Magnetic Resonance Cholangiopancreatography-Derived Anatomic Predictors of Difficult Biliary Cannulation During Endoscopic Retrograde Cholangiopancreatography: A Retrospective Cohort Study

Background Difficult biliary cannulation during endoscopic retrograde cholangiopancreatography (ERCP) is associated with procedural failure, prolonged manipulation of the papilla, and a higher risk of post-ERCP adverse events. Pre-procedural magnetic resonance cholangiopancreatography (MRCP) may provide anatomic markers that help anticipate technical difficulty before ERCP. Objective The main objective of this study is to assess whether MRCP-derived anatomic variables predict difficult biliary cannulation according to European Society of Gastrointestinal Endoscopy (ESGE) criteria, to evaluate their association with the burden of difficult cannulation, and to explore the relationship between difficult cannulation and ERCP failure. Methods We retrospectively analyzed 121 consecutive patients who underwent contrast-enhanced MRCP followed by ERCP. Multiplanar 3D MRCP reconstructions were reviewed by a radiologist blinded to ERCP outcomes. On reconstructed coronal images, the choledochoduodenal angle was measured between the distal common bile duct axis and the adjacent duodenal tangent at the point of biliary termination, while the distal common bile duct diameter was measured using an outer wall-to-outer wall technique within 2 cm proximal to the major papilla; papillary bulging suggestive of periampullary diverticulum was recorded. Difficult cannulation was defined by at least one ESGE criterion: more than five contacts with the papilla, more than five minutes spent attempting to cannulate after papillary visualization, or more than one unintended pancreatic duct cannulation or opacification. Multivariable logistic regression was used for global difficult cannulation and for individual ESGE criteria. Cumulative logit ordinal regression was used to model the number of ESGE criteria met. Because ERCP failure was infrequent, Firth penalized logistic regression was used for failure analyses. Results Difficult cannulation was present in 82/121 patients (67.8%), and ERCP technical success was achieved in 111/121 patients (91.7%). In multivariable analysis, a smaller distal common bile duct diameter independently predicted global difficult cannulation (OR 0.899, 95% CI 0.810-0.999; p = 0.047). For the individual ESGE criteria, a narrower choledochoduodenal angle (OR 0.977, 95% CI 0.955-0.999; p = 0.038) and a smaller distal common bile duct diameter (OR 0.884, 95% CI 0.794-0.984; p = 0.024) were independently associated with more than five papillary contacts, while a narrower choledochoduodenal angle was also associated with cannulation time greater than five minutes (OR 0.978, 95% CI 0.9573-0.9999; p = 0.049). In ordinal regression, both a narrower choledochoduodenal angle (OR 0.980, 95% CI 0.961-0.999; p = 0.040) and a smaller distal common bile duct diameter (OR 0.892, 95% CI 0.811-0.981; p = 0.019) were associated with a higher burden of difficult cannulation. MRCP anatomy alone did not predict ERCP failure. However, all failures occurred in patients with difficult cannulation, and each additional ESGE criterion increased the odds of ERCP failure by 3.66-fold (95% CI 1.67-11.39; p = 0.0003). Conclusions MRCP-derived anatomy, particularly distal common bile duct diameter and choledochoduodenal angulation, is associated with difficult ERCP cannulation and with a greater burden of cannulation difficulty. Final ERCP failure appears to be driven more strongly by the cumulative burden of difficult cannulation than by MRCP anatomy alone.