Identification of significant difficulty of selective deep cannulation by a simple predictive model: an endoscopic scale for teaching ERCP

Abstract

Background: Endoscopic retrograde cholangiopancreatography (ERCP) is an advanced endoscopic procedure associated with a risk of serious complications. This cohort study was designed to assess the accuracy of an endoscopic method aimed at discriminating procedures eligible to teach ERCP: patients with and without significant difficulty of selective deep cannulation (DSDC).

Methods: Clinically relevant variables were analyzed in a cohort of 400 consecutive patients (estimation group = 250 patients; validation group = 150 patients) who underwent an ERCP procedure.

Results: Multivariate analysis identified fixated duodenum, inflamed duodenum, soft major papilla, previous biliary surgery, and papilla with ectopion as independent predictors of DSDC. We constructed a model and a score system combining these five variables. The area under the ROC curve was 0.81 for the estimation group and 0.80 for the validation group. Using the best cutoff score (> 1.63), absence of significant DSDC could be excluded with high accuracy (negative predictive value = 89.2%) in 111 (44.4%) of 250 patients. Similarly, it could be excluded with the same certainty in 77 (51.3%) of the 150 patients in the validation group.

Conclusions: A combination of easily accessible variables accurately predicts the absence of significant DSDC in half the patients who underwent the ERCP procedure. This score system discriminates procedures eligible to teach ERCP.