Comparative analysis of pharmacologic and mechanical methods for labor induction in patients at full-term pregnancy—A systematic review

Introduction

Labor induction methods have been described in academic journals since 1573 when Ambroise Paré implemented amniotomy and detachment of membranes to trigger labor in patients with antepartum hemorrhages, as well as in patients with pelvic dystocia.¹ Over time, these techniques have become common practice to decide on the delivery of patients in whom the continuation of pregnancy involves significant maternal or fetal risks. The modern era adds value to labor induction methods by introducing convenience as an indication.

To date, many methods and combinations have been described for inducing labor. The most reliable methods involve the use of drugs (oxytocin and prostaglandins) and mechanical intracervical devices (Foley balloon, Cook balloon). For these reasons, we aimed to describe and compare features of the different methodologies for triggering labor in patients at full-term pregnancy.

Methods

In this review we created a protocol following the recommendations of preferred reporting criteria for systematic reviews (PRISMA 2020) for the systematic review protocol checklist.¹⁵ Furthermore, we tried to guide our study rationale and create a systematic review of the literature to answer the question: “In patients with full-term gestation - does combining the Foley balloon with dinoprostone for the artificial induction of labor - offer better perinatal results over time from the application of the labor induction method until birth?”

The studies will be included after searching in MEDLINE database (using PubMed). A search of the PubMed database was conducted on 23/08/2023 to identify articles showing the use of dinoprostone + Foley catheter and/or a Foley catheter in full-term patients to induce labor. We used the “AND” and “OR” tools and the search method: “(dinoprostone OR Foley+dinoprostone) AND (artificial induction of labor at full term gestation)” and we limited the search to articles published within the past five years, only in English, excluding reviews, abstracts and others shorts communication.

Two reviewers screened independently articles for relevance, according to the flow chart in Figure 1. A third reviewer was assigned to mediate any issues that might arise in the selection process. Articles were included in which labor was induced through these techniques in full-term patients who met the following criteria: monofetal pregnancy, cranial presentation, intact membranes, and no history of uterine scar - 2 articles.

Figure 1. Flowchart of the study selection process.

A full-text analysis of included articles was performed for comparison between methodologies. The literature search yielded 19 articles that analyzed the aforementioned methods or their association, of which 8 were published in the last 5 years. After applying the proposed criteria, four studies involving patients with spontaneous ruptured membranes before the application of labor induction methods, and another study that included patients with misoprostol-induced labor were excluded. Thus, only three studies met the proposed criteria, all of which were randomized clinical trials.

The outcomes for which data was searched were time from placement of the transcervical Foley catheter to vaginal delivery, caesarean section rate, infectious complications, Apgar scores and neonatal intensive care unit admission. All results that were compatible with each outcome domain from each study were searched, the data were manually extracted by 2 evaluators who worked independently and later compared.

Registration: the review was not registered.

Results and discussion

For the median time for delivery in Foley and Foley+dinoprostone induced labor, two studies were identified. The studies’ characteristics (study type, condition of the cervix, method of labor induction, including population and its repartition (primiparous and multiparous), time to delivery, p-value) are shown in Table 1. The limits of this study are represented by the small number of studies that address the proposed theme, the relatively small number of patients included, the different population characteristics (body mass index, age).

The randomized clinical trial published by Edwards et al. (2021) involved 100 patients with full-term pregnancy (>37 weeks) who were randomly divided into four categories: 26 primiparous women in whom labor was induced with a Foley catheter and dinoprostone, 24 primiparous women in whom labor was induced with a Foley catheter, and two other groups with 25 multiparous patients each as control groups.² The results revealed that the median time for vaginal delivery in Foley and dinoprostone-induced labor was 21.2 hours, compared with 31.3 hours in the patients where labor was induced with a Foley catheter (Wilcoxon p<0.05).² The median time of delivery for multiparous patients triggered with the Foley catheter and dinoprostone was 17.1 hours, while in multiparous patients triggered with the Foley catheter alone, it was 14.8 hours (Wilcoxon p=0.21).² Analyzing the cesarean delivery rate, labor in nulliparous patients induced with the Foley catheter and dinoprostone appears to have lower cesarean delivery rates than those induced only with Foley catheter, but the difference is not statistically significant (Gray test, p=0.34).² In multiparous patients, combining the Foley catheter and dinoprostone produced higher rates of cesarean delivery, compared with those triggered only with the Foley catheter.² However, this difference is not statistically significant (Gray test, p=0.20).²

