ABSTRACT
Aim
To compare maternal and neonatal outcomes of induction strategies initiated with slow‐release vaginal dinoprostone (10 mg) or balloon catheter in multiparous women undergoing term induction of labor with an unfavorable cervix.
Methods
This retrospective cohort study included 802 multiparous women with singleton pregnancies at ≥ 37 weeks' gestation, Bishop score < 8, and a medical indication for induction of labor at a tertiary university‐affiliated center (2020–2025). Women underwent cervical ripening with either a balloon catheter (n = 522) or a slow‐release vaginal dinoprostone insert (n = 280). The co‐primary outcomes were the cervical ripening‐to‐delivery interval and delivery within 24 h. Multivariable analyses identified independent predictors of delivery interval and cesarean birth.
Results
Compared with balloon catheter, dinoprostone was associated with a shorter cervical ripening‐to‐delivery interval (11.6 vs. 19.4 h; p < 0.001). Delivery within 12 h was more frequent with dinoprostone (51.4% vs. 26.1%; p < 0.001), whereas delivery within 24 h did not differ significantly between groups (75.7% vs. 71.1%; p = 0.183). Dinoprostone was also associated with a higher spontaneous vaginal delivery rate (95.0% vs. 90.0%; p = 0.015) and a lower cesarean delivery rate (3.2% vs. 8.2%; p = 0.006). Intrapartum fever occurred less frequently with dinoprostone (1.4% vs. 4.2%; p = 0.036), while no statistically significant differences were observed in measured neonatal outcomes, including NICU admission.
Conclusions
Among multiparous women undergoing term induction of labor, dinoprostone was associated with a shorter cervical ripening‐to‐delivery interval, higher spontaneous vaginal delivery rates, and lower cesarean rates than balloon catheter. Delivery within 24 h was similar between groups, with no significant differences in measured neonatal outcomes.
Keywords: balloon catheter, cesarean delivery, dinoprostone, induction of labor, vaginal delivery
1. Introduction
Induction of labor (IOL) is one of the most common interventions in obstetrics, with global rates approaching 25% [1]. In cases of unfavorable cervix, cervical ripening is often required. This is achieved either pharmacologically—primarily using prostaglandins such as prostaglandin E1 (PGE1) or prostaglandin E2 (PGE2)—or mechanically, most often with single‐ or double‐balloon catheters [2]. Both mechanical and pharmacological interventions are considered safe, but consensus is lacking on which approach yields the most favorable maternal and neonatal outcomes, particularly when stratified by parity [3]. Few trials or cohort studies have directly compared catheter‐ versus prostaglandin‐based IOL in nulliparous and multiparous women separately, despite clear biological and clinical differences between these groups [4, 5].
Nulliparous women typically experience higher rates of IOL, reported between 34% and 43% in the United States, whereas among multiparas the rates are closer to 31% [6]. Although IOL frequency varies by region, institution, and evolving practice guidelines, the consistent finding is that first‐time mothers are induced more often than those with prior vaginal births [7, 8]. This trend may have contributed to the greater research focus on IOL methods in nulliparas, with fewer studies specifically addressing outcomes among multiparous women [9, 10, 11]. Overall, previous heterogeneous studies including both nulliparas and multiparas have reported comparable vaginal delivery rates within 24 h when cervical ripening is performed with either balloon catheters or PGE2 agents [12, 13]. According to previous studies, compared to PGE2, mechanical ripening with balloon catheters has been associated with a 20% reduction in adverse perinatal outcomes—such as neonatal intensive care unit (NICU) admissions, fetal acidosis, and < 7 Apgar scores—and a 65% lower incidence of tachysystole [13, 14, 15].
Despite these insights, the literature is still limited by the lack of studies directly comparing mechanical and pharmacological induction methods in multiparous patients while accounting for the Bishop score prior to cervical ripening [14, 16]. In our department, cervical ripening is initiated using either a balloon catheter or PGE2 (slow‐release vaginal dinoprostone 10 mg), followed by subsequent induction management according to our institutional protocol. The aim of this study was to compare maternal and neonatal outcomes of induction strategies initiated with balloon catheter or dinoprostone among multiparous women with at least one previous vaginal delivery.
2. Methods
2.1. Study Design
This was a retrospective cohort study conducted at a single tertiary university‐affiliated medical center between March 2020 and January 2025. The study included multiparous women (with at least one previous vaginal delivery) with a singleton fetus in cephalic presentation who underwent IOL at term (≥ 37 weeks of gestation) for a medical indication with an unfavorable cervix (Bishop score < 8). Exclusion criteria were nulliparity, previous cesarean delivery, multiple gestations, major fetal anomalies, non‐cephalic presentation, intrauterine fetal demise, prelabor rupture of membranes, or contraindications to vaginal delivery. Data were extracted from the hospital's electronic medical records system. The study was approved by the Institutional Review Board of Galilee Medical Center (approval number 0256‐24‐NHR).
