Induction of labor has been steadily increasing in the United States. From 1990 to 2006, the overall induction rate rose from 10% to 22.5%.1 Although many inductions are performed for medical reasons, a growing number of inductions are elective.2 By definition, elective induction of labor has no accepted medical indication, although there may be reasons that an elective induction of labor is performed, such as patient or physician convenience.
Some have raised concerns with this increase in elective induction of labor given that a wealth of information from observational studies suggests that labor induction is associated with an increased cesarean delivery rate.3–5 Studies consistently have shown this increased cesarean delivery rate in nulliparous women, regardless of whether their cervices are favorable or unfavorable.6,7 However, the large majority of observational studies have used a design in which labor induction is compared with spontaneous labor at a given gestational age.3–7 However, this design is not reflective of the actual clinical situation. Although it is true that any one woman may have an induction or experience a spontaneous labor, it is not true that she may experience either of these events at the same point in time or will experience the latter with certainty. A more appropriate clinical comparison is induction compared with expectant management.
A recent study by Caughey et al8 comparing induction of labor with expectant management demonstrated that nulliparous women who were electively induced had a decreased cesarean delivery rate compared with those who were expectantly managed. However, this study did not evaluate whether cervical status was similar between the two groups. This factor is particularly important because cervical status (eg, as assessed by Bishop score) repeatedly has been shown to be significantly associated with cesarean delivery. If women in the induction of labor group were more likely to have a favorable cervix, the cesarean delivery rate might appear similar as a result of selection bias. Our objective was to compare outcomes of labor between nulliparous women with a favorable cervix who underwent either elective labor induction or expectant management beyond 39 weeks of gestation. We hypothesized that, after accounting for cervical status, women who are electively induced would have a higher chance of cesarean delivery than those who are expectantly managed.
MATERIALS AND METHODS
We conducted a retrospective study of nulliparous women with a favorable cervix who were either electively induced or expectantly managed at term. Inclusion criteria were nulliparity, gestational age 39 0/7 weeks or greater, a singleton vertex gestation, known cervical status at 38 0/7–38 6/7 weeks of gestation, and a modified Bishop Score of at least 5.9 Elective induction is prescribed unless a woman is at least 39 weeks of gestation by the criteria outlined by the American College of Obstetricians and Gynecologists.10 Women undergoing an induction of labor for a medical indication were excluded (diabetes, chronic hypertension, Rh isoimmunization). Institutional Review Board approval for the study was obtained from Northwestern University.
A perinatal database was used to identify all women who met inclusion criteria undergoing an induction of labor from July 1, 2006, to June 30, 2008. Each admission history and physical was reviewed to identify women with a favorable cervix who were being electively induced, and data were abstracted by investigators directly from the chart. Elective inductions were considered to occur when the physician either stated the indication of the induction was “elective” or when the stated indication did not meet accepted medical criteria for labor induction (eg, patient discomfort with pregnancy, patient's home “far from the hospital”). Women induced at 41 0/7 weeks of gestation were not considered electively induced because induction of labor at or after this gestational age is acknowledged by the American College of Obstetricians and Gynecologists to constitute an obstetric indication.11 From all women who underwent elective labor induction, 294 (ie, the number needed to provide the targeted power) were randomly selected based on a computer-generated random-number table. A modified Bishop score was calculated based on the examination at admission for labor induction.
Once the group for elective induction of labor was obtained, a control group was selected. Data extracted from the electronic medical record identified all women meeting inclusion criteria who delivered during the same time period. Women were considered eligible for the control group if they had reached 39 0/7 weeks of gestation and were expectantly managed. These women could have experienced spontaneous labor, been induced for obstetric indications (eg, oligohydramnios, preeclampsia, postterm pregnancy), or could have received an unlabored cesarean delivery for an obstetric indication (eg, nonreassuring fetal status on antepartum testing). Also, women were only considered eligible for the control group if their provider group also had patients in the group of women who were electively induced. This restriction was imposed to limit the possibility that different practice styles could confound the results. Of all women who were expectantly managed, 294 were randomly selected for the control group using a computer-generated random-number table. Cervical examination data between 38 0/7 and 38 6/7 weeks of gestation were obtained on all expectantly managed patients by reviewing office medical records.
