Purpose: To analyse the validity of ultrasonography in predicting the outcomes of labour induction, compared with Bishop score (BS), and to design a predictive model including ultrasound and clinical variables.
Methods: In this prospective, observational study of 151 women who underwent induction of labour, an endovaginal ultrasound was performed to determine cervical length (CL), the wedging pattern and the presence of dilatation and funnel. Foetal head-perineum distance (FHPD) was measured by transperineal ultrasound. Vaginal examination was performed to calculate BS and to determine whether cervical ripening was needed. The outcome variable was the method of delivery.
Results: Caesarean section was performed in 30.5 % of cases. CL (26.1 vs. 31.4) and FHPD (44.7 vs. 51.3) were lower in the vaginal delivery group. The area under the curve obtained for FHPD (0.734) was greatest, followed by CL (0.663) and BS (0.678). The proposed model resulted in correct predictions in 82.8 % of cases, with 15 % false positives.
Conclusions: The FHPD and the CL are useful in predicting the result of the induction labour comparable to Bishop score. Using ultrasound scan is significantly better tolerated than vaginal exam. The predictive model selects women who undergo induction having a risk of caesarean section equivalent to spontaneous delivery.
Keywords: Cervical length; Diagnostic validity; Foetal head–perineum distance; Induction of labour; Ultrasound.