2.2. Protocol and Definitions
The exposure of interest was the method of cervical ripening, either (1) single balloon catheter or (2) slow‐release vaginal dinoprostone 10 mg. Both approaches were available during the study period. Eligible patients were generally counseled regarding both cervical ripening methods, and the choice was made based on patient preference and clinical considerations. Dinoprostone was generally avoided in women with ongoing uterine contractions, in whom balloon catheter ripening was preferred. In the balloon catheter group, cervical ripening was performed by introducing a balloon catheter under direct visualization with a vaginal speculum. The balloon was inflated with 60 mL of sterile water, its position was confirmed by vaginal examination, and the external end was secured to the maternal thigh under gentle traction [17]. The catheter was routinely extracted at 12 h unless it was expelled earlier. In the dinoprostone group, a 10‐mg slow‐release vaginal insert was placed in the posterior fornix. Patients were examined in case of contractions. The dinoprostone insert was left in place for a maximum of 24 h. Earlier removal of the balloon catheter or dinoprostone insert, defined as extraction before 12 h, was undertaken in cases of active labor, tachysystole, abnormal fetal heart rate patterns, vaginal bleeding, membrane rupture, or patient‐reported pain. According to our department protocol, all instances of earlier removal were documented. Amniotomy was performed at cervical dilatation of 3–4 cm or greater, provided the head was well applied to the cervix [18]. According to our department protocol, once the cervical ripening device (balloon catheter or dinoprostone insert) is removed, oxytocin infusion is initiated unless the patient is already in active labor [19].
2.3. Data Collection and Outcomes
Baseline maternal and obstetric characteristics included age, gravidity, parity, body mass index (BMI), comorbidities (diabetes mellitus, hypertension, thrombophilia), and gestational age at IOL. Indications for labor induction were classified as fetal or maternal and were further categorized according to the specific indication. Fetal indications included fetal heart rate decelerations, fetal growth restriction, postterm pregnancy, decreased fetal movements, oligohydramnios, and other fetal indications. Maternal indications included gestational diabetes, hypertensive disorders of pregnancy, intrahepatic cholestasis of pregnancy, thrombophilia, and other maternal indications. Bishop score before and after ripening, need for oxytocin augmentation, and need for additional ripening were recorded. The co‐primary outcomes were cervical ripening to delivery interval (hours), and delivery within 24 h. Secondary outcomes included delivery within 12 h, mode of delivery (spontaneous vaginal, vacuum‐assisted vaginal, or cesarean delivery), duration of cervical ripening, intrapartum fever, indications for cesarean (non‐reassuring fetal heart rate or lack of progress), postpartum complications (fever, hemorrhage), and maternal length of stay. Neonatal outcomes included birthweight, Apgar score at 5 min, NICU admission, neonatal fever, and length of neonatal hospitalization (days).
2.4. Statistical Analysis
Continuous variables were expressed as mean ± standard deviation (SD) or median (interquartile range [IQR]) and compared using Student's t‐test or Mann–Whitney U test, as appropriate. Categorical variables were expressed as n (%) and compared using chi‐square or Fisher's exact test. The distributions of specific fetal and maternal indications for labor induction were compared between the cervical ripening groups using the chi‐square test. Multivariate analyses were conducted to identify independent predictors of the cervical ripening to delivery interval and of cesarean delivery. For the continuous outcome (cervical ripening‐to‐delivery interval), multivariable linear regression was applied, and unstandardized regression coefficients (B) with 95% confidence intervals (CIs) were reported. The model included maternal age, BMI, parity, cervical ripening method (dinoprostone vs. balloon catheter), gestational age at induction, Bishop score at initiation of cervical ripening, and indication for labor induction (fetal vs. maternal). For the dichotomous outcome (cesarean delivery), multivariable logistic regression was performed, with results expressed as odds ratios (OR) and 95% CIs. Given the limited number of cesarean deliveries, a parsimonious model was constructed using prespecified clinically relevant baseline covariates, including maternal age, BMI, parity, cervical ripening method, and indication for labor induction (fetal vs. maternal). Covariates were selected based on their clinical relevance and potential to confound the association between cervical ripening method and the study outcomes. Statistical significance was defined as a two‐tailed p < 0.05.
