To investigate factors affecting ampullary access of endoscopic retrograde cholangiopancreatography (ERCP) in patients undergoing Billroth Ⅱ gastrectomy.

A retrospective analysis was performed on data of 261 patients with Billroth Ⅱ gastrectomy who underwent ERCP at Changhai Hospital from January 2008 to December 2017. Multivariate logistic regression analysis was used to analyze the potential factors affecting successful ampullary access, and receiver operating characteristic (ROC) curve was used to assess the predictive ability of potential factors.

A total of 345 ERCP sessions were collected. The successful ampullary access and cannulation rate were 82.3% (284/345) and 89.1% (253/284), respectively. The main reasons for ERCP procedural failure were unable to reach the duodenal blind end and find the papilla (66.3%, 61/92) and failure of selective cannulation (33.7%, 31/92). The ERCP-related complication rate was 14.2% (49/345), with post-ERCP pancreatitis rate was 3.2% (11/345). Multivariate logistic regression analysis indicated that the first ERCP attempt (OR=7.717, 95%CI: 2.581-23.068. P<0.001), with Braun anastomosis (OR=8.737, 95%CI: 2.479-30.797, P=0.001), and no cap-assisted forward-viewing gastroscope (OR=2.774, 95%CI: 1.283-5.997, P=0.009) were independent risk factors for failure of ampullary access. According to the B value of each risk factor in logistic regression analysis, that is, no cap-assisted as 1 point, the first ERCP attempt as 2 points, and Braun anastomosis as 2 points, the area under ROC curve was 0.773. When the cut-off point was 2.5, the sensitivity and specificity were 75.0% and 70.8%, respectively.

The first ERCP attempt, with Braun anastomosis, and no cap-assisted forward-viewing gastroscope are risk factors for failure of ampullary access of ERCP in Billroth Ⅱ gastrectomy patients. Early identification of high-risk patients may help to improve the success rate of ampullary access.

Cholangiopancreatography, endoscopic retrograde; Gastrectomy, Billroth Ⅱ; Risk factors