关键词: 妊娠期高血压疾病, 妊娠期糖尿病, 初产妇, 不良妊娠结局, 风险预测模型, 列线图

Abstract: Objective  To screen the relevant risk factors for adverse pregnancy outcomes in pregnant women with hypertensive disorders of pregnancy（HDP） combined with gestational diabetes mellitus(GDM) and to construct risk prediction model and evaluate its predictive performance. Methods  Through the electronic medical record system, clinical data of 1 061 singleton primiparas were collected and analyzed retrospectively, who were diagnosed with HDP complicated by GDM in Beijing Obstetrics and Gynecology Hospital, Capital Medical University, from January 2022 to December 2024. A random number method was used to randomly divide the patients into a training set (742 cases) and a validation set (319 cases) according to a 7∶3 ratio. Based the pregnancy outcome the patients were divided into the adverse pregnancy outcome group (APO group) and the non-adverse pregnancy outcome group (non-APO group). The clinical data of two groups of pregnant women were compared to each other, including age, pre-pregnancy weight, pre-pregnancy body mass index (BMI), weight gain during pregnancy, pre-delivery weight, pre-delivery BMI, diabetes family history, hypertension family history, hypothyroidism, whether assisted reproductive technology was used for conception, whether pre-eclampsia (PE) or chronic hypertension with superimposed pre-eclampsia (CPE) occurred. Univariate Logistic regression analysis was used to screen variables, and a prediction model was constructed by multivariate Logistic regression analysis and a nomogram was drawn. The area under the receiver operating characteristic curve (AUC) was used to evaluate the discrimination of the model. The calibration curve and Hosmer-Lemeshow goodness—fit test were used to verify and evaluate the calibration of the model. The decision curve analysis (DCA) was used to evaluate the clinical effectiveness of the prediction model. Results  A total of 1 061 primiparas with HDP  complicated by GDM were enrolled, and the incidence of adverse pregnancy outcomes was 47.6% (505/1061). Six predictive variables were screened out by univariate Logistic regression analysis: age, pre-pregnancy BMI, hypothyroidism, diabetes family history, hypertension family history, and PE/CPE. The multivariate Logistic regression analysis indicated that they were all risk factors for the occurrence of adverse pregnancy outcomes in pregnant women with HDP complicated by GDM (P< 0.05). Draw a column line chart based on these 6 risk factors and construct a risk prediction model. The AUC of the training set patients with adverse pregnancy outcomes were 0.829 （95%CI:0.799-0.859) while that of the validation set patients with adverse pregnancy outcomes were 0.839 (95%CI:0.796-0.883), without significantly difference (P=0.477 6). The Hosmer-Lemeshow goodness-of-fit test showed a good fit (P= 0.323 8 for the training set and P= 0.702 9 for the validation set), and there was a significant agreement between the predicted value and the actual value. DCA indicated that the prediction model demonstrated clinical utility when the threshold probability exceeds 0.05. Conclusion  The risk prediction model can effectively identify high-risk groups with adverse pregnancy outcomes in primiparas with HDP complicated by GDM, which can provide reference for early intervention.

Key words: hypertensive disorders of pregnancy, gestational diabetes mellitus, primiparous women, adverse pregnancy outcomes, a risk prediction model, nomogram