2.5. Sample Size Calculation
Assuming a 10% difference in delivery within 24 h between groups, with 80% power and a two‐sided α of 5%, the required sample size was 771 participants. Our final cohort exceeded this requirement.
3. Results
3.1. Baseline Characteristics
A total of 802 multiparous women were included: 522 (65.1%) in the balloon catheter group and 280 (34.9%) in the dinoprostone group. Maternal age, BMI, parity, maternal comorbidities, gestational age at induction, and gestational age at delivery did not differ significantly between the groups (Table 1). Fetal indications for labor induction were present in 378/522 (72.4%) women in the balloon catheter group and 197/280 (70.4%) in the dinoprostone group. The distribution of specific fetal indications did not differ significantly between the groups (p = 0.624). Maternal indications were present in 144/522 (27.6%) and 83/280 (29.6%) women, respectively, with no significant difference in the distribution of specific maternal indications (p = 0.679) (Table 1).
TABLE 1.
Baseline and obstetric characteristics.
| Variables | Balloon catheter (n = 522) | Dinoprostone (n = 280) | p |
|---|---|---|---|
| Maternal age (years) | 31.0 (27.6–34.1) | 31.0 (27.8–34.6) | 0.371 |
| BMI (kg/m2) | 30.1 (27.0–33.5) | 31.2 (28.3–34.7) | 0.750 |
| BMI > 30 kg/m2 | 284 (54.4%) | 136 (48.6%) | 0.120 |
| Parity | 2.0 (2.0–3.0) | 2.0 (2.0–3.0) | 0.084 |
| Gestational age at delivery (weeks) | 40.1 (39.1–40.5) | 39.6 (39.1–40.4) | 0.290 |
| Assisted reproduction | 25 (4.8%) | 16 (5.7%) | 0.571 |
| Diabetes mellitus | 74 (14.2%) | 30 (10.7%) | 0.164 |
| Hypertension | 31 (5.9%) | 10 (3.6%) | 0.179 |
| Gestational age at induction (weeks) | 39.5 ± 1.1 | 39.5 ± 1.0 | 0.509 |
| Fetal indication for labor induction | 378 (72.4%) | 197 (70.4%) | 0.624 |
| Fetal heart rate decelerations | 76 (14.6%) | 38 (13.6%) | |
| Fetal growth restriction | 28 (5.4%) | 14 (5.0%) | |
| Postterm pregnancy | 115 (22.0%) | 51 (18.2%) | |
| Decreased fetal movements | 70 (13.4%) | 34 (12.1%) | |
| Oligohydramnios | 32 (6.1%) | 21 (7.5%) | |
| Other fetal indication | 57 (10.9%) | 39 (13.9%) | |
| Maternal indication for labor induction | 144 (27.6%) | 83 (29.6%) | 0.679 |
| Gestational diabetes | 55 (10.5%) | 29 (10.4%) | |
| Hypertensive disorders of pregnancy | 22 (4.2%) | 9 (3.2%) | |
| Intrahepatic cholestasis of pregnancy | 12 (2.3%) | 7 (2.5%) | |
| Thrombophilia | 14 (2.7%) | 7 (2.5%) | |
| Other maternal indication | 41 (7.9%) | 31 (11.1%) |
Note: Data are presented as n (%), mean ± standard deviation, or median (interquartile range), as appropriate. p values for fetal and maternal indications represent the overall comparison of the distribution of the specific indications between the balloon catheter and dinoprostone groups.
Abbreviation: BMI, body mass index.
3.2. Primary Outcomes
The cervical ripening‐to‐delivery interval was significantly shorter in the dinoprostone group (11.6 (6.0–23.7) vs. 19.4 (13.9–25.4) h, p < 0.001). Delivery within 24 h was similar between the dinoprostone and balloon catheter groups (212/280 (75.7%) vs. 371/522 (71.1%), respectively; p = 0.183).
