A prospective comparison of magnetic resonance cholangiopancreatography with endoscopic retrograde cholangiopancreatography in the evaluation of patients with suspected biliary tract disease.

AIM: To determine the diagnostic accuracy of magnetic resonance cholangiopancreatography (MRCP) compared with direct cholangiography in the detection of biliary tract disease. PATIENTS AND METHODS: MRCP was performed in 100 patients in whom direct cholangiographic correlation (ERCP, n = 98; PTC, n = 9; intraoperative cholangiography, n = 3) was available for comparison. The MRCP examinations were performed using a two-dimensional multi-slice, fast spin echo (FSE) technique and a local surface coil. The diagnoses at direct cholangiography were choledocholithiasis in 30 patients, benign and malignant strictures in 28 patients and normal bile ducts in 42 patients. The nature of the strictures (benign, n = 2; tumour, n = 18; lymphnode recurrence, n = 3; unknown histology, n = 5) was determined by one or more of the following procedures: surgery (n = 8), biopsy (n = 15), cytology (n = 6) and cross-sectional imaging/follow-up findings (n = 3). RESULTS: MRCP diagnosed choledocholithiasis with a sensitivity of 93%, specificity of 99% and accuracy of 97 %. It resulted in two false-negative and one false-positive findings when compared with direct cholangiography. MRCP accurately diagnosed the presence and level of strictures in all patients. The overall sensitivity, specificity and accuracy of MRCP in the detection of bile duct lesions were 97%, 98% and 97%, respectively. CONCLUSION: MRCP has a high diagnostic accuracy when compared with direct cholangiography in the detection of bile duct disease.