Data were collected regarding maternal and neonatal outcomes as well as regarding the obstetric course (eg, time at delivery, length of labor). Length of labor was defined as the time from admission to the labor and delivery unit to the time of delivery. The primary outcome was cesarean delivery. Given an 80% power at an two-sided α of .05, and to detect an increase in the cesarean delivery rate from 20% in the expectant management group to 30% in the elective induction of labor, 588 women (divided equally between the two groups) were required for analysis. Student t test was used to compare continuous variables, and the χ2 test was used to compare proportions between groups. Logistic regression was used to control for potential confounding with odds ratios (ORs) and 95% confidence intervals (CIs) calculated. All tests were two-tailed, and a P value of <.05 was considered significant. Minitab 13 (College Station, TX) was used for analysis.
Between July 2006 and June 2008, 294 women who underwent elective labor induction were compared with 294 who were expectantly managed after 39 weeks of gestation. These women were compared with 294 randomly selected women who underwent expectant management during the same time period. The elective induction of labor and expectant management groups were similar with respect to demographic data (Table 1) except that women in the elective induction of labor group had a slightly higher body mass index and a slightly greater Bishop score.
Of women who were expectantly managed, 54.4% (160 of 294) delivered between 39 0/7 and 39 6/7 weeks of gestation, 35.4% (104 of 294) delivered between 40 0/7 and 40 6/7 weeks of gestation, and 10.2% (30 of 294) delivered between 41 0/7 and 41 6/7 of gestation. Of these women, 17.7% ultimately required induction, most commonly for reaching 41 weeks of gestation (43.4%) followed by oligohydramnios (22.3%). There were no expectantly managed women who required a cesarean delivery without labor or who were delivered with an intrauterine fetal demise.
There were significant differences in the characteristics of labor experienced by women who were electively induced. They spent more time in labor and delivery, whether measured as a continuous or categorical variable, and, not unexpectedly, received oxytocin more frequently. Conversely, both groups were equally likely to deliver during daytime hours (6:00 am to 6:00 pm) (Table 2). There were no differences with regard to maternal outcomes (Table 3). Specifically, the rate of cesarean delivery was not different between the two groups. Furthermore, there were no differences with regard to other obstetric events (eg, operative vaginal delivery) or complications. Neonatal outcomes were also similar between groups. There was a trend toward a decreased frequency of meconium-stained amniotic fluid; however, this difference was not statistically significant (Table 4).
Multivariable logistic regression was used to further explore the association between elective induction and cesarean delivery (the primary outcome) as well as elective induction and length of labor. This was done to control for potential confounding of body mass index and modified Bishop score, the two variables that were statistically different between the two groups. Despite controlling for these variables, the association of elective induction with a labor longer than 12 hours (adjusted OR 2.78, 95% CI 1.91–4.05) and lack of association with cesarean delivery (adjusted OR 1.03, 95% CI 0.66–1.60) remained stable.
Next, to determine whether elective inductions done at a later gestational age (eg, at 40 3/7 weeks) could be responsible for affecting the observed outcomes, the analysis was redone with the population of electively induced women restricted to those whose elective induction occurred before 40 0/7 weeks of gestation. Of the 294 women who underwent elective induction, 127 were between 39 0/7 and 39 6/7 weeks of gestation. When these women were compared with those who were expectantly managed after 39 0/7 weeks of gestation, the associations of induction with the selected outcomes were unchanged. Women electively induced again were noted to have an increased mean length of labor (12.1±4.3, compared with 9.8±5.1, P<.001) and chance of labor more than 12 hours (40% compared with 30%, P=.048), but no difference in frequency of cesarean delivery (17% compared with 20%, P=.51) than those who were expectantly managed. These results were unchanged in multivariable analysis (data not shown).
The primary objective of this study was to investigate whether elective induction of labor at 39 0/7 weeks of gestation or greater among nulliparous women with favorable cervices, when compared with expectant management at 39 0/7 weeks of gestation, is associated with an increased chance of cesarean delivery. In fact, in this study elective induction was not associated with an increased risk of cesarean delivery.