3.3. Cervical Ripening and Labor Progression
Initial Bishop scores were similar between groups (median 2 (1–3) in both groups; p = 0.360) (Table 2). After cervical ripening, Bishop scores were higher in the dinoprostone group than in the balloon catheter group (median 6 (4–7) vs. 5 (4–7); p < 0.001), with a greater proportion of women achieving a Bishop score > 8 (56/280 (20.0%) vs. 46/522 (8.8%); p < 0.001). Early device removal occurred more frequently in the dinoprostone group (146/280 (52.1%) vs. 47/522 (9.0%); p < 0.001). Reasons for early removal included maternal pain (34/280 (12.1%) vs. 17/522 (3.3%); p < 0.001), tachysystole (26/280 (9.3%) vs. 0/522 (0.0%); p < 0.001), fetal heart rate decelerations (20/280 (7.1%) vs. 8/522 (1.5%); p < 0.001), and labor progression (66/280 (23.6%) vs. 22/522 (4.2%); p < 0.001). Delivery within 12 h of cervical ripening initiation was more frequent in the dinoprostone group than in the balloon catheter group (144/280 (51.4%) vs. 136/522 (26.1%); p < 0.001), whereas delivery within 24 h was similar between the groups (212/280 (75.7%) vs. 371/522 (71.1%); p = 0.183). The interval from device removal to delivery was also shorter with dinoprostone (3.8 (1.4–11.1) vs. 8.8 (5.6–15.3) hours; p < 0.001). Oxytocin use and amniotomy were less frequent following dinoprostone than balloon catheter ripening (40.4% vs. 73.2%; p < 0.001 and 50.7% vs. 62.3%; p = 0.002, respectively).
TABLE 2.
Bishop scores, cervical ripening outcomes, and labor progression by method of induction.
| Variable | Balloon catheter (n = 522) | Dinoprostone (n = 280) | p |
|---|---|---|---|
| Bishop score—before insertion | 2 (1–3) | 2 (1–3) | 0.360 |
| Bishop score—after removal | 5 (4–6) | 6 (4–7) | < 0.001 |
| Bishop delta (Δ score) | 3 (2–4) | 3.5 (2–5) | < 0.001 |
| Bishop score ≥ 8 after removal | 46 (8.8%) | 56 (20.0%) | < 0.001 |
| Time interval from start of cervical ripening to delivery (hours) | 19.4 (13.9–25.4) | 11.6 (6.0–23.7) | < 0.001 |
| Delivery within 24 h of cervical ripening | 371 (71.1%) | 212 (75.7%) | 0.183 |
| Delivery within 12 h of cervical ripening | 136 (26.1%) | 144 (51.4%) | < 0.001 |
| Interval from extraction of balloon catheter/dinoprostone to delivery (h) | 8.8 (5.6–15.3) | 3.8 (1.4–11.1) | < 0.001 |
| Duration of cervical ripening (h) | 11.3 (6.4–12.2) | 9.8 (5.2–12.4) | 0.312 |
| Early removal of cervical ripening device | 47 (9.0%) | 146 (52.1%) | < 0.001 |
| Due to pain | 17 (3.3%) | 34 (12.1%) | < 0.001 |
| Due to tachysystole | 0 (0.0%) | 26 (9.3%) | < 0.001 |
| Due to fetal heart rate decelerations | 8 (1.5%) | 20 (7.1%) | < 0.001 |
| Due to labor progression | 22 (4.2%) | 66 (23.6%) | < 0.001 |
| Need for oxytocin | 382 (73.2%) | 113 (40.4%) | < 0.001 |
| Need for another cervical method | 37 (7.1%) | 18 (6.4%) | 0.771 |
| Amniotomy | 325 (62.3%) | 142 (50.7%) | 0.002 |
Note: Data are presented as n (%) or median (interquartile range), as appropriate.
3.4. Maternal Outcomes
Epidural use was similar between the groups (Table 3). Spontaneous vaginal delivery was more frequent after dinoprostone (95.0% vs. 90.0%, p = 0.015), while cesarean delivery was less common (3.2% vs. 8.2%, p = 0.006). The median interval from membrane rupture to delivery was shorter with dinoprostone (1.25 vs. 2.15 h, p < 0.001). Intrapartum fever occurred less frequently after dinoprostone (1.4% vs. 4.2%, p = 0.036). Postpartum fever, postpartum hemorrhage, and maternal hospitalization length did not differ significantly between groups.
TABLE 3.
Maternal and neonatal delivery outcomes by method of induction.
