The frequency of labor induction has been increasing in the United States and worldwide. In 2006, more than one in five pregnant women underwent induction of labor.1 Transcervical Foley catheter placement has been established as a safe and effective modality in the setting of labor induction.2 Potential mechanisms of cervical ripening include mechanical dilation of the cervix as well as release of endogenous prostaglandins from the fetal membranes. Foley catheter placement before the initiation of oxytocin has been shown to decrease the risk of cesarean delivery when compared with oxytocin alone.3,4 The latest Practice Bulletin published by the American College of Obstetricians and Gynecologists2 on induction of labor reports that there is no difference in the duration of induction to delivery or risk of cesarean delivery when the efficacy of a Foley catheter was compared with that of intravaginal prostaglandins. However, Foley catheter use decreases the risk of tachysystole (with or without fetal heart rate changes) and offers the advantage of lower cost, reversibility, and stability at room temperature.4,5 Potential side effects of Foley catheter for induction of labor include premature rupture of membranes, chorioamnionitis, bleeding, increased patient discomfort, displacement of the presenting part, and future risk of preterm birth.6 Multiple studies have shown no consistent association between Foley catheter use and these risks, although is it generally accepted that low-lying placenta is a relative contraindication for Foley catheter placement because of concern for potential disruption of the placental edge resulting in maternal hemorrhage.7–13
Embrey and Mollison14 first described the use of a transcervical Foley catheter for labor induction in 1967 using a 26 gauge Foley with a 50-mL balloon. Since this initial study, balloon inflation sizes of 30–80 mL have been reported in the literature with inconsistent results regarding induction-to-delivery duration and risk of cesarean delivery. There are only two randomized trials directly comparing 30-mL compared with 80-mL Foley catheter single balloon inflation sizes with delivery outcomes, which showed higher rates of delivery within 24 hours among the 80-mL group.15,16 However, a comparison of the 30-mL Foley balloon with different inflation volumes has not been conducted.
As a result of variations in the available literature on different Foley inflation sizes, the aim of this study was to evaluate if inflation of the Foley catheter balloon to 60 mL for labor induction resulted in a shorter time-to-delivery interval compared with the current standard 30-mL inflation. We also evaluated postexpulsion cervical dilation, oxytocin requirements, mode of delivery as well as maternal morbidity and neonatal outcomes.
MATERIALS AND METHODS
We recruited women admitted for induction of labor at the University of California at San Francisco between October 2006 and June 2009. Inclusion criteria were singleton pregnancies 37 weeks or greater of gestation, vertex presentation, and Bishop score less than 5. Exclusion criteria included spontaneous prodromal or early labor on admission (defined as three or more painful, regular spontaneous contractions in 10 minutes on admission), ruptured membranes, non–English-speaking or non–Spanish-speaking, low-lying placenta with the placental edge less than 2 cm from the internal os as measured by ultrasonography, prior cesarean delivery or hysterotomy, or any contraindication to vaginal delivery. The clinical diagnosis of Braxton-Hicks contractions without labor compared with prodromal or early labor varies across institutions and health care providers; this definition to rule out spontaneous prodromal or early labor was determined by the authors to be the most clear and consistent for this institution. To be clear, all women in the study were being induced, not augmented.
Patients were identified on admission to the labor and delivery department, and informed written consent was obtained by the medical staff. Randomization to receive a Bard 18-French Foley catheter with a 30-mL balloon tip inflated to either 30 mL or 60 mL was computer-generated, performed in blocks of eight, and stratified by parity. The randomization envelopes were opaque, ordered in a strict series, and kept in the labor and delivery department in a closed study box. The randomization envelope was opened and seen only by the labor nurse immediately before Foley balloon placement. Transcervical Foley balloon placement was performed by the digital or speculum method depending on clinician preference. Obstetricians, midwives, and patients were blinded to assigned randomization as follows: after correct placement of the Foley catheter was determined by the obstetrician or midwife (above the level of the internal os, in the extraamniotic space), the balloon was inflated with 30 mL sterile saline for all patients. The obstetrician or midwife then stepped out of the room. The labor nurse had prepared three 10-mL prefilled syringes with sterile saline. If the patient was randomly assigned to the 60-mL arm, the nurse further inflated the Foley balloon to a total of 60 mL. If the patient was randomly assigned to the 30-mL arm, the nurse briefly attached the additional prefilled syringes to the placed Foley balloon but did not actually inflate the balloon with any additional saline. The obstetrician or midwife then was allowed to return to the room. Intravenous oxytocin was started within 30 minutes of Foley balloon placement. Eventual expulsion of the Foley balloon was recorded by the labor nurse to avoid accidental unblinding of patients, obstetricians, or midwives. Follow-up letters to assess blinding were mailed to patients after hospital discharge regarding which size Foley balloon inflation they believed they had received.
