Since 2018, unfortunate misinterpretation of the Arrive Trial has led to the routine protocol of induction of labor at 40 weeks or up to 40+5 weeks gestation, disregarding the 2 week margin of error in EDDs (estimated due dates) and all other evidence against induction of a healthy pregnancy. The results have been so profound to me personally in my work as a Childbirth Educator that at moments I have considered quitting. Of the last 59 inductions of my students, 22 have ended in c-sections (35%), along with 8 babies in the NICU (14%), 5 hemorrhages, 3 blood transfusions, vomiting and fevers of moms, and labors with fetal distress. This is not normal. It is not normal for 59 births and it is not normal in my 29 years of teaching. Watching women leave class feeling strong, confident and informed and then seeing their births unfold as a result of the unnecessary induction has me anguished and despairing. The latest was a woman induced on her due date for having a big baby. The baby weighed 6 lb. 13 oz. and likely was deprived of a week of two of further growth and brain development.
Finally we have a new study from Australia, published January 2024, which sought to evaluate whether Induction of Labor (IOL) is associated with cesarean birth (CB) and infant mortality in uncomplicated first births at term rather than waiting for labor to begin on its own. I wish they would find a catchy name (like the “ARRIVE” trial did) and I pray that doctors will stop inducing healthy mothers and babies and causing complications and cesareans.
Induction of labor and cesarean birth in lower-risk nulliparous women at term: A retrospective cohort study
Sarah E. Butler MD, Euan M. Wallace MD, Andrew Bisits MD, Roshan J. Selvaratnam PhD, Mary-Ann Davey DrPH
First published: 03 January 2024
https://doi.org/10.1111/birt.12806
Abstract
Objective
To evaluate whether induction of labor (IOL) is associated with cesarean birth (CB) and perinatal mortality in uncomplicated first births at term compared with expectant management outside the confines of a randomized controlled trial.
Introduction
Rising rates of cesarean birth in high-income countries continue to be a widespread concern.
Cesarean birth exposes women to inherent surgical risks and complications; and is associated with higher rates of postpartum hemorrhage,1 slower recovery from birth,2 and an increased risk of placental complications in subsequent pregnancies.3 Several factors have been implicated in these trends, including a reduction in vaginal births after cesarean birth (CB),4, 5 an increase in the rates of obesity,6 maternal request for CB,7, 8 and women undergoing induction of labor (IOL).9, 10 In Australia, more than one in three women had a CB in 2020,8 which is among the highest rates worldwide.11 Safely reducing CB in nulliparous women is of particular importance as it also reduces the number of repeat CB in subsequent pregnancies.
A common view in maternity care proposes that inducing labor increases the risk of CB. This has been challenged by several recent randomized controlled trials including the influential ARRIVE trial.12–14 Importantly, in the years following the publication of ARRIVE, some studies have shown an increase in IOL among nulliparous women, along with an increase in neonatal and maternal morbidity.15, 16
While IOL performed for medical indications has been shown to be beneficial,17 IOL without medical indication has also been associated with increased rates of CB.9, 18, 19 With the proportion of women undergoing IOL rising worldwide,20–22 there is a need for a comprehensive understanding of the potential risks and benefits. Specifically, we need guidance for lower-risk pregnancies in the contemporary, local setting due to variation in maternal characteristics and preferences, and management of induced labor in different care settings.
We aimed to examine the association between IOL at 37 and 41 weeks of gestation and CB in lower-risk first births in a contemporary population-based cohort study in which the indication for IOL was known.
Methods
Population-based retrospective cohort study of all births in Victoria, Australia, from 2010 to 2018 (n = 640,191). Preliminary analysis compared IOL at 37 weeks with expectant management at that gestational age and beyond for uncomplicated pregnancies. Similar comparisons were made for IOL at 38, 39, 40, and 41 weeks of gestation and expectant management. The primary analysis repeated these comparisons, limiting the population to nulliparous women with uncomplicated pregnancies and excluding those with a medical indication for IOL. We compared perinatal mortality between groups using Chi-square tests and multivariable logistic regression for all other comparisons. Adjusted odds ratios and 99% confidence intervals were reported. p < 0.01 denoted statistical significance.
