Antepartum ultrasound prediction of failed vacuum-assisted operative delivery: a prospective cohort study

Abstract

Background Failed vacuum-assisted delivery (VD) is associated with increased risk of maternal perineal trauma and neonatal morbidity. Knowledge of the risk factors related to failed VD is essential in the clinical decision-making. Objective To elucidate the strength of association and the predictive accuracy of different ante-partum ultrasound parameters in predicting the risk of failed VD prior to the onset of Labor and to test the diagnostic performance of a multiparametric model including pregnancy and Labor characteristics, ante and intra-partum ultrasound in anticipating failed VD. Study design Prospective study of consecutive singleton pregnancies complicated by VD undergoing a dedicated ultrasound assessment at 36–38 weeks of gestation. Head circumference (HC), estimated fetal weight (EFW) and subpubic angle and (SPA) were recorded before the onset of Labor. At the time of the VD, occiput position, head perineum distance (HPD) and angle of progression (AOP) were also recorded. Multivariate logistic regression and area under the curve (AUC) analyses were used to explore the strength of association and test the diagnostic accuracy of different maternal, Labor and ultrasound characteristics in predicting g failed VD. Results Four hundred eight pregnancies with successful and 26 with failed VD were included in the analysis. Fetuses experiencing failed VD had a larger HC (1.21 versus 1.07 MoM; p = .0001), a higher EFW z-value (0.56 versus 0.33 z values; p = .002) and a narrower SPA (114 versus 122 p = .0001) compared to those having a successful VD. At multivariable logistic regression analysis, maternal height (aOR 0.89 95% CI 0.76–0.98), nulliparity (aOR: 1.14 95% CI 1.06–1.36), HC MoM (aOR: 1.24 95% CI 1.13–1.55) and SPA angle (aOR: 0.82 95% CI 0.67–0.95), but not EFW (p = .08) were independently associated with failed VD. When intrapartum ultrasound variables were added to the multivariate model, fetal occipital position (aOR: 1.45 95th CI 1.11–1.99) and HPD (aOR: 0.77 95th CI 0.44–0.96) were independently associated with failed VD. A multiparametric model integrating pregnancy and Labor characteristics and ante-partum ultrasound variables had an AUC of 0.837 (95% CI 0.797–0.876) for the prediction of failed VE. The addition of intra-partum ultrasound variables to the prediction model, improved the accuracy for failed VD provided by maternal and antepartum ultrasound characteristics with an AUC of 0.913 (0.888–0.937). Conclusion Antepartum prediction of failed VD is feasible. HC, SPA but not EFW are independently associated and predictive of failed VD. Adding these variables to a multiparametric model including maternal and intrapartum ultrasound parameters improves the diagnostic accuracy for failed VD.