Another randomized clinical trial by Vallikkannu et al. (2022) compared the time required until delivery, rate of cesarean section, and satisfaction from induction of labor among term, nulliparous patients, with a Bishop score <5, using a Foley catheter alone and Foley catheter + dinoprostone administered vaginally.³ Consistent with the previous trial, the time from the application of the labor induction method to delivery was shorter in patients who received the combined Foley catheter and dinoprostone method (22.5 hours), compared with the Foley-catheter-only method (35.1 h, p<0.001); no differences were detected in terms of patient satisfaction and the rate of cesarean section in patients receiving the combined method (34%) compared with those where labor was triggered using the Foley catheter alone (49.5 hours, p=0.02, RR: 0.7 (95% CI 0.5–0.9)).³

The third randomized clinical trial that met the search criteria was conducted by Barda et al. (2018). It consisted in a randomization of 300 patients with an unfavorable Bishop score at term to compare the efficacy of triggering labor using a Foley catheter and dinoprostone administered vaginally.⁴ The results showed that the time from the application of the labor induction method was shorter in patients induced with the Foley catheter alone, but this group required the administration of a higher dose of oxytocin.⁴ In addition, inducing labor using a Foley catheter resulted in a lower cesarean section rate, and neonatal indicators were similar in both study groups.⁴

Analysis of papers describing labor induction methods revealed several meta-analyses. A recent study (Dong et al. 2022) of 8796 full-term pregnancies examined the indications and main outcomes of induced labor in patients at 39/40 weeks of gestation.⁵ The findings demonstrated that there was no increase in cesarean section rate or perinatal morbidity when labor was triggered at 39/40 weeks of amenorrhea for non-medical indications.⁵ At the same time, inducing labor at the above-mentioned gestational age is associated with a lower incidence of maternal hypertensive complications during pregnancy, shorter duration of the first period of labor, fewer cases of meconium amniotic fluid during birth, lower mean birth weight, longer maternal hospitalization, and a higher rate of epidural usage.⁵

To assess the effectiveness of labor induction methods, Quach et al. (2022) conducted a meta-analysis of 2990 patients based on the PROBAAT trials, in which they determined the maternal and fetal characteristics that influenced the cesarean section rate following induction of labor in patients with full-term, single-fetal pregnancy, intact membranes, cephalic presentation, and unfavorable cervix.⁶ The authors developed mathematical models to determine the failure rate of induction of labor and conversion of birth to cesarean section.⁶ Of the 2990 patients who underwent labor induction, 10.5% had a cesarean section because of lack of dilation progress, and 7.6% had a cesarean section owing to fetal distress.⁶

The risk of cesarean section was higher in women aged 31–35 years (aOR: 1.15 (95% CI: 1.15–1.99)), nulliparous (aOR: 8.07 (95% CI: 5.34–12.18)), and of sub-Saharan African descent (aOR: 2.09 (95% CI: 1.33–3.28)).⁶ The main indications for cesarean section due to lack of dilation progress were increased BMI (aOR, 1.06 (95% CI: 1.04–1.08)) and fetal growth greater than the 80th percentile (aOR, 4.08 (95% CI: 2.75–6.05)).⁶ The main indications for cesarean delivery due to fetal distress were increased maternal age (aOR: 1.09 (95% CI: 1.05–1.12)), increased BMI (aOR: 1.05 (95% CI: 1.03–1.08)), and fetal growth below the 10th percentile (aOR: 1.93 (95% CI: 1.22–3.05)).⁶ The Bishop score did not demonstrate a statistically significant association with the risk of cesarean delivery, for both lack of progression of labor and fetal distress.⁶ Thus, maternal age, BMI, parity, ethnicity, and fetal growth percentile are predictive of the risk of cesarean section in patients who had a triggered labor; however, the direction and magnitude of the association differ in the weight of determining the decision to perform cesarean section.⁶

