Factors predicting failure of labour induction.

Our objective was to identify risk factors for failed labour induction. From January 2001 to December 2005, we conducted a retrospective case-control study in a Nigerian University Hospital. Cases were women who failed to deliver vaginally after labour induction, and therefore had caesarean section. Controls were patients who were similarly induced and achieved vaginal delivery. Univariate followed by Logistic regression analysis were performed. Failed induction occurred in 37.6%. Variables significantly and independently associated with failed induction were cervical effacement < 70% (adjusted odds ratio [OR] 5.12; 95% confidence interval [CI] 2.65-9.90), Bishop's score < 6 (OR 3.47; CI 1.75 - 6.85), nulliparity (OR 3.91; CI 1.92 - 7.99). Prolonged pregnancy independently reduced failure rate (OR 0.44: CI 0.24 - 0.81). These variables can help determine patients that will require early recourse to caesarean delivery in order to avoid prolonged induction-delivery interval.