The primary outcome for our study was all modes of delivery within 24 hours. Time zero was defined as the time of Foley catheter placement and balloon inflation. Secondary outcomes included all modes of delivery within 12 hours, absolute time to delivery, time to Foley balloon expulsion, cervical dilation after Foley balloon expulsion, time to active labor (defined as three or more painful contractions in 10 minutes, 80% or greater effacement, and cervical dilation 3 cm or greater), maximum oxytocin dose, chorioamnionitis (defined as maternal oral temperature of 38.0°C or higher during labor), meconium, method of delivery, indication for operative vaginal delivery or cesarean delivery, 5-minute Apgar score, umbilical cord artery pH and base excess, and evidence of cervical laceration or placental abruption (determined clinically at the discretion of the obstetrician or midwife at delivery).
Sample size was calculated by estimating a baseline delivery rate of 50% within 24 hours (based on internal data at the University of California at San Francisco). Assuming a two-tailed hypothesis test with 5% type I error and 80% power to detect a 20% difference between study arms (ie, 50% compared with 70%), 180 patients would be required. A χ2 test or Student t test was used for normally distributed categorical and continuous variables, respectively. The nonparametric Wilcoxon rank sum test was used to compare nonnormally distributed variables with outcomes reported using median and interquartile ranges. Normality was evaluated using the Shapiro-Wilk test. Multivariable analysis using a generalized linear model (Poisson family, log link) with robust standard errors was used to estimate relative risk (RR) and 95% confidence intervals (CIs). (“Log link” is the standard linkage function for Poisson regression. Robust standard errors ensured that the CIs avoided large overestimation or underestimation.17,18) P<.05 was used to indicate statistical significance for all analyses. All data were managed by STATA 8.2 (Statacorp, College Station, TX) and analyzed by “intent-to-treat.” The Committee on Human Research at the University of California at San Francisco approved this study (H41147-29401).
During the study period, 232 women were recruited. Thirty-three of these women did not meet study criteria and were not assigned to a study arm, three women elected to discontinue their induction and were discharged home undelivered without completion of data collection, and four women, for various clinical reasons, never had a Foley balloon placed and therefore did not have completion of data collection because time zero was never established. Thus, 192 patients had complete data available for statistical analysis (Fig. 1).
There were 94 patients randomly assigned to the 30-mL study arm and 98 patients randomly assigned to the 60-mL study arm. The two study groups were similar in all demographic aspects except for maternal age, in which the 30-mL arm was on average 2 years older than the 60-mL arm (Table 1). The induction indication “other” refers to the combined groups of nonreassuring antenatal testing, maternal indications such as congenital heart disease, need for timing and coordination of multiple medical services at delivery or for patients who lived long distances from a tertiary care center. There were no complications with placement or inflation of the Foley balloons in any study patient. There was one patient from the control 30-mL group in whom the Foley balloon was removed before spontaneous expulsion because of increased vaginal bleeding.