Results
Among nulliparous, uncomplicated pregnancies at ≥37 weeks of gestation in Victoria, IOL increased from 24.6% in 2010 to 30.0% in 2018 (p < 0.001). In contrast to the preliminary analysis, the primary analysis showed that IOL in lower-risk nulliparous women was associated with increased odds of CB when performed at 38 (aOR 1.23(1.13–1.32)), 39 (aOR 1.31(1.23–1.40)), 40 (aOR 1.42(1.35–1.50)), and 41 weeks of gestation (aOR 1.43(1.35–1.51)). Perinatal mortality was rare in both groups and non-significantly lower in the induced group at most gestations.
Discussion
For lower-risk nulliparous women, the odds of CB increased with IOL from 38 weeks of gestation, along with decreased odds of perinatal mortality at 41 weeks only.
Other excerpts from the analysis, which is broken down in great detail in the paper:
4.1 Main findings
This study found that unplanned CB was more likely among lower-risk, nulliparous women who underwent IOL between 38 and 41 completed weeks of gestation compared with expectant management. This relationship persisted after adjustment for important confounders. Higher odds of CB in the induced groups were also observed when comparing IOL at a specified gestational day (266, 273, and 280 days) with expectant management beyond that day. The increased odds of CB following IOL was seen in both public and private maternity care.
Notably, this is among the first observational studies to examine the relationship between non-medically indicated IOL and CB using a pregnancies-at-risk approach comparing IOL on a specified gestational day with expectant management from that day onwards. The advantage of this is that it more closely mimics clinical practice, where the decision to be made is between “inducing today” versus “not inducing today”.
4.2 Strengths and limitations
There were several strengths to this study. We used a large, validated,25 population-level dataset. Inclusion of all Victorian births eliminates selection bias. The availability of the specified indication for IOL improved upon the methodology of other observational studies23, 24 by avoiding the need to use a proxy for indication. Our study uses a more robust determination of non-medically indicated IOL compared with the exclusion of all women with minor complications of pregnancy that may not have influenced the decision to induce.
4.3 Interpretation
There have been several randomized controlled trials, including the recent ARRIVE trial,12 that have suggested that IOL reduces the risk of CB.12, 14, 17, 29 However, none of these trials examined CB as the primary outcome and only one examined uncomplicated pregnancies. They also may have selected populations that are non-representative of contemporary, lower-risk women experiencing an IOL. For example, in ARRIVE, 73% of women approached to participate in the study declined.12 . . . . Our findings support the findings of other whole-population observational studies.18, 19
With IOL becoming an increasingly common intervention, there is still need for future RCTs that assess the relationship between IOL and CB specifically in the contemporary lower-risk, nulliparous pregnancy. Given the challenges of conducting RCTs on this question, a well-designed prospective cohort study that follows nulliparous women without relevant complications at given gestations and compares IOL with expectant management would be informative. Women’s experiences of IOL and longer-term maternal and neonatal outcomes should also be considered in future research because breastfeeding success33 and developmental outcomes34 have been shown to less favorable following IOL regardless of the indication for IOL.
4.4 Conclusions
In a contemporary lower-risk nulliparous population in Victoria, Australia, we replicated and enhanced an influential retrospective cohort study, Stock 2012, which examined the relationship between IOL and CB. In contrast, in our study, the adjusted odds of CB were increased by 23%–43% following IOL at all gestations from 38 weeks. Perinatal mortality was rare in both the IOL and the expectantly managed groups. It was lower with IOL, but this reduction reached statistical significance only for IOL at 40 or more weeks. These results suggest that IOL in a lower-risk population may be associated with inherent risks. Consequently, maternity care providers should ensure women are informed of all potential risks and benefits of the non-medically indicated IOL.