Construction of a nomogram model for predicting vaginal birth after induction of labor in pregnant women with fetal growth restriction at term

TL;DR

This study created a model to predict vaginal birth success in pregnant women with fetal growth restriction undergoing labor induction.

Contribution

A novel nomogram model was developed and validated for predicting vaginal delivery outcomes in FGR pregnancies.

Findings

The nomogram model included five factors: maternal age, multiparity, oligohydramnios, Bishop score, and Cook’s double balloon.

The model showed good predictive accuracy with an area under the curve of 0.811 in training and 0.760 in validation.

The model demonstrated clinical utility with high net benefit across a wide range of threshold probabilities.

Abstract

The study aimed to develop and validate a predictive nomogram model of vaginal delivery in women with fetal growth restriction (FGR) who undergo labor induction.

A retrospective cohort study was conducted at Fujian Maternity and Child Health Hospital, involving 507 singleton pregnancies complicated with fetal growth restriction (FGR) between October 2017 and December 2022. These pregnancies underwent labor induction with dinoprostone or Cook’s double balloon. The cohort was randomly divided into two groups: 75% of pregnancies (n = 380) were utilized to identify independent factors associated with vaginal delivery using multi-logistic regression and develop a predictive nomogram model, while 25% (n = 127) were allocated for internal validation of the model.

A predictive nomogram model was constructed with five validated factors including maternal age, multiparity, oligohydramnios (borderline significant), Bishop score after cervical ripening and Cook’s double balloon. The trained and validated area under the curve were 0.811 (95% confident interval 0.757–0.865) and 0.760 (95% confident interval 0.669–0.860), respectively. The Hosmer-Lemeshow test indicated no statistically significant difference between the predicted and observed outcomes (P > 0.050). The clinical decision curve demonstrated that both the model and the validation groups achieved the greatest net benefit at threshold probability values ranging from 0.20 to 0.95 and exceeding 0.40, respectively.

The nomogram model could be utilized to inform patients with FGR about their success of vaginal delivery during labor induction. Moreover, this model established a foundation for clinical intervention and the development of personalized medical treatment strategies.

Figures