Risk factors analysis and prediction model establishment for adverse perinatal outcomes in intrahepatic cholestasis of pregnancy
Li Lei, Zhang Linfeng, Sun Ning, et al
(1. Department of the Third Surgery, Hengshui SecondPeople's Hospital, Hebei, Hengshui 053000, China; 2. Department of Gynecology, Hengshui Second People'sHospital, Hebei, Hengshui 053000, China)
Abstract: 【Abstract】 Objective To analyze the risk factors of adverse perinatal outcomes in intrahepatic cholestasisof pregnancy (ICP) and establish a predictive model. Method A total of 228 patients with ICP admittedto the Hengshui Second People's Hospital from September 2021 to June 2023 were retrospectively selectedas the study subjects, and which were categorised into occurrence group (n=90) and non-occurrence group(n=138) according to whether or not adverse perinatal outcomes occurred. Clinical data [age, birth history, ICPdiagnosis gestational week, gestational diabetes mellitus, delivery model, hypertensive disorder complicatingpregnancy, glycocholic acid, fetal gender, alanine aminotransferase, pre-pregnancy body mass index, chronichepatitis B, total bile acid, direct bilirubin, suppressors of cytokine signaling 3 (SOCS3) protein] werecollected between the two groups. Statistical methods performed by t-test and χ2 test. LASSO regressionwas used to screen the characteristic variables of adverse perinatal outcomes in ICP, Logistic regression wasused to analyze the characteristic variables of adverse perinatal outcomes in ICP, and a nomogram predictionmodel was established, and the model was validated. Result The proportion of diagnostic gestationalage of ICP ≥34 weeks in the occurrence group was lower than that in the non-occurrence group [72.22%(65/90) vs 84.06% (116/138), χ2=4.663, P<0.05]. The incidence of gestational diabetes mellitus, hypertensivedisorder complicating pregnancy, and chronic hepatitis B in the occurrence group were significantly higherthan those in the non-occurrence group [21.11% (19/90) vs 10.87% (15/138), 16.67% (15/90) vs 7.97%(11/138), 10.00% (9/90) vs 1.45% (2/138), respectively; χ2=4.503, 4.077, 6.912, all P<0.05]. The levelsof glycocholic acid, total bile acids, direct bilirubin, and alanine aminotransferase in the occurrencegroup were higher than those in the non-occurrence group [(40.58±11.67) μmol/L vs (32.64±10.35)μmol/L, (49.87±12.34) μmol/L vs (38.66±9.47) μmol/L, (9.05±2.23) μmol/L vs (7.49±2.37) μmol/L,(197.60±45.85) μmol/L vs (158.32±39.13) μmol/L; t values was 5.382, 7.738, 4.972, 6.918, respectively,all P<0.05]. SOCS3 protein in the occurrence group was lower than that in the non-occurrence group[(113.82±36.49) pg/L vs (137.95±40.66) pg/L, t=-4.558, P<0.05). Further Logistic regression analysisof the independent variables screened by LASSO regression showed that diagnostic gestational age of ICP≥34 weeks and high SOCS3 protein were independent protective factors associated with adverse perinataloutcomes in ICP, while gestational diabetes mellitus, chronic hepatitis B, hypertensive disorder complicatingpregnancy, low glycocholic acid, and low total bile acid were independent risk factors associated with adverseperinatal outcomes in ICP (P<0.05). Nomogram was constructed according to the influencing factors of adverseperinatal outcomes in ICP, the consistency and prediction efficiency of the model were good. ConclusionThe influencing factors of adverse perinatal outcomes in ICP include diagnostic gestational age of ICP,gestational diabetes mellitus, hypertensive disorder complicating pregnancy, SOCS3 protein, chronic hepatitisB, glycocholic acid and total bile acid. The nomogram model based on the above factors has good predictive value.
李蕾 张林风 孙宁 王晓倩. 妊娠期肝内胆汁淤积症不良围产儿结局的危险因素分析与预测模型建立[J]. 发育医学电子杂志, 2025, 13(3): 188-194.
Li Lei, Zhang Linfeng, Sun Ning, et al. Risk factors analysis and prediction model establishment for adverse perinatal outcomes in intrahepatic cholestasis of pregnancy. Journal of Developmental Medicine(Electronic Version), 2025, 13(3): 188-194.
2025, Vol. 13 
