Objectives: To assess the diagnostic value of magnetic resonance cholangiopancreatography (MRCP) in detecting common bile duct stones in the preoperative investigation of patients electively referred for gallstone disease, to find out the incidence of asymptomatic common duct stones, and to correlate clinical symptoms and history and liver function tests (LFT) with the actual occurrence of common duct stones.
Design: Prospective study.
Setting: General hospital, Denmark.
Patients: 180 consecutive non-jaundiced patients referred with symptomatic gallstones for elective cholecystectomy.
Interventions: LFT, abdominal ultrasonography, MRCP, endoscopic retrograde cholangiopancreatography (ERCP), questionnaire.
Main outcome measures: Positive and negative predictive values and accuracy of MRCP, number of patients with asymptomatic stones, and correlation of symptoms with the presence of stones.
Results: 26/180 patients had common duct stones (14%). Only one (<1%) had an asymptomatic stone. For detection of such stones, MRCP's positive predictive value was 0.95 (95% confidence interval (CI): 0.86 to 1.00), negative predictive value 0.96 (0.93 to 0.99), and accuracy 0.85 (0.93 to 0.99). MRCP missed 5 stones 1-4 mm in size in 5 patients; 17/64 patients with raised LFTs had stones (27%). The probability of stones was highest when the patients had both raised LFTs and a dilated common (>7 mm) bile duct (82%). There were no readmissions with ductal stones in the 6-month postoperative period.
Conclusions: The predicive values of MRCP were fairly good, but MRCP misses some small stones <5 mm in size. Asymptomatic stones in the common duct are not common in this population and should not be screened for. The probability of stones increases with the number of predictive factors. Patients should be questioned carefully about signs of biliary obstruction, and only be offered preoperative MRCP should they have a suspicious history, raised LFTs, or a dilated common duct.