In univariable analyses, there was no difference in the primary outcome of delivery within 24 hours (66% compared with 64%, P=.72). However, the proportion of women delivering within 12 hours from Foley balloon placement was significantly higher in the 60-mL study arm (26% compared with 14%, P=.04). Additionally, median cervical dilation after Foley balloon expulsion was increased in the 60-mL arm as compared with the 30-mL study arm (4 cm compared with 3 cm, P<.01). For other secondary outcomes, there was no difference in cesarean delivery rate, operative vaginal delivery rate, absolute time to delivery, maximum oxytocin dose, perinatal complications, or neonatal outcomes (Table 2). Data on labor outcomes and complications in Table 2 refer to all modes of delivery unless otherwise specified. Among those patients who delivered within 12 hours, there was only one cesarean delivery. The overall vaginal delivery rate (spontaneous and assisted) for delivery within 12 hours was 100% for the 30-mL group (13 of 13 patients) and 96% for the 60-mL group (24 of 25 patients; P=.46). The overall cesarean delivery rate for the study was 22%. This rate is comparable to the overall primary cesarean delivery rate at this institution in women being induced.
In multivariable analysis, there was nearly a twofold increase in the relative “risk” of delivery within 12 hours associated with the 60-mL group (RR 1.84, 95% CI 1.003–3.39, P=.049). When stratified by parity, nulliparas who received the 60-mL Foley balloon were more likely to deliver within 12 hours compared with nulliparas who received the 30-mL Foley balloon (RR 2.88, 95% CI 0.97–8.54, P=.057), although statistical significance with smaller sample sizes when stratifying for parity was borderline (Table 3). Among multiparas, there was no statistical difference in the RR of delivering with 12 hours between the 30-mL and 60-mL Foley balloon inflation groups (RR 1.38, 95% CI 0.73–2.61, P=.32).
Follow-up letters were sent to patients after hospital discharge regarding which size Foley balloon inflation they believed that they received. Sixty-four percent of patients responded; 65% of responding patients guessed their Foley balloon inflation size correctly, 31% guessed incorrectly, and 4% could not decide. As a result of work shifts in the labor and delivery department, often the obstetrician or midwife who placed the Foley balloon was the not the one managing the entire labor course; thus, unblinding follow-up for obstetricians or midwives was not performed.
Our study showed that Foley balloon inflation to 60 mL increased the likelihood of achieving delivery within 12 hours compared with 30-mL inflation. With a 12% absolute risk increase for delivery within 12 hours among the 60-mL arm, the number needed to treat to achieve this outcome is only nine women. There was no difference in the rate of cesarean delivery among those patients delivering within 12 hours, indicating that 60-mL inflation did not shorten time to delivery at the expense of increased cesarean deliveries. Perhaps not surprisingly, the larger inflation size of 60 mL produced larger cervical dilation after Foley balloon expulsion. There were reassuring findings of no increases in cesarean delivery rates or maternal or neonatal morbidity with the larger Foley balloon inflation size. We hypothesized that there might be a higher rate of chorioamnionitis in the 60-mL group; like with larger postexpulsion cervical dilation, health care providers might be inclined to artificially rupture membranes when active labor is not achieved in the setting of larger mechanical dilation. It was reassuring that this hypothesis was not supported by the results from this trial.
The two other randomized trials comparing various Foley single balloon inflation sizes for induction of labor were published by Levy et al in 2004 and Kashanian et al in 2008. (Pennell et al19 in 2009 compared single balloon 30-mL inflation, Atad double balloon 80-mL inflation and PGE gel.) Levy randomly assigned 203 women to 30-mL or 80-mL Foley balloon inflation for induction of labor; oxytocin was started after expulsion of the Foley balloon.15 They found that the larger inflation size was significantly associated with a higher rate of postripening dilation of 3 cm or more (76% compared with 52%); in primiparous women, it resulted in a higher rate of deliveries within 24 hours (71% compared with 49%) as well as decreased need for oxytocin augmentation (69% compared with 90%). Kashanian et al16 randomly assigned 270 women to either 30-mL or 80-mL Foley catheter balloon inflation with oxytocin infusion or a third group of oxytocin alone. The number of favorable Bishop scores, change in Bishop score, overall vaginal delivery rate, and delivery rate within 24 hours were all significantly increased in the 80-mL group as compared with the 30-mL group.
