BACKGROUND

Fluoroscopic intraoperative cholangiography (IOC) has been proposed as a safe and accurate screening method for choledocholithiasis, with a sensitivity and specificity of nearly 100% in selected cases. In the present study we retrospectively reviewed the diagnostic accuracy of IOC in cases highly suggestive of choledocholithiasis.

MATERIALS AND METHODS

Between January 1999 and December 2002, 103 patients underwent IOC as an imaging method for common bile duct (CBD) stone detection. We did not routinely perform IOC in all patients who were submitted to laparoscopic cholecystectomy, reserving the method for patients with a high probability of choledocholithiasis, namely patients with a history or the presence of painful obstructive jaundice at the time of referral, patients with a history of mild acute pancreatitis of biliary origin, and patients with abnormalities in their liver biochemistry profile as measured by liver function tests (LFT).

RESULTS

The mean rates of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for IOC were 98%, 94%, 94.5%, and 98% for the whole series, respectively. The diagnostic accuracy was 100% in patients with a history of obstructive jaundice or liver biochemical derangement, but was less in patients with a history of biliary pancreatitis. There were 3 false positive cases and 1 case of false negative results, all of which occurred in the subgroup of patients with a history of pancreatitis.

CONCLUSION

Selective fluoroscopic IOC is generally feasible and safe, as well as highly accurate (100%) for CBD stone detection in patients with obstructive jaundice or abnormal LFT. The PPV of the method decreases in patients with a history of pancreatitis (75%), while a negative result is highly suggestive of the absence of CBD stones (NPV = 98%). The present study concluded in a higher incidence of false results in patients with a normal size CBD, suggesting that the diagnostic accuracy of IOC is probably related to the size of the CBD rather than the indication for its performance.