DESIGN: A comparative clinical trial.

SETTING: A tertiary hospital in Selangor, Malaysia.

PARTICIPANTS: 294 women planned for elective induction of labour for various indications were included. All women had transvaginal ultrasound to assess the cervical length and digital vaginal examination to assess the Bishop cervical scoring by separate investigators before induction of labour.

PRIMARY OUTCOME MEASURE: To evaluate the association of the cervical length by transvaginal ultrasound scan and Bishop score in predicting successful induction of labour.

SECONDARY OUTCOME MEASURE: Variables associated with successful induction of labour and patients' tolerability and acceptance for transvaginal ultrasound scan of cervix.

RESULTS: There was no statistically significant difference among the vaginal and Caesarean delivery groups in terms of mean maternal age, height, weight, body mass index, ethnicity and gestational age at induction. Vaginal delivery occurred in 207 women (70.4%) and 87 women (29.6%) delivered via Caesarean section. There was a high degree of correlation between the cervical length and Bishop score (r-value 0.745; p <0.001). Sonographic assessment of cervical length demonstrated a comparable accuracy in comparison to Bishop score. Analysis using ROC curves noted an optimal cut-off value of ≤27mm for cervical length and Bishop score of ≥ 4, with a sensitivity of 69.1% vs 67%, specificity 60.9% vs 55%, and area under the curves (AUCs) of 0.672 and 0.643 respectively (p <0.001). Multivariate logistic regression analysis demonstrated that parity (OR 2.70), cervical length (OR 0.925), Bishop score (OR 1.272) and presence of funnelling (OR 3.292) were highly significant as independent predictors of success labour induction. Women also expressed significantly less discomfort with transvaginal ultrasound compared with digital vaginal examination.