Abstract
BACKGROUND: While there are guidelines for appropriate training in ERCP, these are non-existent for needle-knife precut. The aim of this study was: (1) evaluate the experience curve of three endoscopists in needle-knife fistulotomy (NKF); (2) propose a minimum number of NKF procedures to attest proficiency. METHODS: Between November 1997 and March 2011, the first 120 consecutive NKF performed by three endoscopists (A, B, and C) were selected (360 patients) from three centers. Each group of 120 patients was chronologically ordered into three subgroups of 40. The main outcomes were: NKF use, NKF success, and post-ERCP adverse events. RESULTS: The need for NKF did not decrease over time. The NKF success rate in the first attempt for endoscopist A and C in each of the three subgroups was 85 %/85 %, 87.5 %/87.5 %, and 87.5 %/90 %, respectively. Furthermore, both demonstrated a high NKF success in their initial 20 NKFs (85 % and 80 %, respectively). Endoscopist B however presented a different pattern as the success rate initiated at 60 %, then rose to 82.5 % and 85 % for the last group (P = 0.03). Adverse events were mild (28 of the 32 occurrences) with no clear reduction with increased experience. CONCLUSIONS: A skillful endoscopist may expect to master NKF easily with few adverse events. While some endoscopists could begin on their own because of their innate skills, a minimal training is needed for all, as we cannot predict skills in advance. We propose a minimum of 20 NKF precuts to attest a trainee's competence in this procedure.