Our study was similar to the studies by Levy et al and Kashanian et al in that Foley balloon inflation sizes of larger than 30 mL may decrease induction to delivery time. However, the two prior studies showed this difference at 24 hours, whereas our study showed only a difference in delivery rates at 12 hours. We also did not show a decreased need for oxytocin augmentation in the 60-mL Foley balloon group. Our results of increased efficacy of larger inflation sizes in nulliparous women were similar to the study by Levy et al.
It was surprising that delivery rates within 24 hours were 64% and 66% for the study and control groups, respectively. The women in this study, the majority of whom were nulliparous with unfavorable cervices, had higher rates of delivery within 24 hours than our baseline assumption. We hypothesize that both medical providers and nursing staff, knowing that these women were involved in a trial, may have been more attentive with labor management and oxytocin to produce this level of increase for deliveries within 24 hours. With this large of an increase in the primary outcome, this study was not powered to detect such small differences between the study groups.
Mechanisms for decreased time to delivery with larger Foley balloon inflation may include increased release of endogenous prostaglandins with the increased surface area of the balloon. The increased postexpulsion cervical dilation may be an effect of mechanical dilation with a larger balloon, an effect of increased prostaglandin release, or other mechanisms yet undiscovered. Implications for practice management include a possible decrease in hospitalization costs with a shorter time from induction to delivery with larger Foley balloon inflation. Future research could address quantitative levels of vaginal prostaglandins compared across various Foley balloon inflation sizes, examine issues such as type and model of Foley balloon as well as the use of extraamniotic saline infusion.19 Such analyses could also include financial analyses of hospitalization costs with varying inflation sizes and time to delivery.
Strengths of our study include the randomized control design as well as a diverse study population at the University of California at San Francisco. Although it is a tertiary care referral center with a slightly older average maternal age than many other parts of the United States, our patient population consists of a substantial number of low-risk women, women across multiple race and ethnicity groups as well as insurance coverage categories. The larger racial, ethnic, and socioeconomic diversity contributes to the generalizability of this study across various populations. The study also has several potential limitations. First, blinding was not completely successful among patients, and providers potentially could have been unblinded during expulsion of the Foley balloon. Although such unblinding should not have affected some of the objectively measured outcomes such as length of labor, duration of induction, or cesarean delivery, it might affect the clinical decision process regarding labor management. Because oxytocin dosing protocols at the study institution are standardized and independent of this study design, accidental unblinding likely would not infer significant bias for this outcome. Finally, although we conducted an a priori sample size calculation to ensure adequate power to examine length of labor, we were underpowered to detect small differences in cesarean delivery or effect modification and interaction.
Another potential limitation was the inclusion of both nulliparous and multiparous women. If we had recruited only nulliparous women, this would have increased the statistical power for outcomes among that group. However, it was a deliberate methodological decision to include both multiparous and nulliparous women to observe if there was a difference in efficacy by parity. It was also anticipated that there would be a proportion of patients who had received some form of prostaglandin before Foley balloon placement. Similarly, this was intentional to reproduce the common situations for which health care providers choose to use a Foley balloon as the next method for induction of labor. Prostaglandin use before Foley balloon placement was common in both study groups (Table 1) and was evenly distributed across both study arms. Thus, prostaglandin use could not have contributed any confounding bias and was not adjusted for in statistical analysis.
In conclusion, it appears from this and other studies that inflation of a transcervical Foley balloon for induction of labor to 60 mL or 80 mL is a more effective method of labor induction as compared with 30-mL inflation. For health care providers who only have the 30-mL Foley balloon tip, our study suggests that inflating to a volume of 60 mL is safe and potentially beneficial. We recommend considering a 60-mL Foley balloon for induction of labor, particularly in nulliparous women. For delivery within 12 hours, a number needed to treat of only nine women provides an achievable intervention for health care providers without increasing cesarean delivery rates and maintaining both maternal and neonatal safety.
1. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Kirmeyer S, et al. Births: final data for 2006. Natl Vital Stat Rep 2009;57:1–102.
