Su1562 NEEDLE KNIFE SPHINCTEROTOMY FOR DIFFICULT BILIARY CANNULATION: FACTORS PREDICTING ITS FAILURE

Abstract

Needle-knife sphincterotomy (NKS) is the most popular technique for accessing the bile duct when conventional cannulation methods fail. The success rate of NKS has been reported from 71-92%. Certain factors may influence the NKS results but have not been reported in the literature. This retrospective study is to analyze predicting factors for NKS failure. Between 2004 and 2016, 397 NKS procedures (6.3% of the ERCP procedures for patients with an intact papilla) were performed for difficult cases of biliary cannulation. All procedures were performed by five endoscopists (A-E), and only endoscopist A had NKS experience before the study. Deep bile duct cannulation was successfully performed in 300 patients after NKS (75.6%, NKS success group) and failed in 97 patients (24.4%, NKS failure group). To identify predicting factors, both patient and technique factors were compared. Patient factors included age, sex, CBD diameter, major papilla status (periampullary diverticulum, enlarged or swelling papilla, low set papilla, impacted stone at ampulla of Vater, periampullary tumor, and surgically altered anatomy), and final diagnosis (choledocholithiasis, malignant/benign biliary stricture, bile leak, and suspected Sphincter of Oddi dysfunction). Technique factors included endoscopists’ experience and major bleeding during NKS. Major bleeding was the bleeding that could interfere with or eventually caused the interruption of NKS. Logistic regression analysis was performed to identify predictors of NKS failure. There was no significant difference in the median patient age (71 vs. 72 years) and male patients (52.6% vs. 55.4%) between the groups. Endoscopists A, B, C, D, and E performed 188, 93, 51, 45, and 20 NKS procedures, respectively, during the study. Their NKS success rates were 79.8%, 76.3%, 72.5%, 64.4%, and 65%, respectively. To calculate the endoscopists’ experience, endoscopist A was used as a reference; endoscopists B and C were considered as intermediate experience; the endoscopists D and E together, they were considered inexperienced. Base on univariate analysis, periampullary diverticulum, surgically altered anatomy, endoscopists’ experience, and major bleeding during NKS were the predictors of NKS failure. By multivariate analysis, surgically altered anatomy (OR: 2.42, 95% CI: 1.04-5.64, p = 0.041), endoscopists’ experience (OR: 3.38, 95% CI: 1.65-6.92, p = 0.001 for endoscopists D+E), and major bleeding during NKS (OR: 22.68, 95% CI: 10.333-49.78, p <0.001) remained significant predictors. Surgically altered anatomy, endoscopists’ experience, and major bleeding during NKS were the predictors of NKS failure. Major bleeding during NKS was the strongest predictor. Therefore, any bleeding during NKS has to be stopped before major bleeding begins. NKS may be reserved for experienced endoscopists.View Large Image Figure ViewerDownload Hi-res image Download (PPT)