双胎妊娠发生胎膜早破的危险因素分析及风险预测模型的构建OA

Objectives:To analyze the risk factors for premature rupture of membranes(PROM)in twin pregnancies and establish a quantifiable risk prediction model based on these risk factors. Methods:A retrospective analysis was conducted on the data of 965 pregnant women with twin pregnancies who delivered at Dongguan People's Hospital from January 2018 to December 2024.They were divided into the non-occurrence group(n=789)and the occurrence group(n=176)based on whether PROM occurred.The general data and gestational complications of the two groups were collected and compared.The univariate and multivariate logistic regression analyses were used to explore the risk factors for PROM in twin pregnancies.A nomogram was used to construct a risk prediction model and the model validation was conducted.The pregnancy outcomes and neonatal health conditions of the two groups were compared,such as the rate of preterm birth,transfer to the neonatal intensive care unit(NICU),neonatal pneumonia,neonatal respiratory distress syndrome(NRDS),and the incidence of antibiotic application. Results:The univariate and multivariate logistic analyses found that gestational diabetes[OR=2.103,95％CI(1.105,3.732)],gestational hypertension[OR=2.278,95％CI(1.186,4.376)],Group B streptococcus(GBS)infection[OR=3.522,95％CI(1.416,8.761)],infection during pregnancy[OR=1.786,95％CI(1.065,3.238)],and prenatal fever[OR=4.443,95％CI(1.081,18.260)]were all risk factors for PROM in pregnant women with twin pregnancies(P＜0.05).A nomogram prediction model was constructed based on five independent risk factors.The results of the decision curve(DCA)indicated the clinical effectiveness of this prediction model was significant when the threshold was more than 0.19.The receiver operating characteristic(ROC)curve indicated that the area under the ROC curve(AUC)of the training set and the validation one were 0.884 and 0.827,respectively.The calibration curve indicated that the consistency index of the training set and the validation set and the Hosmer-Lemeshow goodness of fit were 0.830(x2=5.439,P=0.066)and 0.815(x2=5.930,P=0.052),respectively.The preterm birth rate in the occurrence group was higher than that in the non-occurrence group(P＜0.05).The incidence of transfer to the NICU,neonatal pneumonia and NRDS as well as the application rate of antibiotics in the occurrence group were higher than those in the non-occurrence group(P＜0.05). Conclusions:Gestational diabetes,gestational hypertension,GBS infection,prenatal fever,and infection during pregnancy are all risk factors for PROM in twin pregnancies.A risk prediction model based on these five indicators can accurately predict the risk of PROM occurrence,assisting physicians in screening high-risk mothers with twin pregnancies.