Labor Induction in Low-risk Nulliparous Women at 39 Weeks of Gestation to Reduce Cesarean: A Randomized Trial of Induction Versus Expectant Management in France (FRENCH-ARRIVE)

The nadir of the neonatal risks seems to be at 39 weeks of gestation and there is an augmentation of maternal complications after 39 weeks. Nonetheless, planned induction of labor at 39 weeks was not encouraged in common practice due to a suspected higher risk of frequency cesarean delivery and other adverse maternal outcomes, especially among nulliparous women with unfavorable cervix, compared with expectant management. This suspected increased maternal morbidity especially the cesarean rate associated with the induction of labor was based on observational studies which suffer from methodological limitations or based on underpowered small randomized clinical trials.

A recent multicenter, randomized, controlled, unmasked trial conducted in United States of America in 2014-2017 (ARRIVE trial), provides new results with a high level of evidence. This trial conducted among 6,106 low-risk nulliparous women who were randomized, 3062 assigned to labor induction at 39 weeks 0 day to 39 weeks 4 days and 3,044 assigned to expectant management, demonstrates that induction of labor at 39 weeks did result in a trend but not significant lower frequency of the primary outcome - a composite adverse perinatal outcome - (relative risk [RR] 0.80, 95% confidence interval [CI] 0.64-1.00), but did result in a significantly lower frequency of cesarean delivery (RR 0.84, 95% CI 0.76-0.93).

Although the cesarean delivery rate was a secondary outcome of the study and the absence of any comparison for women's characteristics of those who were eligible and declined to participate and those who were randomized, the recommendations of the Society of Maternal-Fetal Medicine (SMFM), American College of Obstetricians and Gynecologists (ACOG) and numerous authors have already considered that induction of labor ≥ 39 weeks should be proposed in low-risk nulliparous women. In other words, on the basis mainly on the results of a secondary outcome from only one single randomized controlled trial, they have proposed to modify the management of the pregnant women who reach 39 weeks gestation (i.e. the large majority of the pregnant women) while until now induction of labor for those women was proposed only if abnormal medical condition occurred or beyond 41 weeks of gestation. Nevertheless, the findings of the ARRIVE trial are so important because the reduction of the cesarean rate is a worldwide goal and the elective induction of labor ≥ 39 weeks seems to be today the most effective way to achieve it. However, French context and obstetrical practices are different than USA, with a lower rate of cesarean. Women's characteristics are also different (i.e. lower body mass index in France). Consequently, the external validity of the ARRIVE trial needs to be confirmed. Moreover, in the ARRIVE trial, the labor management has to respect American recent guidelines with a longer duration of the latent phase and administration of oxytocin if needed for at least 12 hours after membrane rupture before deeming the induction a failure. A generalization of labor induction at 39 weeks without a strictly respect of the eligibility criteria or labor management could result to an increase of unexpected adverse maternal or neonatal outcomes.

Thus, the expected benefits of labor induction at 39 weeks have to be confirmed in other context, in particular in French settings with a lower cesarean rate than in USA by replicating the ARRIVE study conducting another randomized controlled trial for which the primary outcome will be the cesarean section rate.

This replication is crucial before leading to a such important change in daily practice that would be considering routine induction of labor for low-risk nulliparous women at 39 weeks of gestation and consequently modifying the organization of all maternity wards in order to achieve this policy.

In addition to the replication of the ARRIVE trial in a French context, a prospective observational cohort with 4,200 women is associated with this trial to identify a potential participant selection bias.