Failed Cannulation After Initial ERCP: Benefit of a Tertiary Referral ERCP Service

Abstract

Background: Competent biliary endoscopists fail selective biliary cannulation between 5 and 10% of cases and unsuccessful cannulation is reported in up to 20% of published series. Published data confirms that patients with previously failed cannulation at ERCP are often referred for alternative therapeutic techniques with inferior safety and accuracy profiles. Prospective data on outcomes after referral to a tertiary high volume ERCP center with greater experience is absent, however if this clinical pathway was confirmed as viable and effective then substantial cost savings and improved clinical outcomes might be anticipated. Methods: Prospective recruitment of patients with previously failed ERCP cannulation referred by other specialist biliary endoscopists. Prospective data collection included initial procedural indication, reason (if any) for failed cannulation, along with detailed cannulation parameters at the tertiary center. This group was compared to all other ERCP procedures performed at the same institution during the same period. 24 hour and 30 day follow up was obtained in all patients. Results In the 24 months to November 2008, a total of 1676 ERCP procedures were performed. 35 patients (20 female, mean age 68(range 40-88)) were referred after previous failed ERCP cannulation. Indications were biliary stones(16), pancreatic malignancy(6), biliary strictures(5), bile leak(2) and cholangitis(2). Successful cannulation was achieved in 100% of cases, 10(29%) requiring needle knife sphincterotomy (NKS). Identifiable reasons for initial failed cannulation included long intraduodenal papilla(5), peripapillary diverticulum(5), difficult/unstable duodenal position(4) and small papilla(4). There was one episode of mild post ERCP pancreatitis. There were no deaths or perforations. In the non-tertiary referral procedural group (1676) the cannulation rate for naïve papilla (651) was 97.8% with a NKS rate of 12.9%. Conclusions: Referral to a high volume centre following an initial unsuccessful ERCP procedure is associated with high technical success and a favorable complication rate compared to routine ERCP procedures. This data suggests that this type of clinical pathway should be developed as a part of an ERCP treatment algorithm. Tabled 1 Number Cannulation success NKS rate Post ERCP pancreatitis Other ERCP complications Non tertiary referral naïve papilla procedures 651 97.8% 12.9% 6.3% Perforation 0.15% Bleeding 0.6% Tertiary referral ERCP procedures 35 100% 29% 2.9% 0% p value 0.15 0.01 0.71 Open table in a new tab