Regarding maternal age, another meta-analysis (Fonseca et al. 2020) involving 81,151 patients revealed that induction of labor in patients aged 35 years or older was not associated with a higher risk of cesarean delivery (OR: 0.97 (95% CI: 0.86–1.1)), instrumental delivery (OR: 1.12 (95% CI: 0.96–1.32)), or postpartum hemorrhage (OR: 1.11 (95% CI: 0.88–1.41)).⁷ If the indications for triggering labor are not clear based on maternal health status, the primary consideration in inducing labor from the fetal point of view is fetal distress and the risk of deterioration of fetal well-being in utero.⁷

A Cochrane meta-analysis was conducted by Middleton et al. (2020) on approximately 21,000 patients based on 34 randomized clinical trials to investigate the impact of triggering labor at or after 37 weeks of gestation in terms of antenatal fetal death.⁸ According to published results, triggering labor at or after 37 weeks of gestation significantly reduces the risk of antenatal fetal death (0.4 deaths versus 3 deaths reported per 1000 pregnancies (induced births versus expectant management, respectively)).⁸ It is also important to note that the cesarean section rate was even lower in patients in whom labor was induced compared with those who received expectant management (RR: 0.90 (95% CI: 0.85–0.95)) and probably a lower rate or no difference in the instrumented birth rate (RR: 1.03 (95% CI: 0.96–1.10)).⁸ On the other hand, triggering labor may increase the risk of perineal trauma (severe perineal trauma RR: 1.04 (95% CI: 0.85–1.26); 5 trials, 11,589 patients) and postpartum hemorrhage (RR: 1.02 (95% CI: 0.91–1.15)).⁸ Neonatal admission rates to intensive care units were lower in patients who received induction of labor compared with patients who received expectant management (RR: 0.88 (95% CI: 0.80–0.96)), based on a study of 17 trials and 17,826 neonates.⁸ Additionally, there were fewer newborns with an APGAR score <7 in patients who received labor induction compared with those who received expectant management (RR 0.73 (95% CI: 0.56–0.96)), according to a review of 20 trials accounting for 18,345 neonates.⁸ No statistically significant differences were observed in the length of hospital stay, rate of outpatient newborn breastfeeding, neonatal encephalopathy, or neonatal trauma.⁸ Additionally, no randomized clinical trials have reported neurodevelopmental disorders or childhood depressive disorders during the follow-up.⁸ Thus, a discussion on induction of labor at or after 37 weeks of gestation should focus on the risk profiles of each patient, complemented by paraclinical results and patient preferences.⁸

A similar meta-analysis examining elective labor induction at 39 weeks of gestation versus expectant management was conducted by Grobman and Caughey (2019). The authors identified 375 studies with a total of 66,019 patients who underwent elective triggering of labor at 39 weeks and 584,390 patients who received expectant management.⁹ The study revealed that triggering labor at 39 weeks, compared with expectant management after this gestational age, was associated with a significant decrease in the rate of cesarean delivery (26.4% versus 29.1%, RR 0.83 (95% CI: 0.74–0.93)) and peripartum maternal infections (2.8% versus 5.2%, RR 0.53 (95% CI: 0.39–0.72)), and an improvement in other outcomes including neonatal intensive care unit admission (3.5% versus 5.5%, RR 0.8 (95% CI: 0.72–0.98)), respiratory morbidity (0.7% versus 1.5%, RR 0.71 (95% CI: 0.59–0.85)), and mortality (0.04% versus 0.2%, RR 0.27 (95% CI: 0.09–0.76)).⁹

The role of early amniotomy in triggering labor was studied in a systematic review and meta-analysis (Kim et al. 2019) that included seven studies and a total of 1775 patients.¹⁰ Two Cochrane meta-analyses that independently investigated the role of early amniotomy alone and early amniotomy associated with oxytocin infusion were included in the review.¹⁰ The authors showed that early amniotomy does not produce any benefit in terms of shortening the first period of labor and does not influence the risk of cesarean delivery, and amniotomy associated with oxytocin infusion can shorten the duration of the first period of labor by up to 1.28 hours compared with expectant management.¹⁰ It should be noted that the studies were performed in patients who presented with spontaneous labor and were subsequently associated with amniotomy or amniotomy plus oxytocin infusion.¹⁰ The role of early amniotomy is less clear. Therefore, several randomized trials have investigated the maternal and neonatal outcomes of early amniotomy in the setting of triggered labor; however, data are inconclusive regarding the time to delivery, rate of cesarean delivery, and rate of infectious complications.

