Selective Embolization for Post-Endoscopic Sphincterotomy Bleeding: Technical Aspects and Clinical Efficacy

Abstract

Objective: The objective of this study was to evaluate the technical aspects and clinical efficacy of selective embolization for post-endoscopic sphincterotomy bleeding.Materials and Methods: We reviewed the records of 10 patients (3%; M:F = 6:4; mean age, 63.3 years) that underwent selective embolization for post-endoscopic sphincterotomy bleeding among 344 patients who received arteriography for nonvariceal upper gastrointestinal bleeding from 2000 to 2009. We analyzed the endoscopic procedure, onset of bleeding, underlying clinical condition, angiographic findings, interventional procedure, and outcomes in these patients.Results: Among the 12 bleeding branches, primary success of hemostasis was achieved in 10 bleeding branches (83%). Secondary success occurred in two additional bleeding branches (100%) after repeated embolization. In 10 patients, post-endoscopic sphincterotomy bleedings were detected during the endoscopic procedure (n = 2, 20%) or later (n = 8, 80%), and the delay was from one to eight days (mean, 2.9 days; ± 2.3). Coagulopathy was observed in three patients. Eight patients had a single bleeding branch, whereas two patients had two branches. On the selective arteriography, bleeding branches originated from the posterior pancreaticoduodenal artery (n = 8, 67%) and anterior pancreaticoduodenal artery (n = 4, 33%), respectively. Superselection was achieved in four branches and the embolization was performed with n-butyl cyanoacrylate. The eight branches were embolized by combined use of coil, n-butyl cyanoacrylate, or Gelfoam. After the last embolization, there was no rebleeding or complication related to embolization.Conclusion: Selective embolization is technically feasible and an effective procedure for post-endoscopic sphincterotomy bleeding. In addition, the posterior pancreaticoduodenal artery is the main origin of the causative vessels of post-endoscopic sphincterotomy bleeding.