Our results are contrary to those of the large majority of previous investigations of labor induction. Most studies have found at least a doubling of the cesarean delivery rate among nulliparous women who are induced.12 In one of the largest studies, Dublin et al13 compared 1,020 women who underwent labor induction with 3,622 women who spontaneously labored and found an increase in the cesarean delivery rate from 10% to 19%. This study did not stratify by parity or cervical ripeness. Nevertheless, investigators who have stratified by or examined these factors have found that nulliparous women who are electively induced or who are induced with a favorable cervix are still at increased risk of cesarean delivery compared with those who are spontaneously laboring.7
Yet the design of these retrospective studies has been criticized because of the choice of the comparison group.14 The alternative to elective induction is not spontaneous labor, but expectant management. If a woman were not to be induced, she would not labor but would await labor with several different possible outcomes, including the eventual need for labor induction. Thus, the proper comparison for a woman who is being induced is a woman who is being expectantly managed. Indeed, this is exactly the design that typically has been used when randomized controlled trials evaluating labor induction have been performed.15,16 The largest of these studies compared induction at 41 weeks of gestation with expectant management and found no difference in cesarean delivery rates.17 However, only smaller randomized studies have evaluated the effects of induction at earlier gestational ages, and these studies do not have adequate power to discern a clinically relevant difference in cesarean delivery rate.18
Caughey et al8 used a retrospective cohort design with the appropriate comparison groups (ie, induction compared with expectant management) and did not find an increased chance of cesarean delivery when an induction at or after 38 weeks of gestation was undertaken. However, they did not control for cervical status, which could bias the results toward the null hypothesis if, for example, women with unfavorable cervices were more likely to be expectantly managed. Also, their population was not restricted to women undergoing elective induction, thus limiting the applicability of the conclusions to this population. The present study examined only electively induced women and accounts for the important potential confounding variable of cervical status by only including women whose cervix was favorable at the time when a management choice—either induction or expectant management—was undertaken. Yet, even after accounting for this factor, the results are similar to those of Caughey et al,8 namely, that the chance of cesarean delivery is similar between women who were electively induced and women who were expectantly managed.
This is not to say, however, that both groups were identical, and it should be noted that those who were induced had significantly longer duration of stay on labor and delivery and significantly greater use of a specific intervention such as oxytocin. These findings are consistent with those of Seyb et al6 who found a 33% increase in time spent in labor and delivery among women who underwent labor induction. They found that increased duration of labor equated to a 17% increase in cost. This increase in cost is not just related to increased cesarean delivery rate. Maslow et al19 found that costs associated with vaginal delivery after labor induction were 25% higher than vaginal delivery after spontaneous labor.
The present study has several strengths. A sample size, based on an a priori power calculation, was obtained that should limit the possibility of a type II error with regard to the primary outcome of cesarean delivery. Furthermore, charts were individually reviewed to ensure that patients were correctly classified as elective inductions and that their cervices were favorable at the time when either management choice could be made.
Nevertheless, limitations should be noted. There are 19 different provider groups who perform deliveries at our institution. This permits a variety of practice styles, including with regard to the undertaking of labor induction and the threshold for cesarean delivery. There is the possibility, therefore, that an association between induction and cesarean delivery, if it did exist, could be obscured by a selection bias related to different management by different provider groups. Although this possibility cannot be summarily dismissed, many different provider groups were represented in this study, all of whom cared for women in both groups, thereby lessening the possibility of bias. Also, ideally women electively induced at 39 0/7 weeks of gestation and a small number of days after this gestational age would be compared with those expectantly managed after 39 0/7 weeks of gestation. However, the variation in clinical medicine outside of a randomized trial would not allow for an adequate sample size if only women who were induced in this narrow window were eligible for inclusion. Yet, there is reason to believe that the inclusion criteria did not obscure a positive association between elective induction and cesarean delivery. First, when only women who were induced before 40 weeks of gestation were included in analysis, the results were no different. Second, the inclusion of women who were electively induced between 40 and 40 5/7 weeks of gestation, if anything, should make labor induction appear more highly associated with cesarean delivery, because increasing gestational age is associated with an increased risk of cesarean delivery.11 Lastly, this study, although of adequate size for the study of the primary outcome, did not have sufficient power to detect clinically relevant changes in secondary outcomes such as those regarding the neonate.
