ObjectiveTo investigate the independent risk factors for acute biliary pancreatitis (ABP) induced by gallstones, and to develop a nomogram prediction model for the onset of ABP, thereby enabling early identification of high-risk patients and the formulation of individualized management plans, so as to provide a scientific basis for improving the clinical management of ABP. MethodsThe patients with gallstones who were admitted to the Department of Hepatobiliary and Pancreatic Surgery, Suining Central Hospital from January 2022 to January 2024 were collected. The patients were randomly divided into a training set and a validation set in a 7∶3 ratio. Univariate and multivariate logistic regression analyses were applied to identify independent risk factors for ABP and to develop a nomogram prediction model. The discriminative ability, calibration, and clinical utility of the prediction model were evaluated using the receiver operating characteristic (ROC) curve, calibration curve, Hosmer-Lemeshow test, and decision curve analysis. The accuracy and stability of the prediction model were further tested using the validation set. ResultsA total of 1 103 patients with gallstones were included, comprising 772 in the training set and 331 in the validation set. Among them, 137 (17.75%) and 60 (18.13%) patients in the training and validation sets, respectively, were complicated with ABP. Multivariate logistic regression analysis in the training set revealed that age [OR (95%CI)=1.03 (1.01, 1.04), P=0.002], normal gallbladder size [OR (95%CI)=5.36 (2.70, 10.65), P<0.001], gallstone diameter ≤1 cm [OR (95%CI)=6.26 (3.23, 12.14), P<0.001], round-shaped gallstones [OR (95%CI)=11.29 (5.42, 23.51), P<0.001], multiple gallstones [OR (95%CI)=1.37 (1.17, 1.81), P=0.013], and common bile duct diameter ≥1 cm [OR (95%CI)=4.44 (1.37, 14.41), P=0.013] were independent risk factors for ABP induced by gallstones. Based on these risk factors, a nomogram prediction model was constructed. The model demonstrated area under the ROC curve for distinguishing patients with ABP induced by gallstones were 0.873 [95%CI (0.842, 0.905)] in the training set and 0.858 [95%CI (0.810, 0.905)] in the validation set, respectively. The calibration of the model by the Hosmer-Lemeshow test indicated a good fit between predicted and actual probabilities (training set: χ2=14.061, P=0.080; validation set: χ2=7.656, P=0.468). Internal validation via the Bootstrap method (1 000 resamples) yielded calibration curves for both the training and validation sets that closely aligned with the ideal diagonal line. In both the training and validation sets, the patients with gallstone would benefit from intervention according to the nomogram at threshold probabilities of 0.03–0.88 and 0.89–0.91, respectively. ConclusionsThis study confirms that age, normal gallbladder size, gallstone characteristics (diameter ≤1 cm, round shape, and multiple stones), and common bile duct dilation (diameter ≥1 cm) are independent risk factors for ABP induced by gallstones. The nomogram prediction model constructed based on these factors demonstrates good discrimination and calibration in both the training and validation sets, and decision curve analysis confirmes its ideal clinical utility. This model provides clinicians with an intuitive and reliable quantitative tool for early identification of high-risk ABP patients and the development of individualized prevention and management strategies.
