Pre-induction sonographic measurement of cervical length in prolonged pregnancy: the effect of parity in the prediction of the need for Cesarean section.

OBJECTIVE To examine the effect of parity on the relationship between pre-induction cervical length and the risk of Cesarean section in women undergoing induction of labor for prolonged pregnancy. METHODS In 382 singleton pregnancies, induction of labor was carried out at 41 + 3 to 42 + 1 weeks of gestation. The cervical length was measured by transvaginal sonography before induction. In 71 of these cases a Cesarean section was required for either fetal distress or failure to progress and 311 women delivered vaginally. The effect of parity and pre-induction cervical length on the risk of Cesarean section was examined. RESULTS In nulliparous women the incidence of Cesarean section was significantly higher than in multiparous women (28.1% vs. 8.9%). In both nulliparae and multiparae there was a significant association between pre-induction cervical length and the rate of Cesarean section. Logistic regression analysis demonstrated that cervical length and parity provided significant independent prediction of all Cesarean sections and Cesarean sections for failure to progress. The odds of Cesarean section increased by about 10% with each increase of 1 mm in cervical length, over the mean cervical length of 20 mm for nulliparae and 18 mm for multiparae, and the odds was about 75% lower in multiparae, compared to nulliparae with the same cervical length. Receiver-operating characteristics curves (ROC) demonstrated that cervical length was better than the Bishop score in the prediction of all Cesarean sections (area under ROC = 0.72 vs. 0.68) and Cesarean sections for failure to progress (area under ROC = 0.76 vs. 0.69). CONCLUSION In women undergoing induction of labor for prolonged pregnancy, cervical length and parity provide significant independent prediction of the likelihood of Cesarean section.