EP23.17: A Multivariable Decision Model Based on Maternal and Fetal Sonographic Variables for the Prediction of a Spontaneous Vaginal Delivery in the Term Nulliparous Parturient Before the Onset of Labour

To create a multivariable decision model based on maternal and fetal sonographic variables for the prediction of a spontaneous vaginal delivery in the term nulliparous parturient before the onset of labour. Nulliparous women carrying a singleton fetus in vertex presentation at 37.0-42.0 weeks of gestation were recruited at the post date clinic. Sonographic measurements included estimated fetal weight and pelvimetric measures: pubic arch angle and angle of progression. Descriptive and comparative analysis was performed between the two outcome groups: spontaneous vaginal delivery (SVD) unplanned operative delivery (UOD) which included vacuum assisted, forceps assisted and Caesarean deliveries. Logistic regression was used in univariate and multivariate models of risk for unplanned operative delivery UOD. Out of 234 patients 160 delivered spontaneously, and 74 patients had an unplanned operative delivery. Analysis of maximum likelihood estimates revealed a multivariate model for prediction of UOD including the parameters of maternal age, height, sonographic PAA, AOP and EFW with an area under the curve of 0.707. A multivariate decision model for prediction of an UOD based on maternal and fetal sonographic parameters was created. In this study we present a combination between measurements performed before the onset of labour that represent the pelvic configuration and the maternal-fetal interface performed on real time 2D ultrasound images. This data may aid the obstetrician in the counselling process prior to delivery and may have significance in rural or indigent areas at which appropriately timed transfer to a district hospital have great impact on maternal and neonatal morbidity. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.