Changes in obstetrical practices and pregnancy outcomes following the ARRIVE trial.

Background: The ARRIVE trial demonstrated the benefit of induction of labor at 39 weeks gestation. Obstetrics departments across the United States faced the challenge of adapting clinical practice in light of these data while managing logistical constraints.

Objective: To determine if there were changes in obstetrical practices and perinatal outcomes in the United States after the ARRIVE trial publication.

Study Design: This was a population-based retrospective cohort study of low-risk, nulliparous women who initiated prenatal care by 12 weeks gestation with singleton, nonanomalous pregnancies delivering at ≥39 weeks. Data were obtained from the US Natality database. The pre-ARRIVE group were women who delivered between January 1, 2015 and December 31, 2017. The post-ARRIVE group consisted of women who delivered between January 1, 2019 and December 31, 2019. Births that occurred in 2018 were excluded. Practice outcomes were rates of induction of labor, timing of delivery, and cesarean delivery rate. Adverse maternal outcomes were blood transfusion and admission to medical intensive care unit. Adverse neonatal outcomes were need for assisted ventilation (immediate and >6 hours), 5-minute APGAR score <3, neonatal intensive care unit admission, seizures, and surfactant use. Univariate and multivariate analyses were performed. Trends were tested across the time period represented by the pre-ARRIVE group using Cochran-Armitage trend test.

Results: There were 1,966,870 births in the pre-ARRIVE group and 609,322 in the post-ARRIVE group. The groups differed in age, race, body mass index, marital status, infertility treatment, and smoking history (P<.001). After adjusting for these differences, the post-ARRIVE group was more likely to undergo induction (36.1% vs 30.2%; adjusted odds ratio, 1.36 [1.36-1.37]) and deliver by 39+6 weeks of pregnancy (42.8% vs 39.9%; adjusted odds ratio, 1.14 [1.14-1.15]). The post-ARRIVE group had a significantly lower rate of cesarean delivery than the pre-ARRIVE group (27.3 % vs 27.9%; adjusted odds ratio, 0.94 [0.93-0.94]). Patients in the post-ARRIVE group were more likely to receive a blood transfusion (0.4% vs 0.3%; adjusted odds ratio, 1.43 [1.36-1.50]) and be admitted to medical intensive care unit (0.09% vs 0.08%; adjusted odds ratio, 1.20 [1.09-1.33]). Neonates in the post-ARRIVE group were more likely to need assisted ventilation at birth (3.5% vs 2.8%; adjusted odds ratio, 1.28 [1.26-1.30]) and >6 hours (0.6% vs 0.5%; adjusted odds ratio, 1.36 [1.31-1.41]). The neonates in the post-ARRIVE group were more likely to have low 5-minute APGAR scores (0.4% vs 0.3%; adjusted odds ratio, 0.91 [0.86-0.95]). Neonatal intensive care unit admission did not differ between the 2 groups (4.9% vs 4.9%; adjusted odds ratio, 1.01 [0.99-1.03]). There were no differences in neonatal seizures (0.04% vs 0.04%; adjusted odds ratio, 0.97 [0.84-1.13]), and surfactant use (0.08% vs 0.07%; adjusted odds ratio, 1.05 [0.94-1.17]) between the 2 groups.

Conclusion: There were more inductions of labor, more deliveries at 39 weeks' gestation, and fewer cesarean deliveries in the year after the ARRIVE trial publication. The small but statistically significant increase in some adverse maternal and neonatal outcomes should be explored to determine if they are related with concurrent changes in obstetrical practices.