comorbidity index, baseline performance status or presence of ascites (Table 1). There was significant difference in the median Gastric Emptying t1/2 between those who underwent EUS-GJ and ES (86.5 mins vs 133 mins)(p=0.036) (Figure 1). Abnormal emptying was seen in 1 (8.3%) patient who underwent EUS-GJ and 6 (42.8%) patients undergoing ES (p=0.08). Clinical success at 2 weeks was achieved in 91.6% patients undergoing EUS-GJ and 71.4% patients undergoing ES (p=0.33).
Conclusion: There is a significantly delayed gastric emptying after ES as compared to EUS-GJ. Narrowing secondary to tumor itself may impact clearance of stomach contents. Rapid emptying of gastric contents may explain the higher clinical success rate achieved by EUS-GJ as compared to ES.

Introduction
Despite the advantages of EUS-guided Gastroenterostomy (EUS-GE) over enteral stenting (ES) for malignant Gastric Outlet Obstruction (mGOO) in retrospective series, ES is still advocated for patients with shorter life expectancy, whilst no prospective comparison is available.

Methods
All consecutive patients with mGOO treated between March-2021 and June-2022 in a tertiary, academic center were allocated to EUS-GE versus ES after multidisciplinary discussion and included in a prospective registry (PROTECT, NCT04813055) with monthly follow-up.
Technical/Clinical Success, Adverse Events, Symptom’s recurrence, and Survival were compared after 1:1 matching for primary disease, disease stage, ASA score and Charlson Comorbidity Index (CCI).

Results
During study period, 52 EUS-GE and 28 ES were performed (higher baseline ASA score [p=0.02] in the ES group).
After matching, 22 patients per arm were analysed, with no baseline differences in age, sex, BMI, primary disease (pancreatic cancer=86%) and stage (metastatic=69%), CCI (8 [5-9]) and ASA score.
Technical success was 100% in both arms (p=1). Patients treated with EUS-GE experienced higher clinical success (ability to eat at least a soft solid: 100% vs 73%, p=0.01) and shorter refeeding time (2[1-2] versus 3.5[2-7] days, p=0.002), with a trend to reduced adverse events (5% vs 14%, p=0.3). During a median FU of 75[42-103] and 45[18-143] days respectively, symptoms’ recurrence was 5% vs 29% (p=0.04) without any difference in overall survival.

Conclusions
In this first, prospective, matched comparison including frailer patients with more advanced neoplasms, EUS-GE confirmed higher and faster clinical success than ES, with reduced dysfunction and without any increased invasiveness.

Introduction: Endoscopic-guided gastrojejunostomy (EUS-GJ) is a novel procedure typically used for enteral decompression in patients with gastric outlet obstruction (GOO). The procedure involves placement of a lumen apposing metal stent (LAMS) from the stomach to a loop of jejunum distal to the obstruction. Patients with GOO commonly develop biliary obstruction that may require endoscopic intervention. However, due to gastric or duodenal obstruction, passage of the duodenoscope through the pylorus to the duodenum is often not feasible. Subsequently, it has been recognized that a previously created EUS-GJ can be utilized for biliary access by passing an endoscope from the stomach across the LAMS to the jejunum with retrograde advancement to the papilla. The utilization of EUS-GJ for biliary intervention has not yet been fully evaluated in the literature.

Methods: A multi-center retrospective chart review was conducted at five tertiary care hospitals. Patients who previously underwent EUS-GJ and subsequently had the GJ utilized for biliary access were identified. Data regarding patient demographics, procedural details and complications were collected. The primary outcome was success accessing the bile duct utilizing the GJ. Secondary outcomes included technical success of the biliary intervention and complications.

Results: EUS-GJ was utilized for biliary access in 45 patients (Table 1). Success accessing the bile duct utilizing the GJ was achieved in 42/45 patients (93%) all of whom also had technical success of the intended biliary intervention. 32 of the 42 successful cases accessed the bile duct with a duodenoscope. The average time between the EUS-GJ procedure and utilization of the GJ for biliary access was 76 days (range 0-519). Four patients underwent attempted utilization of the GJ for biliary access on the same day it was placed with 3/4 cases being successful. In two cases where biliary access was not achieved, the papilla was reached however the bile duct was not able to be cannulated. In the third unsuccessful case, the scope was able to reach the second portion of the duodenum however the papilla could not be identified due to friability and edema. Overall, complications were minimal with one case of bleeding which was managed conservatively and one stent migration. The migration occurred 219 days after the EUS-GJ and the stent was repositioned without perforation.

