Prospective analysis of a scoring system to predict choledocholithiasis

Background: The management of choledocholithiasis in the laparoscopic era r emains debatable. A common policy is to perform preoperative endoscopic ret rograde cholangiopancreatography (ERCP) on patients suspected of having com mon bile duct (CBD) stones, using standard risk criteria. The aim of this s tudy was to evaluate prospectively a scoring system designed to improve the accuracy of CBD stone prediction before laparoscopic cholecystectomy. Methods: Known clinical, biochemical and radiological risk factors for CBD stones were analysed retrospectively in 233 patients. The presence (n = 77) or absence (n = 156) of CBD stones was determined by preoperative ERCP and /or laparoscopic cholangiography. Using multivariate analysis, the signific ant risk factors for CBD stones were identified and a new preoperative scor ing system was developed. A score of 3 or more was taken as the cut-off poi nt to suggest CBD stones and the need for preoperative ERCP. This scoring s ystem was then tested prospectively in 211 consecutive patients with sympto matic gallstones requiring surgery. Patients whose bile ducts could not be demonstrated by ERCP or operative cholangiography were excluded. Results: Fifty-five patients scored 3 or more (predicted ERCP rate of 29 pe r cent), of whom 23 (42 per cent) had proven CBD stones. Intraoperative cho langiography was successful in 87 per cent. Five patients (4 per cent) who scored less than 3 had small stones (less than 5 mm) demonstrated at operat ive cholangiography. The overall sensitivity and specificity of this scorin g were 82 and 80 per cent respectively. Conclusion: Formal risk assessment of the presence of CBD stones using this scoring system is simple and may be used for preoperative selection of pat ients for biliary tract imaging by magnetic resonance cholangiography or ER CP.