关键词: 胆总管结石, 腹腔镜经胆囊管胆总管取石术, 经胆囊管, 血清总胆红素, 预测模型

Abstract: Objective The aim of this study was to construct and evaluate a predictive model by abnormal total serum bilirubin (TSB) level at 72 hours after laparoscopic transcystic common bile duct exploration (LTCBDE). Methods A retrospective analysis was conducted on the clinical data of 158 patients with common bile duct stone who underwent LTCBDE at Gaoyou People's Hospital between January 2023 and December 2024. Lasso and multivariate Logistic regression analysis were employed to identify clinical risk indicators for elevated TSB levels, by which a Logistic predictive model was established. The discriminative ability of the model was assessed by using receiver operating characteristic (ROC) curves, while calibration curves were used to verify its predictive accuracy. Decision curve analysis (DCA) was applied to evaluate its clinical utility. Internal validation of the model was performed by Bootstrap method. Results TSB levels elevated in 44 cases (27.8%) post-operationally in our series; Lasso regression and multivariate Logistic regression analysis revealed that common bile duct stone diameter ≥10 mm (OR=12.272, 95% CI: 3.853-39.084, P<0.001), multiple stones (OR=11.317, 95% CI: 3.956-32.377, P<0.001) and serum albumin level (OR=0.841, 95% CI: 0.761-0.930, P<0.001) were the impacting factors for elevated TSB levels 72 hours after LTCBDE; the three predictors were incorporated into multivariate Logistic regression model, and the constructed predictive model demonstrated a significant diagnostic efficacy, with the area under the ROC curve (AUC) of 0.900 (95% CI: 0.8509-0.9484); the optimal cutoff value was 0.232, achieving a balance between specificity (0.798) and sensitivity (0.886); the calibration curve showed a high agreement with actual values, and the Brier score was 0.115, indicating excellent predictive performance; additionally, the decision curve revealed net clinical benefit; during internal validation, the model exhibited an AUC of 0.892 (95% CI: 0.8399-0.9442), demonstrating robust discriminative ability and calibration. Conclusion The clinical predictive model based on CBDS diameter, multiple CBDS and serum albumin level could effectively identify patients at risk of elevated TSB at 72 hours after LTCBDE, providing valuable reference for clinical decision-making and individualized assessment.

Key words: Common bile duct stone, Laparoscopic transcystic common bile duct exploration, Via cystic duct, Total serum bilirubin, Predictive model