Another meta-analysis (Orr et al. 2020) summarized 30 randomized clinical trials involving 6465 patients that compared labor induction using a Foley catheter, Foley catheter and oxytocin, Foley catheter and prostaglandins, and single-use prostaglandins (misoprostol and dinoprostone).¹¹ The authors concluded that combining oxytocin with a Foley catheter versus a Foley catheter alone reduced the time to delivery by 4.2 hours (median: -4.2 hours (95% CI: -6.5 to -1.9)), and combining prostaglandins with a Foley catheter versus a Foley catheter alone reduced the time to delivery by 2.9 hours (median: -2.9 hours (95% CI: -5.7 to 0.0, p=0.05)).¹¹

There were no differences in the rates of cesarean delivery, chorioamnionitis, use of epidural analgesia, tachysystole, postpartum hemorrhage, meconium amniotic fluid, neonatal admissions to ATI units, and 5-min APGAR scores for the Foley catheter-oxytocin and Foley catheter-dinoprostone combinations; instead, the rate of endometritis was higher in patients who received the Foley catheter-prostaglandin combination.¹¹ The Foley catheter-misoprostol, Foley catheter-dinoprostone, and Foley catheter-oxytocin combinations did not demonstrate a statistically significant difference in the duration of delivery; however, the Foley catheter-dinoprostone combination was associated with a longer duration of delivery compared with the Foley catheter-oxytocin and Foley catheter-misoprostol combinations (p=0.05).¹¹ It should be noted that the randomization criteria did not consider parity and term of pregnancy (>24 weeks of gestation), and it was not mentioned whether the volume of fluid used to fill the Foley balloon was taken into account. There were significantly fewer patients who received dinoprostone compared with the other study groups.

The Foley catheter and Cook balloon are the most commonly used mechanical methods for induction of labor. Liu et al. (2019) conducted a meta-analysis on 1326 articles including 7 randomized clinical trials and 1159 patients.¹² The primary result for comparison was the cesarean section rate, and the secondary outcomes were effectiveness, efficiency, safety, and patient satisfaction.¹² No statistically significant differences for the primary outcome (RR: 0.88 (95% CI: 0.65–1.2)) or secondary outcomes were noted.¹²

A meta-analysis by Zhu et al. (2018) compared a Foley catheter and dinoprostone for labor induction.¹³ The primary outcome was the time from the application of the method to delivery, and the secondary outcomes were maternal (uterine hyperstimulation, postpartum hemorrhage, and infectious complications) and neonatal (1- and 5-min APGAR scores) conditions.¹³ Eight randomized clinical trials included 1191 patients in labor who were induced with a Foley catheter and 1199 patients who received dinoprostone.¹³ No statistically significant difference was observed for the duration of labor based on the induction method (mean difference: 0.71 h (95% CI: -2.5 to 3.91, p=0.67)) or for cesarean delivery rate (RR 0.91 (95% CI: 0.78–1.07, p=0.24)).¹³ Additionally, the rates of maternal infectious complications, postpartum hemorrhage, uterine hyperstimulation, and fetal outcomes did not show statistically significant differences.¹³

The Foley catheter has clear advantages over dinoprostone in terms of wide availability, simple storage, and low cost.¹³ Low-volume Foley balloons (30 mL) and high doses of dinoprostone (≥6 mg) may provide the shortest time from triggering labor to delivery.¹³

A pilot study conducted by the author in 2017 at Brugmann University Hospital, Brussels, on a cohort of 68 patients with a scarred uterus and 61 patients without a scarred uterus, triggered with a Foley catheter and oxytocin, revealed that 69% of the patients with scarred uterus managed to give birth vaginally while 70.5% of the control group had vaginal births. Thus, the Foley catheter-oxytocin combination also represents an interesting option for inducing labor in patients with a scarred uterus after a cesarean section.

Conclusions

Full-term pregnancies tend to become high-risk pregnancies, as it is often necessary for the birth to be induced artificially.¹⁴ Scientific studies have not yet identified an ideal method of artificial labor induction, a fact that offers an interesting track for scientific research.

Patients with overdue pregnancy have a significant risk of antenatal fetal mortality,¹⁴ which is why, according to simple rules in obstetrics, after exceeding the due date for delivery by 7–10 days, pregnant women are more carefully monitored (every 48 hours) to opportunistically induce labor using the appropriate method, if necessary.