The learning curve for safety and success of precut sphincterotomy for therapeutic ERCP: a single endoscopist's experience.

ABSTRACT

STUDY AIMS: The aims of this study were to evaluate the efficacy and safety of precut sphincterotomy in relation to the experience of a single endoscopist, and to establish the number of procedures required before achieving an effective and safe precut sphincterotomy. METHODS: A total of 200 consecutive patients underwent precut sphincterotomy carried out by a single endoscopist (T.A.) between January 2003 and December 2005. All of the procedures were divided into four chronological groups of 50 (Group I, II, III, and IV). Medical records and patient data were retrospectively reviewed and included procedure indications, outcomes, and complications. All patients were admitted for observation after the procedure in case of complications. RESULTS: A total of 200 patients (23.3%) (mean age 58.5 years; 101 men) underwent precut sphincterotomy (161 with needle-knife technique, 32 with septotomy technique, and seven with Erlangen technique). There was no mortality. The success rates of prompt bile duct cannulation after precut sphincterotomy were 88%, 86%, 94%, and 82%, respectively ( P > 0.05). Immediate bleeding requiring a submucosal adrenaline injection was observed in combined group I - II (28%) and combined group III - IV (7%) ( P < 0.05). One patient (2%) from each of group I, III, and IV required further endoscopic treatment for rebleeding. Duodenal perforation (2%) was detected and conservatively treated in one patient from group II. Mild pancreatitis was found in one patient (2%) in group III. CONCLUSIONS: The success rates of bile duct cannulation by precut sphincterotomy were not associated with the experience of the endoscopist. The postprocedural complications significantly decreased after the first 100 procedures. An experience of at least 100 procedures is suggested to achieve a safe precut sphincterotomy.