960 Maximizing success in single-session edge - predictive factors of stent migration
Shinn B, Boortalary T, Raijman I, Nieto J, Khara HS, Kumar SV, Confer B, Diehl DL, El Halabi MAAN, Ichkhanian Y, Runge TM, Kumbhari V, Khashab MA, Tyberg A, Shahid HM, Sarkar A, Gaidhane M, Bareket R, Kahaleh M, Piraka C, Zuchelli T, Law R, Sondhi AR, Kedia P, Robbins JS, Calogero C, Bakhit M, Chiang AL, Schlachterman A, Kowalski TE, and Loren DE. 960 Maximizing success in single-session edge - predictive factors of stent migration. Gastrointestinal Endoscopy 2020; 91(6):AB80.
Background: Roux-en-Y gastric bypass anatomy poses a challenge in performing ERCP due to the inability to access the excluded stomach and duodenum. EUS-directed gastro-gastric ERCP (EDGE) uses a lumen apposing metal stent (LAMS) to access the excluded stomach and is most often performed in two-stages with initial placement of the LAMS followed by ERCP weeks later. The need for urgent ERCP in this patient population has led to the advent of single-session EDGE with LAMS and ERCP on the same day. The most serious and common complication of this procedure is intra-procedural migration of the LAMS. The purpose of this study was to identify predictive factors of LAMS complications during single-session EDGE in a large multicenter cohort. Methods: A multi-centered, retrospective chart review was conducted at nine tertiary medical centers. Single-session EDGE procedures were identified and data including age, gender, LAMS diameter, dilation, route of LAMS placement (gastric or jejunal), stent fixation and procedure complications were collected. The primary outcome was intra-procedural LAMS migration. Groups were compared using Fisher’s exact test for univariate analysis and binary logistic regression analysis to predict independent factors associated with stent migration. Results: 131 patients were included in the study who underwent single-session EDGE. Median age=58 y, 74.8% were female. LAMS migration occurred in 12 patients (9.1%). In univariate analysis, statistically fewer LAMS migrations occurred in patients with a 20mm diameter LAMS (2/84) vs 15mm (7/42), p=0.006, those with suture fixation, p=0.032, and those who underwent LAMS dilation, p=0.026. Location of LAMS placement did not affect the rate of stent migration (2/40 jejunal vs 10/91 gastric, p=0.343) although a trend towards more migration in the trans-gastric route was observed. The use of electrocautery-enhanced stents did not influence the rate of migration. Double pigtail stents for anchoring did not reduce the rate of migration. Binary logistic regression analysis of stent diameter, dilation, suture fixation, access route and electrocautery identified that 15mm stents were significantly more likely to migrate than 20mm (OR=7.9; 95% CI:1.3-47; p=0.024). Of the patients that experienced LAMS migration, 3 required surgery, 2 were rescued with esophageal stent bridging and 2 with a second LAMS. Bleeding occurred in 2 patients and was managed endoscopically at the time of the procedure. No deaths occurred in the cohort. Conclusions: Same day EDGE procedures can be performed safely with an acceptable complication rate. Larger diameter 20mm stents are the strongest predictor of a non-migrated LAMS, whereas stent fixation and dilation after deployment may also improve procedural success. Expanding on this cohort will offer further insights into the optimal technique for single-session EDGE.