Background: Common bile duct stones (CBDS) are present in 3-20% of patients with cholelithiasis. Intraoperative cholangiography has high sensitivity in detecting CBDS but its routine use is associated with increased costs and operating room time. The aim of our study was to define an accurate and simple model for the prediction of CBDS using preoperative variables. Methods: The study consisted of a retrospective analysis followed by a prospective study. Multivariate analysis of the retrospective data was used to create a predictive model for the presence of concurrent CBDS in patients undergoing cholecystectomy. The predictive model was then validated in a prospective series of 160 patients undergoing laparoscopic cholecystectomy. Results: Among the 19 potentially predictive variables for CBDS, only four were found to be statistically significant and independent: X1-alkaline phosphatase levels (UI/L); X2-number of gall-bladder stones; X3-total serum bilirubin (mg/dL); and X4-CBD diameter (mm). Using these four variables, the multivariate analysis created the equation: score = 0.002 × X1 + 0.485 × X2 + 0.232 × X3 + 0.220 × X4 - 4.167 to define the risk of CBDS in each patient. The predictive model, tested prospectively in 160 patients undergoing laparoscopic cholecystectomy (LC), showed an elevated index of correlation (r = 0.75) among the predicted and the observed frequencies (χ2 = 126.6; P <0.0001). The predictive model sensitivity and specificity were 92.9% and 99.3%, respectively. Conclusions: In patients undergoing cholecystectomy, accurate prediction of the risk for concurrent CBDS can be achieved using four preoperative variables. The use of this predictive model can contribute to reducing the number of unnecessary common bile duct explorations.
|Number of pages||6|
|Journal||Journal of Laparoendoscopic and Advanced Surgical Techniques|
|Publication status||Published - Oct 2004|
ASJC Scopus subject areas