PREDICTING OUTCOME OF LABOUR INDUCTION USING SONOGRAPHIC CERVICAL LENGTH AND BISHOP SCORE AT UNIVERSITY OF ILORIN TEACHING HOSPITAL

Abstract

Introduction: Failed induction is associated with increased maternal and perinatal morbidity and increased need for operative deliveries. The pre-induction cervical status is the most important determinant of induction success. The traditional Bishop scoring system used in the assessment of cervical status has come under some criticism and newer methods such as

sonographic cervical length measurements are being offered.
Aim: To compare the accuracy of transvaginal sonographic cervical length with Bishop score in predicting outcome of induction of labour using misoprostol.
Methods: This was a prospective study involving 133 women admitted at term and beyond for induction of labour with misoprostol at the Obstetrics Department of University of Ilorin Teaching Hospital between August 2016 and March 2017. Interviewer-administered, pretested proforma were used to collect data. Subjects had sonographic and digital cervical assessment for cervical length and Bishop score respectively. The pain score associated with each method was obtained. Induction outcomes were recorded. Successful induction of labour was defined as vaginal delivery occurring within 24 hours of onset of induction. Statistical analysis was done using SPSS (version 21.0; Chicago, Illinois, USA). Analyses were by t-test, Chi-square, Pearson’s correlation, receiver–operating characteristics (ROC) curves and multivariate logistic regression.
Results: The mean sonographic cervical length was 25.19 ± 10.09 mm (range 0- 44 mm) and was 8mm higher than mean cervical length on digital palpation (p <0.001). Successful induction occurred in 78.9% of patients while 21.1% had failed induction. The mean Bishop score was 3.71 ± 1.98. The mean induction-delivery interval was 15 hours 38 minutes ± 13 hours 06 minutes (range = 4 hours 23 minutes – 94 hours 23 minutes). Bishop score, parity and birth weight independently predicted successful induction of labour. The Bishop score had a linear correlation with the induction-delivery interval (r = -0.267, p = 0.002). The sonographic cervical length did not have a linear correlation with successful induction or the induction-delivery interval. Analyses of the ROC curves showed that the Bishop score predicted successful labour induction better than sonographic cervical length (area under the curve 0.6617 vs 0.6199, p = 0.0046 vs p = 0.059). The optimum cut-off values for predictingsuccessful induction of labour were sonographic cervical length and Bishop score of ≤24mmand >4 respectively. At these cut-offs, Bishop score showed a higher specificity (82.1% vs 78.6%) and positive predictive value (90.9% vs 90.2%) while sonographic cervical length showed a higher sensitivity (52.4% vs 47.6%) and higher negative predictive value (30.6% vs 29.5%). More than 90% of subjects had successful vaginal delivery at these cut-off points. Sonographic cervical length measurement was significantly less painful (mean difference in pain score 3.47, p = <0.001) and was more acceptable (65.3% of subjects).

Conclusion: Bishop scoring system is more accurate than sonographic cervical length in predicting successful labour induction in women at term and beyond undergoing induction of labour with misoprostol. Sonographic assessment is less painful than digital cervical assessment. Bishop score > 4 and sonographic cervical length ≤ 24mm predict success inwomen undergoing induction of labour with misoprostol.

Recommendation: When induction of labour with misoprostol is planned, Bishop score should be used for pre-induction cervical assessment. Further studies in the form of randomized controlled trials are warranted.

PREDICTING OUTCOME OF LABOUR INDUCTION USING SONOGRAPHIC CERVICAL LENGTH AND BISHOP SCORE AT UNIVERSITY OF ILORIN TEACHING HOSPITAL