Unnecessary Mrcp Prior To Ercp In Patients With Choledocholithiasis: The Role Of On-Site Ercp

INTRODUCTION: The American Society of Gastrointestinal Endoscopy (ASGE) recommends endoscopic retrograde cholangiopancreatography (ERCP) as the first step in the management of patients with high probability for choledocholithiasis, defined as the presence of a common bile duct stone on imaging, clinical acute cholangitis, a serum bilirubin greater than 4 mg/dL, or a serum bilirubin between 1.8 and 4 mg/dL with concomitant common bile duct dilation greater than 6 mm. However, magnetic resonance cholangiopancreatography (MRCP) is commonly obtained in high probability patients. This study investigates the frequency of these potentially “unnecessary” MRCP in general, and in those who were initially admitted to a center with available ERCP versus those who were initially admitted to a center with no on-site ERCP. METHODS: A retrospective review of an electronic endoscopy database and medical records was conducted. Patients with a diagnosis of choledocholithiasis on ERCP performed at the University of Maryland Medical Center between 1/1/2017 and 2/1/2020 were identified. ASGE guidelines were utilized to characterize patients who were high risk for choledocholithiasis and hence may not have needed MRCP prior to ERCP. The frequency of these “unnecessary” MRCP was calculated and then compared between patients who were initially admitted to the tertiary center, where on-site ERCP is available, and those who were initially admitted to a community hospital with no on-site ERCP. RESULTS: Overall, out of 226 patients who were high risk for choledocholithiasis and had unnecessary MRCP prior to ERCP, 97 patients (42.9%) underwent MRCP. Of 66 patients who were high risk for choledocholithiasis and initially admitted to a center with on-site ERCP, 17 (25.8%) underwent unneeded MRCP. While, of 160 high risk patients who were initially admitted to a center without on-site ERCP, 80 (50%) had an unneeded MRCP. The difference between the two groups was statistically significant (P = 0.001). CONCLUSION: This study is limited by its retrospective design which cannot account for all factors influencing the decision to perform MRCP. Nonetheless, the results suggest that centers with on-site ERCP perform fewer unneeded MRCPs compared to those without on-site ERCP. However, even in a center with available ERCP, a substantial number of patients may have had unnecessary MRCP. This provides an opportunity for quality improvement efforts and interventions that could lead to fewer unneeded MRCPs.