Discussion: Although typically placed to allow for resumption of oral intake in patients with GOO, EUS-GJ can also be used to access the bile duct. Data from this multi-center cohort demonstrates a 93% success rate accessing the bile duct with 100% technical success rate of biliary intervention utilizing the EUS-GJ with minimal complications. These outcomes suggest that utilization of EUS-GJ is a safe and effective way to perform biliary interventions in patients with GOO.

Table 1. Demographics, Procedure Details and Outcomes with EUS-GJ for Biliary Access

Table 1. Demographics, Procedure Details and Outcomes with EUS-GJ for Biliary Access

Background/Aims: Endoscopic ultrasound-directed transgastric ERCP (EDGE) is a safe and efficacious procedure to treat pancreaticobiliary diseases in patients with Roux-en-Y gastric bypass (RYGB). Stent fixation during the procedure can likely decrease the risk of stent migration. We aim to compare the rate of stent migration in patients who underwent EDGE with/without stent fixation.

Methods : Patients with RYGB who underwent Single Session-EDGE between 2017 and 2022 from 8 institutions were captured in a dedicated registry. Patient demographics, procedural details, and long-term clinical outcomes were measured and analyzed with descriptive statistics.

Results: 113 patients were included (mean age 60, 30% male). 54 (48%) patients has had stent fixation during the procedure – ( group A), while 59 (52%) did not – (group B) (See Table 1 and Table 2). The primary procedure indication was choledocholithiasis/cholangitis in 38% of
patients. 64% of the procedures were done acutely as an inpatient. Technical success of LAMS placement was 103/113 (91.1%). Technical success of intervention was 87% in group A and 95% in group B. 54 patients (100%) and 51 (95%) patients had a
20mm LAMS in groups A and B respectively. The mean procedure time was 91 and 79 minutes in groups A and B respectively.
The most commonly reported complication was stent dislodgement/migration with 7 stent migrations (12%) in group B compared to 2 (3.7%) in group A. One patient required surgical removal of the LAMS due to migration to the terminal ileum causing small bowel obstruction.

Conclusion Stent fixation with suturing during EDGE prevents stent migration and potentially consequent adverse events. Further prospective studies are needed to help evolve EDGE into a more standardized procedure.

Introduction: Endoscopic ultrasound guided trans-gastric ERCP (EDGE) is used to access the remnant stomach and biliary tree in select patients with gastric bypass. One risk of this procedure includes unintentional weight gain as a persistent fistula functionally reverses the bypass. Our aim was to assess rates of confirmed EDGE fistula closure after closure by primary and secondary intent as well as to assess weight regain after EDGE.
Methods: A retrospective chart review was performed for patients who underwent EDGE procedure between 2018 to 2022. Continuous variables were described as mean or median ± Standard deviation (SD), categorical variables were described as percentages. Means were compared using independent t-test and medians were compared using the Mann-Whitney U test. We described rates of primary fistula closure, diagnostic tests performed for evaluation of closure and reported cases of confirmed closure. Pre-EDGE (baseline) weight was compared to weight at 6 months. Weight change was also compared at 1-year post-EDGE between those who underwent endoscopic fistula closure (Fistula closed or FC group) and those whose fistula was left patent (Fistula open or FO group). Weight is reported in kg.
Results: 35 patients were included in the analysis, 12 FC and 23 FO patients. Mean age was 63±13. 75% were female. 42.9% of the patients had a diagnostic test performed to follow up the fistula, most commonly an upper GI series. [SACM1] In the FO group, 52% had a diagnostic test of which 75% had a persistent fistula. In the FC group, a third of patients had a diagnostic test and all had confirmed fistula closure. There were no adverse events related to endoscopic fistula closure. Regarding weight changes, mean baseline weight in kg was not significantly different between the FC group (78.9) and the FO group (83.5), p=0.63. There was no difference between mean overall baseline weight (76.5) and weight at 6 months (76.6), p=0.94. %Total body weight change (%TBWC) at 1 year was not significantly different between both groups when taking weight at 6 months as baseline (+0.83 versus +1.74 for FC and FO groups respectively, p=0.79). When excluding patients with weight altering comorbidities, %TBWC at 1 year was also not significantly different between both groups and was +2.80% for the FC group versus +2.53% for the FO group, p=0.95. The average duration between EDGE and confirmation of fistula closure was 182 days.
Conclusion: Spontaneous fistula closure is not guaranteed after EDGE procedure and endoscopic closure is safe and highly effective. While our limited data shows no significant difference in weight regain, a trend towards increased weight gain after leaving the fistula open was seen and larger studies are needed to further clarify this risk.

Mihir Wagh

University of Colorado

Appearances