Endoscopic Retrograde Cholangiopancreatography at a Teaching Medical Center

Abstract

Background: Endoscopic Retrograde Cholangiopancreatography (ERCP) has facilitated the diagnosis and treatment of pancreatobiliary diseases for a few decades. The procedure is relatively difficult to perform. One guideline indicates that to meet the minimum eligibility criteria to perform the procedure alone, a trainee should perform at least 180 to 200 ERCP under supervision. However, information about how to teach ERCP is sparse. Aims: To collect information that will help determine the optimal length of time that an attending physician should allow a trainee to cannulate the common bile duct (CBD). Methods: From October 2002 to March 2003, 4 ERCP fellows and from October 2003 to March 2004, another 4 ERCP fellows participated in the study. Each fellow had performed about 40 ERCPs under supervision prior to the study. During the first study period, we allowed each fellow to have 10 minutes for cannulation and in the second period 5 minutes. If the CBD could not be cannulated deeply within the time limits, an attending continued the cannulation attempts. We recorded the deep CBD cannulation time for each ERCP procedure as the time from the initiation the cannulation attempts until the time when the catheter was passed deeply into the CBD. The same 2 experienced pancreatobiliary-attending physicians participated in the study for both years. Results: In the first period, the fellows deeply cannulated 5 out of 45 patients. In the second period, fellows deeply cannulated 6 of 35 patients. The deep cannulation rate by the fellows in the 2 periods was not different (P>0.05). The mean deep CBD cannulation time was 30.24±16.3 minutes in the first period that was significantly longer than 15.1±13.2 minutes in the second period (P<0.01). Conclusions: For these small groups of relatively inexperienced fellows, attempts at cannulation longer than 5 minutes did not increase their cannulation rate and may result in a more difficult cannulation subsequently for the attending physician. Background: Endoscopic Retrograde Cholangiopancreatography (ERCP) has facilitated the diagnosis and treatment of pancreatobiliary diseases for a few decades. The procedure is relatively difficult to perform. One guideline indicates that to meet the minimum eligibility criteria to perform the procedure alone, a trainee should perform at least 180 to 200 ERCP under supervision. However, information about how to teach ERCP is sparse. Aims: To collect information that will help determine the optimal length of time that an attending physician should allow a trainee to cannulate the common bile duct (CBD). Methods: From October 2002 to March 2003, 4 ERCP fellows and from October 2003 to March 2004, another 4 ERCP fellows participated in the study. Each fellow had performed about 40 ERCPs under supervision prior to the study. During the first study period, we allowed each fellow to have 10 minutes for cannulation and in the second period 5 minutes. If the CBD could not be cannulated deeply within the time limits, an attending continued the cannulation attempts. We recorded the deep CBD cannulation time for each ERCP procedure as the time from the initiation the cannulation attempts until the time when the catheter was passed deeply into the CBD. The same 2 experienced pancreatobiliary-attending physicians participated in the study for both years. Results: In the first period, the fellows deeply cannulated 5 out of 45 patients. In the second period, fellows deeply cannulated 6 of 35 patients. The deep cannulation rate by the fellows in the 2 periods was not different (P>0.05). The mean deep CBD cannulation time was 30.24±16.3 minutes in the first period that was significantly longer than 15.1±13.2 minutes in the second period (P<0.01). Conclusions: For these small groups of relatively inexperienced fellows, attempts at cannulation longer than 5 minutes did not increase their cannulation rate and may result in a more difficult cannulation subsequently for the attending physician.