For Healthy Pregnancies, Inducing Labor at 39 Weeks Is Safe

Babies are “due” after 40 weeks of gestation, but evidence suggests that infant mortality and complications are lowest for those delivered at 39 weeks, when a fetus is considered full term. Some obstetricians have recommended inducing labor at 39 weeks to reduce the risk of complications.

But the practice isn’t routine. Physicians worried that elective induction at 39 weeks might increase the rate of C-sections, and no randomized studies had determined whether induction might be safer than the usual care—letting nature take its course until around 41 weeks and inducing labor after that.

A new clinical trial, recently published in the New England Journal of Medicine, provides some answers. The CUIMC Newsroom spoke with Annette Perez-Delboy, MD, a maternal-fetal medicine expert at Columbia University Vagelos College of Physicians and Surgeons and NewYork-Presbyterian/Columbia University Irving Medical Center and a member of the trial’s research team.

Q: What was the purpose of this clinical trial, and what were the results?

This trial was designed to compare maternal and perinatal outcomes in women who were induced at 39 weeks’ gestation and those who either went into labor spontaneously or were induced at 41 to 42 weeks if spontaneous labor did not occur.

Based on previous retrospective studies, our thought was that delivery after 40 weeks may be associated with a higher risk of complications, such as maternal preeclampsia and low Apgar scores for neonates. To avoid complications, some doctors have been interested in inducing labor at 39 weeks, when the risk of complications is lowest. But we didn’t have enough evidence to determine if elective induction in low-risk first-time mothers could improve outcomes, so we conducted a randomized clinical trial to find out. We also questioned the older literature that induction before 40 weeks might increase the number of deliveries by cesarean section, so we looked at this outcome as well.

We found that maternal and perinatal outcomes in both groups were similar: The group that delivered later had slightly more neonatal deaths and complications, although the difference wasn’t statistically significant. However, the group that was induced at 39 weeks had significantly fewer C-sections and lower rates of preeclampsia, which is one of the leading causes of maternal morbidity.

Q: Will these findings change clinical practice?

These findings will give women and their providers options that once were considered taboo. They show that elective induction at 39 weeks is now a valid option. This is particularly important for women known to be at risk for preeclampsia.

Of course, while the study showed that induction at 39 weeks is safe, some women may still prefer to let nature take its course and allow labor to begin naturally. It’s a personal choice, but women have the option to decide.

Q: Were the findings surprising, and, if so, why?

The findings were not what we had predicted. We had expected neonatal outcomes to be better in the group induced at 39 weeks. While the numbers pointed in the direction of benefit, the findings were not statistically significant. We were also surprised at the marked benefit to maternal health by reducing the rate of preeclampsia and other pregnancy-associated hypertensive disorders.

We were also glad to validate our secondary hypothesis that induction at 39 weeks was associated with fewer C-sections. Many of the women in this group had a low Bishop’s score—a rough measure of a woman’s readiness to deliver vaginally. In general, a low Bishop’s score is associated with a higher risk of delivery by C-section, so doctors tend to avoid induction in women with a low score. But having a low score didn’t seem to affect the likelihood of delivering vaginally. This adds a new piece of information to the puzzle about whether and when to induce.

Q: Are your results generalizable to all women with healthy pregnancies? Or are there certain women for whom the results don’t apply?

Our study was the largest randomized trial to study this question. It included 6,100 healthy women from 41 urban and community hospitals, which have varying protocols for labor induction and management. In addition, there were no significant differences in results according to the mom’s race or ethnicity, age, or body mass index. However, the results are only generalizable to the type of women who were eligible for the study. Importantly, they had to be low-risk, with no other medical or obstetric issues that would impact delivery, and they had to be delivering their first baby.

Additionally, we only enrolled women with a reliably estimated due date. We aren't always able to obtain an accurate estimate, so we can’t be sure if the results apply when we don’t have good dating.

Q: What questions remain unanswered?

There are several unanswered questions, which will be addressed in secondary analyses of this study. One question is the cost associated with a 39-week induction. Labors that are induced are generally longer than spontaneous labors, so induction may be associated with an increase in cost. But because the induced group had fewer cesareans, there may also be cost savings. Other questions concern the use of Pitocin (the drug to induce labor), length of labor, and resource utilization. It may be challenging for some centers to accommodate a potential increase in inductions, though fewer women delivering after going into labor spontaneously may offset that.


Annette Perez-Delboy, MD, MBA, is an associate professor of obstetrics & gynecology at Columbia University Vagelos College of Physicians and Surgeons.

The study, “Labor Induction versus Expectant Management in Low-Risk Nulliparous Women,” was published Aug. 9 in the New England Journal of Medicine.

Additional study investigators from the Department of Obstetrics & Gynecology at VP&S included Ronald Wapner, MD, professor of obstetrics & gynecology, and Cynthia Gyamfi-Bannerman, MD, the Ellen Jacobson Levine and Eugene Jacobson Professor of Women’s Health. Other authors are included in the paper.

The work was supported by the National Institutes of Health. Conflicts of interest are included in the paper.