Abstracts
1997 Digestive Disease Week

Magnetic resonance cholangiopancreatography accurately predicts the presence or absence of choledocholithiasis.

SN Hochwald, SG Marcus, KS Naik, M Dobryansky, P Shamamian, N Rofsky, G Coppa. Departments of Surgery and Radiology, New York University Medical Center, NY, NY.

Objective: Non-invasive methods to evaluate calculi in the common bile duct (CBD) have limited accuracy and rely mainly on ultrasound and computed tomography. Accurate CBD imaging in patients with biliary calculi is an important determinant of specific therapy. This retrospective study was undertaken to assess the accuracy of MRCP in predicting the presence or absence of CBD stones in patients with biliary calculous disease. Methods: 26 patients with a final diagnosis of biliary calculous disease underwent MRCP over an 11 month period, starting in November 1995. Three groups were identified: CBD stones (n=11), gallstone pancreatitis (n=7) and cholelithiasis (n=9). One patient with gallstone pancreatitis had CBD stones. In all patients, the presence or absence of CBD calculi, as determined by MRCP, was correlated with the final diagnosis obtained from ERCP, intraoperative cholangiogram, CBD exploration or clinical follow up. Sensitivity, specificity and accuracy of MRCP were determined. Median follow up time was 4 months. Results: The major clinical indications for MRCP in the 26 patients were abnormal liver function tests (LFTs) followed by abdominal pain. LFTs were abnormal in 6 of 9 patients with cholelithiasis. Eleven patients were diagnosed with CBD stones and 15 were not.

CBD Stones                              Gallstone      MRCP       MRCP
Final Diagnosis   n    Cholelithiasis   Pancreatitis   Positive   Negative
Positive          11     0                1             10          1
Negative          15     9                6             1          14

MRCP correctly predicted the presence of CBD stones in 10 of 11 patients and failed to detect CBD stones in one patient with gallstone pancreatitis, who was found to have CBD stones at operation. MRCP incorrectly predicted the presence of CBD stones in 1 of 15 patients ultimately found not to have CBD stones. This patient had gallstones only. MRCP correctly predicted the absence of CBD stones in the other 14 patients, including 6 patients with gallstone pancreatitis. For all patients, MRCP had a sensitivity of 91%, specificity of 93% and accuracy of 92%. Conclusions: MRCP is an accurate, non-invasive test in evaluating the CBD duct for the presence or absence of calculi in patients suspected of having CBD stones. Our data support the use of MRCP in the preoperative evaluation of these patients as findings may influence therapeutic decisions.