Read the full EBM Focus and earn CME credit.
Reference: ARRIVE trial (N Engl J Med 2018; 379:513) (level 2 [mid-level] evidence)
There have been concerns, based primarily on observational data, that induction of labor was associated with worse maternal and fetal outcome. In the ARRIVE multicenter randomized unblinded trial, low risk nulliparous women were randomized at the 38 week visit to either expectant management (3,044 women) or induction of labor (3,062 women) between 39 weeks and 0 days to 39 weeks and 4 days. Induction was with oxytocin as well as cervical ripening if modified Bishop score was
< 5. Women assigned to expectant management were asked to forego delivery until 40 weeks and 5 days and if still pregnant, were induced by at least 42 weeks and 2 days. Baseline characteristics of both groups were similar with the exception that a significantly higher proportion of women in the expectant management group had a history of previous pregnancy loss. A similar number of patients in each group (4.6-6 %) crossed over due to patient or provider preference. Participants, including those women in the expectant management group who underwent induction of labor after 40 weeks and 5 days, were analyzed using intention-to-treat.
The primary outcome studied was a composite of perinatal death or severe complications, which included need for respiratory support, birth trauma, and a number of other largely patient-centered outcomes. There was no significant difference among treatment groups for the primary outcome (relative risk [RR] 0.80, CI 0.64-1.00 in favor of labor induction). However, need for mechanical ventilation and continuous positive airway pressure was lower in the induction of labor group (RR 0.71, CI 0.55-0.93). The rate of cesarean delivery was also lower in the induction group than the expectant management group (RR 0.84, CI 0.76-0.93), with a number needed to treat of 28 to prevent one cesarean delivery. Based on subgroup analyses, women in the induction group also delivered sooner, had less pain, were more satisfied, had a shorter postpartum length of stay (but longer course on labor and delivery), and less risk of hypertensive disorders of pregnancy. Rates of postpartum hemorrhage, chorioamnionitis, third-degree and fourth-degree laceration, and operative vaginal delivery were similar in both groups.
While the study design generally appears solid and the results convincing, there are several caveats worth noting. The inclusion of women delivered beyond 42 weeks in the expectant management group, a gestation at which there are established neonatal risks, may overestimate benefits of elective induction at 39 weeks. Sampling and ascertainment biases are also possible, as participating trial centers have heavy maternal-fetal-medicine influence and patients self-selected to participate. Another concern is that 73% of eligible women declined to participate, and this may be another potential source of selection bias. Also noteworthy are the lower rates of cesarean sections in both groups compared to the national average.
Focus Point: Within the confines of this trial protocol, induction of labor for low risk nulliparous women at 39 weeks appears to reduce the risk of cesarean delivery compared to expectant management with similar neonatal outcomes. While engaging in shared decision-making about elective induction at 39 weeks may seem reasonable based on these data, keep in mind that your pregnant patients may have a bias toward earlier delivery due to the phenomenon of TOBP (tired of being pregnant.)
For more information, see the topic Labor induction in DynaMed Plus. DynaMed users click here.
DynaMed Plus EBM Focus Editorial Team
Alan Ehrlich, MD: Executive Editor at DynaMed Plus and Associate Professor in Family Medicine at the University of Massachusetts Medical School
Monica Zangwill, MD, MPH: Associate Deputy Editor at DynaMed Plus
Katharine DeGeorge, MD, MS: Clinical Editor at DynaMed Plus and Assistant Professor in Family Medicine at University of Virginia School of Medicine
Carina Brown, MD: Fellow at University of Virginia School of Medicine
Sarah Dalrymple, MD: Fellow at University of Virginia School of Medicine