Su1317 “AHHH, THIS INTERVAL IS JUST RIGHT!” THE STORY OF EDGE AND THE THREE INTERVALS. CAN A SHORTENED (SEVERAL DAY) INTERVAL BETWEEN EUS-GASTROGASTROSTOMY AND TRANSGASTRIC ERCP MITIGATE THE RISK OF STENT (LAMS) DISLODGMENT WHILE FACILITATING A TIMELY ERCP?

Abstract

EUS-directed transgastric ERCP (EDGE) is the newest endoscopic modality for treating pancreaticobiliary disorders in the setting of Roux-en-Y gastric bypass (RYGB). EDGE consists of EUS-directed gastrogastrostomy (GG) / jejunogastrostomy (JG) creation (step 1), followed by transgastric ERCP (step 2). These 2 procedural steps can be performed concomitantly (same session) or separately (different sessions). Single-session EDGE (EUS-GG and transgastric ERCP during index procedure) is immediately therapeutic but risks perforation via stent (LAMS) dislodgment. Dual-session EDGE (transgastric ERCP performed after GG fistula maturation) does not risk perforation, but the clinical malady festers during the 2-3 week interval required for fistula maturation. We propose a “shortened-interval dual-session” EDGE (2 – 4 day interval) to balance this dilemma. Our study compares 20-mm LAMS dislodgment risk between single-session and shortened-interval dual-session EDGE. We conducted a single-center retrospective study of 19 RYGB patients who underwent EDGE using 20-mm by one advanced endoscopist between 3/2018 – 9/2019. Given the small sample size, a permutation of regressor residuals test was conducted to investigate the association between EDGE interval type and LAMS dislodgment, controlling the effect of fistula type. All statistical analyses were performed on R (Version 3.6.1). Ten patients (5 ♀; mean 54 years) underwent single-session EDGE via GG (n = 6) or JG (n = 4) (Table 1). LAMS dislodgment occurred in 5 cases (50%), followed by successful intra-procedural rescue with a bridging esophageal FCSEMS. One adverse event (moderate severity) and 1 technical failure (10%) occurred in the same patient because transgastric ERCP was aborted and postponed after the rescue maneuver. Nine patients (7 ♀; mean 60 years) underwent a shortened-interval dual-session EDGE via GG (n = 5) or JG (n = 4). The median interval between EUS-GG and transgastric ERCP was 3-days (range 2-4). LAMS dislodgment occurred in 1 case (11%), followed by successful rescue with a bridging FCSEMS. Technical success was 100% and no adverse events occurred in this group. The permutation test showed that the odds of LAMS dislodgment using single-session EDGE was 841% that of using shortened-interval dual-session EDGE (OR 8.41), after controlling for the effect of fistula type (Table 2). There was significantly greater odds of stent (20-mm LAMS) dislodgment during single-session EDGE compared to shortened-interval dual-session EDGE, after controlling for the effect of fistula. We hypothesize that a shortened interval dual-session EDGE may decrease the risk of LAMS dislodgment by allowing for mucosal anchoring and complete expansion of the LAMS, prior to transgastric ERCP.