A Predictive Tool for Choledocholithiasis in Patients Undergoing Emergency Cholecystectomy

Area of Science:

• Gastroenterology and Hepatobiliary Surgery
• Surgical Diagnostics
• Predictive Modeling in Medicine

Background:

• Management of acute cholecystitis often involves emergency laparoscopic cholecystectomy.
• Accurate detection and management of concurrent choledocholithiasis remain challenging and debated.
• Improved predictive models are needed to guide clinical decisions.

Purpose of the Study:

• To develop a more accurate predictive model for choledocholithiasis in patients undergoing emergency cholecystectomy for acute cholecystitis.
• To identify preoperative factors associated with the presence of common bile duct stones.

Main Methods:

• A 9-year retrospective audit of patients undergoing emergency cholecystectomy.
• Evaluation of preoperative clinical, laboratory, and imaging factors.
• Development of predictive models using logistic regression and Receiver Operating Curve (ROC) analysis with Youden Index.

Main Results:

• 1601/1828 patients analyzed; choledocholithiasis associated with fever, elevated C-reactive protein, and higher median bilirubin (25.0 μmol/L vs 11.0 μmol/L).
• Multivariate analysis identified fever, biliary dilatation, and ultrasound-detected common bile duct stones as significant predictors (excluding bilirubin).
• Including bilirubin, levels of 20-39 μmol/L (OR 2.44) and ≥40 μmol/L (OR 4.84) significantly increased choledocholithiasis likelihood. ROC model achieved a C-statistic of 0.796.

Conclusions:

• A perfect predictive model for concurrent choledocholithiasis in acute cholecystitis is not yet achievable.
• The developed model can establish high and low-risk groups for choledocholithiasis, aiding clinical management.
• Preoperative factors, particularly bilirubin levels, are valuable in predicting common bile duct stones.