Objective To investigate the risk factors of acute biliopancreatic complica-tions in patients of pregnancy combined with gallbladder stone and construction of prediction model.

Methods The retrospective case-control study was constructed. The clinical data of 98 patients of pregnancy combined with gallbladder stone who were admitted to the First Hospital of Lanzhou University from September 2011 to October 2022 and 53 patients of pregnancy combined with gallbladder stone who were admitted to Gansu Provincial Hospital May 2014 to October 2021 were collected. The age of 151 patients was 29(25,32)years. Observation indicators: (1) situations of patients of pregnancy combined with gallbladder stone; (2) risk factors of acute biliopancreatic com-plications in patients of pregnancy combined with gallbladder stone; (3) construction of prediction model for acute biliopancreatic complications in patients of pregnancy combined with gallbladder stone. Measurement data with normal distribution were represented as Mean±SD, and comparison between groups was conducted using the independent t test. Measurement data with skewed distribution were represented as M(Q₁,Q₃), and comparison between groups was conducted using the Mann-Whitney U test. Count data were described as absolute numbers, and comparison between groups was conducted using the chi-square test. Comparison of ordinal data was conducted using the rank sum test. Univariate and multi-variate analyses were conducted using the Logistic regression model. Nomogram prediction model was conducted, and the receiver operating characteristic (ROC) curve was used to evaluate discri-mination of the nomogram predic-tion model. The calibration curve and clinical decision curve were used to evaluate calibration and net clinical benefit of the nomogram prediction model. Internal validation of the prediction model was performed by applying 10-fold cross-validation.

Results (1) Situations of patients of pregnancy combined with gallbladder stone. The total cholesterol, triglyceride, high density lipoprotein cholesterol, low density lipoprotein cholesterol, prepregnancy body mass index (<18.5 kg/m², 18.5‒24.0 kg/m², >24.0 kg/m²), gesta-tional period (early, mid, late), primipara (positive, negative), stone type (solitary, non solitary), diameter of stone (≤10 mm, >10 mm), gallbladder wall thickness (≥4 mm, <4 mm) were (4.9±1.4)mmol/L, 1.88(1.22,2.93)mmol/L, 1.48(1.22,1.83)mmol/L, (2.8±0.9)mmol/L, 13, 75, 58, 37, 45, 69, 86, 65, 37, 114, 89, 62, 38, 113 in the 151 patients of pregnancy combined with gallbladder stone. Of the 151 patients, the age, prepregnancy body mass index (<18.5 kg/m², 18.5‒24.0 kg/m², >24.0 kg/m²), primipara (positive, negative), stone type (solitary, non solitary), diameter of stone (≤10 mm, >10 mm), gallbladder wall thickness (≥4 mm, <4 mm) were 31(28,37)years, 3, 30, 36, 29, 40, 32, 37, 26, 43, 4, 65 in 69 cases without symptom, versus 27(24,31)years, 10, 45, 22, 57, 25, 5, 77, 63, 19, 34, 48 in 82 cases combined with acute biliopancreatic complications, showing significant differences in the above indicators between them (Z=‒3.636, ‒2.385, χ²=11.544, 32.862, 23.729, 25.310, P<0.05). Five of the 82 patients of pregnancy combined with gallbladder stone missed data of prepregnancy body mass index. Of the 82 patients, there were 42 patients of simple acute cholecystitis, 40 patients of common bile duct stone and/or acute biliary pancreatitis including 18 cases of common bile duct stone, 13 cases of acute biliary pancreatitis and 9 cases of common bile duct stone combined with acute biliary pancreatitis. (2) Risk factors of acute biliopancreatic complications in patients of pregnancy combined with gallbladder stone. Results of multivariate analysis showed that primipara, non solitary stone, diameter of stone ≤10 mm, gallbladder wall thickness ≥4 mm were independent risk factors of acute biliopancreatic complications in patients of pregnancy combined with gallbladder stone (odds ratio=3.102, 6.305, 3.674, 6.686, 95% confidence interval as 1.280‒7.519, 1.886‒21.080, 1.457‒9.265, 1.984‒22.528, P<0.05). Results of multivariate analysis in further analysis showed that primipara, non solitary stone, gallbladder wall thickness ≥4 mm were independent risk factors of simple acute cholecystitis in patients of pregnancy combined with gallbladder stone (odds ratio=3.671, 8.905, 7.137, 95% confidence interval as 1.386‒9.723, 2.332‒34.006, 1.902‒26.773, P<0.05), and age, non solitary stone, diameter of stone ≤10 mm, gallbladder wall thickness ≥4 mm were independent risk factors of common bile duct stone and/or acute biliary pancreatitis in patients of pregnancy combined with gallbladder stone (odds ratio=0.883, 5.361, 5.472, 8.895, 95% confidence interval as 0.789‒0.988, 1.062‒27.071, 1.590‒18.827, 2.064‒38.325, P<0.05). (3) Construction of prediction model for acute biliopancreatic complications in patients of pregnancy combined with gallbladder stone. The nomogram prediction model for acute biliopancreatic complications in patients of pregnancy combined with gallbladder stone was constructed based on the clinical factors of age, primipara, stone type, diameter of stone and gallbladder wall thickness. The area under the curve (AUC) of ROC curve of prediction model was 0.869 (95% confidence interval as 0.813‒0.923), indicating that the prediction model with good predictive ability. Results of Hosmer-Lemeshow test showed a good fit (χ²=5.680, P>0.05), indicating that the prediction model with good calibration. Results of decision curve analysis showed the prediction model with high net clinical benefit. Results of internal validation of the prediction model based on 10-fold cross-validation showed the AUC of ROC curve for the cross-validation sample was 0.833, indicating that the prediction model with good stability.

Conclusions Primigravida, non solitary stone, diameter of stone ≤10 mm, gallbladder wall thickness ≥4 mm are independent risk factors of acute biliopancreatic complications in patients of pregnancy combined with gallbladder stone. The prediction model for acute biliopancreatic complications has good predictive ability.