PTH-021 Proximally migrated pancreatic stents – endoscopic management and outcome at a tertiary centre. (June 2019)

• Record Type: Journal Article
• Authors: Goodchild, George; Potts, Jonathan; Martin, Harry; Thorburn, Douglas; Chapman, Michael; Johnson, Gavin; Pereira, Stephen; Webster, George
• Is Part Of: Gut. Volume 68(2019)Supplement 2
• Page Start: A23
• Page End: A24
• Publication Date: 2019-06
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2019-BSGAbstracts.46
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: Pancreatic stents (PS) are increasingly used to aid biliary cannulation and reduce the risk of pancreatitis during endoscopic retrograde cholangiopancreatography (ERCP). Proximal PS migration is uncommon (reported in 2–5%) but poses a significant problem due to the risk of pancreatitis and pain. We describe our experience in the endoscopic management of migrated PS. Methods: A retrospective review of all ERCPs at our tertiary unit from January '13 –February '19 was performed to identify cases of migrated PS. Results: Twelve patients were identified; 10 received PS placement at other UK hospitals, 1 abroad and 1 at our unit. PS indications were prophylaxis of post-ERCP pancreatitis (PEP) in 9 and following pancreatic endotherapy in 3 (one for sphincter dysfunction, one for ductal stricture and one for cyst drainage). The migrated PS were 5 cm (n=11) or 7 cm (n=1) long with diameter of 3Fr (n=1), 5Fr (n=10) and 7Fr (n=1). Seven were straight and 5 single pigtail. 9 PS migrated after reported correct placement (of which 6 (67%) were straight). Two were inadvertently deployed within the pancreatic duct (PD); in one case there was no information regarding placement. The site of PS migration was within the pancreatic head (n=3), body (n=3) and tail (n=6). After MDM review one patient was referred directly for surgery where a PD stricture precluded endoscopic access. Endoscopic removal was attempted in 11 patients and successful in 9 (82%). 7 patients required two procedures. The median interval from stent placement to retrieval was 5 weeks (range 2–20). All patients had a previous pancreatic sphincterotomy. Techniques used for successful retrieval were: stent grabbers (n=3), extraction balloon/stent grabbers (n=1), snare/stent grabbers (n=2) and pancreatoscopy with SpyBite TM forceps (n=3). Endoscopic removal was unsuccessful in 2/11 patients where pigtail stents had hooked into a side branch. In all cases a 5 cm/5Fr prophylactic PS was placed post retrieval and 8 received rectal diclofenac (one contraindicated). Mild post-ERCP pancreatitis (PEP) occurred in one case. Conclusions: Endoscopic retrieval of proximally migrated PS was successful in 82% of patients. Migrated PS retrieval is challenging, and should only be attempted by endoscopists experienced in pancreatic endotherapy following case review within a specialist MDM. The mechanism of migration is unclear and spontaneous proximal migration of an appropriately placed single pigtail PS is unlikely to occur. Apparent migration in these cases most likely arises from incorrect deployment, potentially due to difficulty distinguishing the stent from the pushing device, emphasising the need for caution during PS placement. The optimal retrieval technique depends upon PS type, position, and PD anatomy. In our experience use of an extraction balloon alone was never successful. Failure of endoscopic retrieval was highest with pigtail stents and those >2 cm from the papilla.