This study calls into question the previously documented relationship between elective labor induction and cesarean delivery in nulliparous women. This does not imply that the threshold for elective induction should be lowered. There remains evidence that it increases use of healthcare resources (eg, duration in labor and delivery), and there is no compelling evidence that it improves health outcomes. Ultimately, however, our conclusions as well as those of others largely are based on retrospective data and a properly powered randomized trial will be required to truly appreciate both the maternal and neonatal consequences of electing to undertake an induction without a medical indication rather than to continue expectant management.
1. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Kirmeyer S, et al. Birth: final data for 2006. Natl Vital Stat Rep 2007;56:1–103.
2. Zhang J, Yancey MK, Henderson CE. US national trends in labor induction, 1989–1998. J Reprod Med 2002;47:120–4.
3. van Gemund N, Hardeman A, Scherjon SA, Kanhai HH. Intervention rates after elective induction of labor compared to labor with a spontaneous onset. A matched cohort study. Gynecol Obstet Invest 2003;56:133–8.
4. Vrouenraets FP, Roumen FJ, Dehing CJ, van den Akker ES, Aarts MJ, Scheve EJ. Bishop score and risk of cesarean delivery after induction of labor in nulliparous women. Obstet Gynecol 2005;105:690–7.
5. Vahratian A, Zhang J, Troendle JF. Labor progression and risk of cesarean delivery in electively induced nulliparas. Obstet Gynecol 2005;105:698–704.
6. Seyb ST, Berka RJ, Socol ML, Dooley SL. Risk of cesarean delivery with elective induction of labor at term in nulliparous women. Obstet Gynecol 1999;94:600–7.
7. Yeast JD, Jones A, Poskin M. Induction of labor and the relationship to cesarean delivery: a review of 7001 consecutive inductions. Am J Obstet Gynecol 1999;180:628–33.
8. Caughey AB, Nicholson JM, Cheng YW, Lyell DJ, Washington AE. Induction of labor and cesarean delivery by gestational age. Am J Obstet Gynecol 2006;195:700–5.
9. Lange AP, Secher NJ, Westergaard JG, Skovgård I. Prelabor evaluation of inducibility. Obstet Gynecol 1982;60:137–47.
10. Induction of labor. ACOG Practice Bulletin No. 107. American College of Obstetricians and Gynecologists. Obstet Gynecol 2009;114:386–97
11. Management of postterm pregnancy. ACOG Practice Bulletin No. 55. American College of Obstetricians and Gynecologists. Obstet Gynecol 2004;104:639–46.
12. Heffner LJ, Elkin E, Fretts RC. Impact of labor induction, gestational age, and maternal age on cesarean delivery rates. Obstet Gynecol 2003;102:287–93.
13. Dublin S, Lydon-Rochelle M, Kaplan RC, Watts DH, Critchlow CW. Maternal and neonatal outcomes after induction of labor without an identified indication. Am J Obstet Gynecol 2000;183:986–94.
14. Caughey AB, Sundaram V, Kaimal AJ, Gienger A, Cheng YW, McDonald KM, et al. Systematic review: elective induction of labor versus expectant management of pregnancy. Ann Intern Med 2009;151:252–63. W53–63.
15. Cole RA, Howie PW, Macnaughton MC. Elective induction of labour. A randomised prospective trial. Lancet 1975;1:767–70.
16. Egarter C, Kofler E, Fitz R, Husslein P. Is induction of labor indicated in prolonged pregnancy? Results of a prospective randomised trial. Gynecol Obstet Invest 1989;27:6–9.
17. Hannah ME, Hannah WJ, Hellmann J, Hewson S, Milner R, Willan A. Induction of labor as compared with serial antenatal monitoring in post-term pregnancy. A randomized controlled trial. The Canadian Multicenter Post-term Pregnancy Trial Group. N Engl J Med 1992;326:1587–92.
18. Nielsen PE, Howard BC, Hill CC, Larson PL, Holland RH, Smith PN. Comparison of elective induction of labor with favorable Bishop scores versus expectant management: a randomized clinical trial. J Matern Fetal Neonatal Med 2005;18:59–64.
19. Maslow AS, Sweeny AL. Elective induction of labor as a risk factor for cesarean delivery among low-risk women at term. Obstet Gynecol 2000;95:917–22.