Analysis of maternal and fetal outcomes and establishment of prediction model of vaginal delivery in pregnant women with pre-eclampsia complicated with fetal growth restriction

L. Lin, Y.-N. Guo, X. Xu, L.-P. Huang, Q.-P. Yang, J.-Y. Yan

College of Clinical Medicine for Obstetrics and Gynecology and Pediatrics, Fujian Medical University, Department of Obstetrics, Fujian Maternity and Child Health Hospital, Fuzhou, Fujian, China. yanjy2019@fjmu.edu.cn

---

• Obstetrics

OBJECTIVE: This study aimed to analyze the maternal and fetal outcomes of pregnant women with pre-eclampsia (PE), complicated with fetal growth restriction (FGR), and establish a prediction model of vaginal delivery to guide the selection of the delivery mode.

PATIENTS AND METHODS: The study included 208 pregnant women with PE complicated with FGR. Of them, 49 patients were in the vaginal delivery group, and 159 patients were in the cesarean section group. The relevant maternal and fetal outcomes were analyzed. Patients were randomly divided into the training sample group and the test group with a ratio of 2:1. The three-layer neural network was used to select 24 maternal and infant outcome factors as the input nodes of the neural network to build a vaginal delivery prediction model.

RESULTS: Results showed that the gestational age, the highest systolic and diastolic blood pressure, body weight, body length, and placental weight of the newborns in the vaginal delivery group were significantly higher than those in the cesarean section group. Incidence of preterm birth, amniotic fluid grade III, oligohydramnios, and severe small-for-gestational-age (sSGA) neonates were significantly lower in the vaginal delivery group compared to the cesarean section group (p < 0.05). A three-layer neural network delivery prediction model was constructed, and the accuracy rate of fitting with test samples was 91.80%.

CONCLUSIONS: There is no significant difference in the incidence of maternal and fetal complications in PE complicated with FGR in different delivery methods. The three-layer neural network prediction model has good prediction ability for vaginal delivery of PE complicated with FGR and may be applied in clinical practice.

Free PDF Download