Transvaginal ultrasound measurement of cervical length in the supine and upright positions versus Bishop score in predicting successful induction of labor at term

Objectives To examine the predictive value of cervical length as measured by transvaginal sonography (TVS) in supine and upright maternal positions for the mode of delivery and induction-to-delivery interval after induction of labor at term, and to compare these measurements with the Bishop score and its predictive value. Methods TVS for cervical length measurement in the supine and upright positions and digital examination of the cervix were performed in 68 nulliparous and 34 parous women before induction of labor at term. In assessing the predictive value of the Bishop score and TVS parameters for a vaginal delivery after labor induction only nulliparous women were included in the analysis, since all the parous women delivered vaginally. Both nulliparous and parous women were included in the analysis ofthe induction-to-delivery interval. The method of labor induction, oxytocin or prostaglandin, was determined on the basis ofthe pre-induction Bishop score. Results Logistic regression analysis showed in nulli-parous women that only the cervical length measured in the upright position was a significant predictor of the need for Cesarean section (odds ratio 1.14; 95% CI, 1.02-1.27). The areas under the receiver-operating characteristics curve in predicting the need for Cesarean section because of failure to progress were higher for the cervical length, both in supine and upright positions, than for the Bishop score (0.66, 0.68 and 0.46, respectively). Only the Bishop score correlated significantly with the induction-to-delivery interval in both nulliparous and parous women. However, this may have been due to a selection bias, as no significant correlation with Bishop score was found when the oxytocin and prostaglandin induction-to-delivery intervals were analyzed separately. Conclusion Our results suggest that maternal postural change might improve the accuracy of sonographically-measured cervical length for predicting a vaginal delivery after induction of labor at term. However, our results need to be confirmed in a larger and more homogeneous population.