Construction Of Nomogram To Predict Spontaneous Preterm Birth In Twin Pregnancies

Objective: To study the risk factors for the occurrence of spontaneous preterm birth in twin pregnancies in our hospital,to develop a prediction model and to validate it,to provide some value for the prediction of spontaneous preterm birth in twin pregnancies.Methods: According to the inclusion and exclusion criteria,526 twin pregnancies who underwent outpatient checkups and delivered in our hospital from January 2020 to December 2022 were collected as the study subjects and divided into the following groups according to their gestational weeks of delivery: 28～36+6 weeks(preterm group,123 cases)and ≥37 weeks(control group,403 cases).The general data of these 2 groups,such as age,pre-pregnancy BMI,gestational weight gain,the number of pregnancies and deliveries,obstetric examinations and pregnancy comorbidities,were screened for possible risk factors for spontaneous preterm birth in twin pregnancies by univariate analysis and independent risk factors by multi-factor logistic regression analysis,and column plots were drawn.The predictive ability of the model was then assessed by the area under the ROC curve,the calibration was evaluated using the graphical calibration method and the Hosmer-Lemeshow goodness-of-fit test,and finally the clinical utility of the model was evaluated by the DCA curve.Results:1.The factors influencing the occurrence of spontaneous preterm delivery in twin pregnancies were analyzed successively by univariate and multivariate analyses,which showed that: previous history of preterm delivery(OR = 4.684,95% CI: 1.649 to13.650,P < 0.05),monochorionicity(OR = 2.312,95% CI: 1.347 to 3.966,P < 0.05),,and cervical funneling(OR = 2.423,95% CI: 1.065 to 5.512,P < 0.05)were independent risk factor for spontaneous preterm birth,pre-pregnancy BMI(OR = 0.876,95% CI: 0.808 to 0.950,P < 0.05)and midtrimester cervical length(OR = 0.882,95%CI: 0.846 to 0.920,P < 0.05)were protective factors for spontaneous preterm birth.2.The nomogram developed in this study to predict the risk of spontaneous preterm birth in twin pregnancies had good internal validation results and good discrimination;the calibration curve and consistency test suggested that the predicted probability converged with the actual probability of occurrence;and the decision curve indicated a large net clinical benefit of the column line graph model.Conclusions: Previous history of preterm labour,monochorionicity,mid-pregnancy cervical length were identified as independent risk factors for spontaneous preterm birth in twin pregnancies,,pre-pregnancy BMI and midtrimester cervical length were protective factors for spontaneous preterm birth.The column line graph model developed by combining the above factors is predictive of the occurrence of spontaneous preterm birth in twin pregnancies with regular maternity visits.