Construction and application of neonatal asphyxia risk prediction model in cesarean section / 中国实用护理杂志

ABSTRACT

Objective:To establish a risk prediction model for neonatal asphyxia in cesarean section and test its application effect.Methods:This was a retrospective study. We retrospectively analyzed the clinical data of 2 244 infants (modeling group) who were delivered by cesarean section in Affiliated Hospital of Weifang Medical University from April 2021 to December 2021. Newborns were divided into asphyxia group ( n=176) and non-asphyxia group ( n=2 068) according to the occurrence of neonatal asphyxia. Logistic regression was used to screen the risk factors of neonatal asphyxia in cesarean section and a line chart model was established to predict the risk. Another 683 neonates were selected as validation group for external validation of the model from January to March in 2022. Results:Five factors including preterm birth, fetal distress, fetal growth restriction, abnormal S/D value of umbilical artery and umbilical cord around the neck were included in the prediction model. The area under ROC curve of the modeling group was 0.902, the Youden index was 0.687, the sensitivity was 0.837, and the specificity was 0.850. Hosmer-lemeshow test showed that χ2=1.79, and P=0.877. In the validation group, the area under ROC curve was 0.823, the Youden index was 0.555, the sensitivity was 0.835, and the specificity was 0.720. It showed that the model had a good fitting effect and identification validity. Conclusions:The risk prediction model has a good clinical application value in the prediction of neonatal asphyxia in cesarean section, and provides reference for obstetricians to take preventive management measures of neonatal asphyxia in time.