Outcomes of single-endoscopist-performed needle-knife fistulotomy for selective biliary access in 842 consecutive patients: learning curve and changes over a 14-year period in a retrospective study

Abstract

Background and aims

Needle-knife fistulotomy (NKF) has emerged as the preferred precut technique. From a late strategy, NKF has shifted to an early rescue technique and has been used recently as a primary method for biliary access. It is unknown how these changes have affected NKF outcomes. We analyzed the outcomes of NKF over time in a large cohort of patients.

Methods

Multicenter retrospective cohort study of 842 patients who underwent NKF for biliary access between 2006 and 2019. Patients were divided into four study periods according to a late or early cannulation strategy and to the use of post-ERCP pancreatitis prophylaxis (Period 1–Period 4). We assessed outcomes of NKF, learning curves and shifts over time.

Results

Bile duct access was obtained in 88.0% of the patients. The initial cannulation rate increased significantly from 77.5% in P1 to 92.0% in P4 (p < .001). An endoscopist can obtain 80% success rate after performing 100 NKF procedures (95% CI: 0.79–0.86) and a 95% success rate after 830 procedures (95% CI: 0.92–0.98). Adverse events and pancreatitis were observed in 6.5% and 4.9% of patients respectively. The rate of pancreatitis was not significantly different during the 4 periods (p = .190). A decline in the pancreatitis rate was observed from 2006 until 2016 (no trainees) and then an increase until 2019 (trainees involved). The presence of trainees increased the rate of pancreatitis in the last period by 9.9%.

Conclusions

The success of NKF has increased significantly over the years, initially in a rapid manner and then more slowly. It is associated with a low rate of complications, which tend to decrease with experience. The involvement of trainees is associated with an increased rate of pancreatitis.

Keywords:

• Needle-knife fistulotomy
• cannulation rate
• post-ERCP pancreatitis
• large cohort
• ERCP adverse events

Disclosure statement

The authors declare that they have no competing interests relevant to this research.

Data availability statement

The datasets generated and/or analyzed in the present study are available from the corresponding author on reasonable request.