Endoscopic retrograde cholangiopancreatography (ERCP) is the gold standard in evaluation of the biliary tree for choledocholithiasis. Formal indications for magnetic resonance cholangiopancreatography (MRCP) in suspected choledocholithiasis are lacking. Our objective was to determine if MRCP affects management of patients who otherwise would undergo ERCP. A review was conducted of all MRCPs and ERCPs at our institution from 2008 to 2012 with suspected choledocholithiasis. Patients who underwent MRCP and ERCP were compared with those who underwent ERCP alone. Demographic data were collected and notation of whether a post-MRCP ERCP occurred was the primary variable. MRCP was performed in 107 patients for choledocholithiasis. Eighty-eight patients were negative for choledocholithiasis (82%) and 76 were discharged without ERCP (71%). Thirty-one patients received a diagnosis of choledocholithiasis and were referred for ERCP. Of the 19 patients with MRCP-diagnosed common bile duct stones, 95 per cent were confirmed by ERCP (odds ratio 18.0, P < 0.05; agreement 77%, sensitivity 0.76, specificity 0.86, positive predictive value 0.95, negative predictive value 0.50). Length of stay was similar for all groups. A total of 131 patients underwent ERCP without a preprocedural MRCP. Choledocholithiasis was found in 116 patients (92%), whereas 12 patients (9%) had no common bile duct stones and three had an alternate diagnosis. In conclusion, MRCP significantly affected the management of patients who would have undergone ERCP. MRCP did not increase length of stay and contributed to the 95 per cent positivity rate of subsequent ERCPs. These data illustrate the utility of MRCP in suspected choledocholithiasis patients at a low cost with regard to risk and time.