Outcomes of Endoscopic Retrograde Cholangiopancreatography (ERCP) and Sphincterotomy for Suspected Sphincter of Oddi Dysfunction (SOD) Post Roux-En-Y Gastric Bypass.

BACKGROUND: Sphincter of Oddi dysfunction (SOD) is thought to be a cause of chronic abdominal pain post Roux-en-Y gastric bypass, and current practice of performing endoscopic retrograde cholangiopancreatography (ERCP) with or without sphincterotomy is not supported by evidence. In addition to the complexity and risks of the procedure in patients with Roux-en-Y anatomy, the outcomes are uncertain and debatable. We performed a retrospective review and analysis of post-gastric bypass patients who had undergone ERCP with sphincterotomy to determine the effectiveness in patients with suspected SOD.
METHODS: Over a period of 5 years at the University of Minnesota, we retrospectively reviewed a prospectively collected database of a cohort of patients whom had a previous Roux-en-Y gastric bypass and whom had a subsequent ERCP for suspected SOD. Patients were categorized by modified Milwaukee classification, and outcomes were evaluated by patients' self-reporting of symptoms at follow-up.
RESULTS: We identified 50 patients who underwent laparoscopic-assisted gastrostomy for endoscopic retrograde cholangiopancreatography post Roux-en-Y gastric bypass over the study period. Within this group, 35 patients (70%) were suspected to have SOD. Nine patients (25.7%) were classified as type I, 19 patients (54.3%) type II, and seven patients (20%) type III. Thirty-four patients (97.1%) had biliary sphincterotomy, and 17 patients (48.6%) had both biliary and pancreatic sphincterotomy. Fourteen (40%) had repeated ERCP. At median follow-up of 11.5 months, type I SOD had two responders (25%), type II had nine responders (52.9%), and type III had one responder (14.3%). A subgroup analysis did not show significant differences in improvement of symptoms between patients whom had single versus repeated ERCP or biliary sphincterotomy alone versus both biliary and pancreatic sphincterotomy. Three patients (9%) had post-ERCP pancreatitis.
CONCLUSIONS: SOD in patients post Roux-en-Y gastric bypass is complex due to multiple confounding factors. Rome III and Milwaukee classification systems assist us in the diagnosis and treatment of sphincter dysfunction until we have a better way to predict treatment response post sphincterotomy. Current treatment is based on the type of disorder and anatomy of biliary ducts. Types I and II sphincter dysfunction particularly associated with dilated biliary duct on imaging have the best response to endoscopic sphincterotomy and therefore should be considered taking into account the risks and benefit. Repeated sphincterotomy and concurrent pancreatic sphincterotomy is generally not useful.