Skip the sphincterotomy before bile duct stent placement

For patients with unresectable pancreatic cancer, self-expandable metallic stents (SEMS) are preferred to smaller-caliber, plastic stents for the palliation of jaundice. SEMS have longer patency rates and reduce the overall costs of care by obviating the need for stent exchanges in the majority of cases. Further, SEMS have largely replaced surgical bypass of the bile duct because of reduced morbidity and acceptable durability since most patients with unresectable pancreatic cancer are not expected to survive more than 1 year.

The decision to place a SEMS with complete or partial coating, primarily designed to reduce the likelihood of tumor ingrowth within the interstices of the stent, is more complex. Coating reduces ingrowth and modestly prolongs patency, but increases the likelihood of stent migration and possibly cholecystitis (due to occlusion of the cystic duct). Presumably due to their larger diameter than plastic stents, previous studies suggested a higher rate of post-ERCP (post-endoscopic retrograde cholangiopancreatography) pancreatitis, irrespective of preexisting endoscopic sphincterotomy.

Dr. Gregory Cote

Hayashi and his colleagues compared complication rates, time to stent occlusion, and survival time in a randomized clinical trial using partially covered SEMS with or without endoscopic sphincterotomy for biliary drainage of patients with unresectable pancreatic cancer. While the results favor the null (that is, no benefit of endoscopic sphincterotomy), there are a few take-home pearls. First, a difference in procedure time of 189 seconds (about 3 minutes) is longer than expected for most high-volume endoscopists who perform sphincterotomy. However, if primary cannulation is accomplished using a catheter and sphincterotomy requires exchange for a sphincterotome, the prolonged time (and costs associated with using two versus one device) is understandable.

Second, pancreatitis rates of 8%-9% are higher than expected for a low-risk indication such as malignant bile duct obstruction. The higher rates in this trial may be related to study design (patients in clinical trials are expected to have higher event rates due to meticulous follow-up compared with retrospective studies), use of SEMS in all patients, and specificity of the definition of post-ERCP pancreatitis in this population: Patients with pancreatic cancer are more likely to complain of pain following ERCP that is related to expansion of the metal stent, and a rise in serum pancreatic enzymes that may not necessarily connote true acute pancreatitis. This requires further investigation.

Third, bleeding and perforation are uncommon complications of ERCP with or without sphincterotomy when a high-volume physician performs the procedure. These data corroborate this - the argument that skipping the sphincterotomy will lower an already rare set of events is challenging. In fact, the only observed perforation occurred in a patient randomized to no sphincterotomy! Last, the higher rates (albeit not statistically significant) of hepatic abscess and cholecystitis among patients who did not undergo a sphincterotomy should not be ignored.

In some cases, sphincterotomy facilitates passage of the stent delivery catheter and tissue sampling devices, if needed. Nevertheless, endoscopists should be reassured by the equivalent rates of pancreatitis in this clinical trial. Based on these results, it seems appropriate to withhold sphincterotomy, at least in cases where bleeding may be high risk (e.g., with ongoing use of antiplatelet or anticoagulant medications).

Dr. Gregory A. Cote is an assistant professor of medicine in the department of medicine, division of gastroenterology and hepatology, Indiana University School of Medicine, Indianapolis. He is a consultant and member of the advisory board for Boston Scientific and a member of the advisory board for Olympus America.