ABSTRACT
OBJECTIVES: To investigate the diagnostic efficacy of magnetic resonance cholangiopancreatography (MRCP) in choledocholithiasis and to determine whether use of MRCP may eliminate the need for purely diagnostic endoscopic retrograde cholangiopancreatography (ERCP). PATIENTS AND METHODS: A total of 116 patients with suspected biliopancreatic pathology were studied prospectively between November 1996 and February 1998. Choledocholithiasis was initially suspected in 61 patients and rated before ERCP and MRCP as being of low, intermediate, or high probability based on clinical, laboratory, and/or imaging findings (Cotton criteria). RESULTS: The sensitivity of choledocholithiasis diagnosis was 91%, with a global efficacy of 90%. The level of duct stone obstruction was visualized in all patients. Suprastenotic dilatation also showed a good correlation to ERCP. Choledocholithiasis was found in 32 patients (65%) and 3 patients (33%) in the high- and intermediate-probability groups, respectively. None of the low-probability patients had choledocholithiasis. Endoscopic retrograde cholangiopancreatography was performed for only a diagnostic (not therapeutic) purpose in 3 patients (6%) and 2 patients (22%) of the high- and intermediate-probability cases, respectively. CONCLUSIONS: Magnetic resonance cholangiopancreatography seems to be effective in diagnosing choledocholithiasis. It plays a fundamental role in patients with a low or intermediate risk of choledocholithiasis, contributing to the avoidance of purely diagnostic ERCP.