2. Induction of labor. ACOG Practice Bulletin No. 107. American College of Obstetricians and Gynecologists. Obstet Gynecol 2009;114:386–97.
3. Leiberman JR, Piura B, Chaim W, Cohen A. The cervical balloon method for induction of labor. Acta Obstet Gynecol Scand 1977;56:499–503.
4. Boulvain M, Kelly A, Lohse C, Stan C, Irion O. Mechanical methods for induction of labour. The Cochrane Database of Systematic Reviews 2001, Issue 4. Art. No. CD001233. DOI: 10.1002/14651858.CD001233.
5. Gelber S, Sciscione A. Mechanical methods of cervical ripening and labor induction. Clin Obstet Gynecol 2006;49:642–57.
6. Sherman DJ, Frenkel E, Tovbin J, Arieli S, Caspi E, Bukovsky I. Ripening of the unfavorable cervix with extraamniotic catheter balloon: clinical experience and review. Obstet Gynecol Surv 1996;51:621–7.
7. Sciscione AC, McCullough H, Manley JS, Shlossman PA, Pollock M, Colmorgen G. A prospective, randomized comparison of Foley catheter insertion versus intracervical prostaglandin E2 gel for preinduction cervical ripening. Am J Obstet Gynecol 1999;180:55–60.
8. Abramovici D, Goldwasser S, Mabie BC, Mercer BM, Goldwasser R, Sibai BM. A randomized comparison of oral misoprostol versus Foley catheter and oxytocin for induction of labor at term. Am J Obstet Gynecol 1999;181:1108–12.
9. Barrilleaux PS, Bofill JA, Terrone DA, Magann EF, May WL, Morrison JC. Cervical ripening and induction of labor with misoprostol, dinoprostone gel, and a Foley catheter: a randomized trial of 3 techniques. Am J Obstet Gynecol 2002;186:1124–9.
10. Chung JH, Huang WH, Rumney PJ, Garite TJ, Nageotte MP. A prospective randomized controlled trial that compared misoprostol, Foley catheter, and combination misoprostol–Foley catheter for labor induction. Am J Obstet Gynecol 2003;189:1031–5.
11. Sciscione A, Larkin M, O'Shea A, Pollock M, Hoffman M, Colmorgen G. Preinduction cervical ripening with the Foley catheter and the risk of subsequent preterm birth. Am J Obstet Gynecol 2004;190:751–4.
12. Prager M, Eneroth-Grimfors E, Edlund M, Marions L. A randomized controlled trial of intravaginal dinoprostone, intravaginal misoprostol and transcervical balloon catheter for labour induction. BJOG 2008;115:1443–50.
13. Hill JB, Thigpen BD, Bofill JA, Magann E, Moore LE, Martin JN Jr. A randomized clinical trial comparing vaginal misoprostol versus cervical Foley plus oral misoprostol for cervical ripening and labor induction. Am J Perinatol 2009;26:33–8.
14. Embrey MP, Mollison BG. The unfavourable cervix and induction of labour using a cervical balloon. J Obstet Gynaecol Br Commonw 1967;74:44–8.
15. Levy R, Kanengiser B, Furman B, Ben Arie A, Brown D, Hagay ZJ. A randomized trial comparing a 30-mL and an 80-mL Foley catheter balloon for preinduction cervical ripening. Am J Obstet Gynecol 2004;191:1632–6.
16. Kashanian M, Nazemi M, Malakzadegan A. Comparison of 30-mL and 80-mL Foley catheter balloons and oxytocin for preinduction cervical ripening. Int J Gynaecol Obstet 2009;105:174–5.
17. McNutt LA, Wu C, Xue X, Hafner JP. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol 2003;157:940–3.
18. Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004;159:702–6.
19. Pennell CE, Henderson JJ, O'Neill MJ, McCleery S, Doherty DA, Dickinson JE. Induction of labour in nulliparous women with an unfavourable cervix: a randomised controlled trial comparing double and single balloon catheters and PGE2 gel. BJOG 2009;116:1443–52.