OBJECTIVE The aim of this study was to compare prospectively the image quality of magnetic resonance cholangiopancreatography (MRCP) using manganese chloride tetrahydrate (Bothdel Oral Solution 10) (MCT), a new negative oral contrast agent; ferric ammonium citrate (FerriSeltz powder 20%) (FAC); and no agent. METHODS MRCP images (TE 970 ms) of patients administered MCT (n = 19) or FAC (n = 20) at random, and 18 patients without an agent were evaluated. The subjective image quality of the overall, extrahepatic bile duct, and pancreatic duct and the degree of elimination of gastrointestinal fluid scored by two radiologists blinded to information regarding the agent were compared using Mann-Whitney's U-test. RESULTS The degrees of elimination of gastroduodenal fluid of MCT and FAC were significantly better than those without an agent (P < 0.01 and P < 0.01). The subjective image quality of MCT of the overall and extrahepatic bile duct were significantly better, although no significant differences for FAC were observed compared with those without an agent (P < 0.01 and P = 0.21, P = 0.02 and P = 0.16). There were no significant differences for the pancreatic duct (P = 0.12 and P = 0.19), nor were there any significant differences in the evaluations between MCT and FAC (P = 0.19-0.98). CONCLUSIONS MCT has shown performance comparable to that of conventional FAC in terms of pancreatic and biliary depiction and safety.