The overall likelihood of vaginal delivery is lower after labor induction than after spontaneous labor, especially when labor induction is attempted in a nulliparous woman with an unfavorable cervix. Institutions should have a clear policy regarding labor induction, including a list of acceptable indications, and should specify the definitions of a favorable cervix, options for cervical ripening in the presence of an unripe cervix, oxytocin infusion protocols, and criteria for the diagnosis of failed induction. Labor induction with an unfavorable cervix should not be undertaken unless delivery is indicated for clear maternal or fetal benefit. Any time induction is undertaken, it should be clear that the goal is vaginal delivery.
Because an unfavorable cervix can negatively effect the labor course and increase the potential for cesarean delivery, this factor should be considered in decision making regarding the method of labor induction. However, the decision for induction should be considered first, and should be separate from the decision about whether or not to employ cervical ripening. Pragmatically, while the potential maternal and fetal risks related to induction with an unfavorable cervix should be incorporated into the overall risk-benefit evaluation when considering medically-indicated labor induction, the decision to proceed with induction should be made independent of the condition of the cervix and based on the specific indication(s). There is no single definition used to differentiate a favorable (“ripe”) from an unfavorable (“unripe”) cervix, whether in research or in clinical practice. In general, the Bishop’s score has most often been used to describe cervical ripeness. A Bishop’s score greater than 8 generally confers the same likelihood of vaginal delivery with induction of labor as that following spontaneous labor, and thus has been considered to indicate a favorable cervix (10
). Conversely, a Bishop’s score of 6 or less has been used to denote an unfavorable cervix in many studies and has been associated with a higher risk of cesarean delivery when labor is induced compared with spontaneous labor. Since the Bishop’s score was originally developed to predict the likelihood of multiparous women at term to enter spontaneous labor, making it less predictive of outcome after labor induction in nulliparas, the affects of maternal parity and gestational age were also considered during the Workshop. Cervical ripening may be considered when there is a medical indication for induction. Since inductions without medical indication should not be done with an unripe cervix, cervical ripening would not be an option. (Box 1
, ). Although cervical ripening agents have generally not been demonstrated to reduce the likelihood of cesarean delivery in prospective interventional trials, their use can effect the duration of labor.
Because the goal of labor induction is vaginal delivery, adequate time to enter into or progress in labor should be allowed provided the mother and baby are stable. The prudent use of labor induction, and the expectation that well-defined criteria be met before cesarean is performed for failure of induction or failure of progress in labor, may actually prevent many unnecessary first cesareans. During this evaluation, it is important to differentiate between “failed induction” and “arrest of labor” in the first stage. The diagnosis of failed induction should be reserved for those women who have failure to generate regular (e.g. every 3 minutes) contractions and cervical change after at least 24 hours of oxytocin administration, with artificial membrane rupture if feasible (after completion of cervical ripening, if performed; Box 1
, ). Studies have shown that over half of the women undergoing labor induction remain in the latent phase for at least 6 hours, and nearly one-fifth remain in the latent phase for 12 hours or longer (11
). In a multi-center study, nearly 40% of the women still in the latent phase after 12 hours of oxytocin and membrane rupture successfully delivered vaginally. These data suggest that induction should not be defined to have failed in the latent phase unless oxytocin has been administered for at least 24 hours, or for 12 hours after membrane rupture. (12
) Numerous approaches to induction and cervical ripening have been published, and no single approach is considered superior to all others. Individual circumstances should be considered for each patient. The algorithm offered in provides a general approach once the decision has been made to proceed with labor induction.
There is much debate as to how long induction should be allowed to continue, and whether it is appropriate to “rest” the patient who does not progress after 12 or more hours of induction but who does not otherwise have a maternal or fetal reason for immediate delivery. In cases where induction is undertaken for specific maternal or fetal conditions that can worsen with time, then stopping the induction is not an appropriate option. Examples of such cases include preeclampsia, fetal growth restriction, diabetes, and ruptured membranes. On the other hand, induction is sometimes undertaken when neither the maternal nor fetal condition is expected to deteriorate rapidly. Examples include induction for social reasons or induction at 41 weeks. Despite this being a common obstetrical dilemma, guidance available from professional organizations do not provide clarity (14
). Published trials allowed cervical ripening over a period ranging from a single dose to several doses over two days. In a trial of the Maternal-Fetal Medicine Units network, the study design specified at least 24 hours from start of oxytocin before declaring a failed induction. All trials have found good outcomes in the induction group despite waiting for at least 24 hours before failed induction was declared (Box 1
, ). It is also important to note that in all induction trials, rupture of membranes was undertaken as soon as feasible and safe. Based on this indirect evidence, it is considered appropriate to temporize before declaring that an induction has failed in women being induced for conditions that are not likely to worsen with time and whose membranes remain intact. An arrest disorder should not be confused with failed induction. The diagnosis of an arrest disorder in women undergoing induction should not be made unless the woman has entered the active phase of labor, requiring that there be documented cervical change preceding the arrest in dilation (Box 1
).Once 6 cm cervical dilation is reached and the active phase is entered, labor progress during induction is similar to the patient in spontaneous labor. However, the duration of the phase before 6 cm dilation is longer in women undergoing induction (17