| Variables | Balloon catheter (n = 522) | Dinoprostone (n = 280) | p |
|---|---|---|---|
| Epidural anesthesia | 399 (76.4%) | 211 (75.4%) | 0.729 |
| Spontaneous vaginal delivery | 470 (90.0%) | 266 (95.0%) | 0.015 |
| Vacuum‐assisted vaginal delivery | 9 (1.7%) | 5 (1.8%) | 1.000 |
| Cesarean delivery | 43 (8.2%) | 9 (3.2%) | 0.006 |
| Indications for cesarean delivery | |||
| Non‐reassuring fetal heart rate | 26/43 (60.5%) | 8/9 (88.9%) | 0.139 |
| Non‐progressive labor | 12/43 (27.9%) | 1/9 (11.1%) | 0.420 |
| Malpresentation diagnosed during induction | 3/43 (7.0%) | 0/9 (0.0%) | 1.000 |
| Patient request during prolonged induction | 2/43 (4.7%) | 0/9 (0.0%) | 1.000 |
| Membrane rupture‐to‐delivery interval (hours) | 2.15 (0.8–4.6) | 1.25 (0.3–2.9) | < 0.001 |
| Intrapartum fever | 22 (4.2%) | 4 (1.4%) | 0.036 |
| Postpartum fever | 5 (1.0%) | 8 (2.9%) | 0.073 |
| Postpartum hemorrhage | 29 (5.6%) | 9 (3.2%) | 0.164 |
| Maternal hospitalization length (days) | 2 (2–2) | 1 (1.7–2) | 0.125 |
| Birthweight (g) | 3360.0 (3054.0–3645.0) | 3415.0 (3100.0–3664.0) | 0.157 |
| 5‐min Apgar score < 7 | 3 (0.6%) | 0 (0.0%) | 0.555 |
| NICU admission | 16 (3.1%) | 11 (3.9%) | 0.541 |
| Neonatal fever | 7 (1.3%) | 0 (0.0%) | 0.103 |
| Neonatal hospitalization length (days) | 2 (2–3) | 2 (1.8–2.3) | 0.818 |
Note: Data are presented as n (%) or median (interquartile range), as appropriate.
Abbreviation: NICU, neonatal intensive care unit.
3.5. Neonatal Outcomes
No statistically significant differences were observed between the groups in 5‐min Apgar score < 7, NICU admission, neonatal fever, or neonatal hospitalization length (Table 3).
3.6. Multivariate Analysis
In the multivariable linear regression analysis (Table 4), an induction strategy initiated with dinoprostone remained associated with a shorter cervical ripening‐to‐delivery interval after adjustment for maternal age, BMI, parity, gestational age at induction, Bishop score at initiation of cervical ripening, and indication for labor induction (B = −5.19 h, 95% CI −7.57 to −2.82; p < 0.001). A higher Bishop score at initiation of cervical ripening was also associated with a shorter interval (B = −1.78 h, 95% CI −2.83 to −0.73; p = 0.001). Maternal age, BMI, parity, gestational age at induction, and indication for labor induction were not significantly associated with the cervical ripening‐to‐delivery interval. In the multivariable logistic regression analysis, an induction strategy initiated with dinoprostone remained associated with lower odds of cesarean delivery compared with balloon catheter (OR 0.22, 95% CI 0.08–0.59; p = 0.002). Increasing maternal age was associated with higher odds of cesarean delivery (OR 1.15, 95% CI 1.06–1.24; p < 0.001), whereas higher parity was associated with lower odds (OR 0.49, 95% CI 0.28–0.86; p = 0.013). BMI and indication for labor induction were not significantly associated with cesarean delivery (Table 5).
TABLE 4.
Multivariable linear regression analysis of factors associated with cervical ripening‐to‐delivery interval.
| Variables | B (hours) | Standard error | 95% CI | p |
|---|---|---|---|---|
| Maternal age (years) | 0.14 | 0.13 | −0.11 to 0.40 | 0.271 |
| BMI (kg/m2) | 0.13 | 0.11 | −0.09 to 0.35 | 0.258 |
| Parity (number of births) | −1.18 | 0.85 | −2.86 to 0.50 | 0.168 |
| Gestational age at induction (weeks) | −0.98 | 0.55 | −2.07 to 0.10 | 0.075 |
| Bishop score at initiation of cervical ripening | −1.78 | 0.54 | −2.83 to −0.73 | 0.001 |
| Maternal vs. fetal indication for induction | 2.63 | 1.35 | −0.02 to 5.29 | 0.052 |
| Dinoprostone vs. balloon catheter | −5.19 | 1.21 | −7.57 to −2.82 | < 0.001 |
Note: Data are presented as unstandardized regression coefficients (B), standard errors, and 95% CIs. Reference categories are fetal indication for labor induction and balloon catheter.
Abbreviations: BMI, body mass index; CI, confidence interval.
TABLE 5.
Multivariable logistic regression analysis of factors associated with cesarean delivery.
| Variable | Odds ratio | 95% CI | p |
|---|---|---|---|
| Maternal age (years) | 1.15 | 1.06–1.24 | < 0.001 |
| BMI (kg/m2) | 1.05 | 0.99–1.13 | 0.116 |
| Parity (number of births) | 0.49 | 0.28–0.86 | 0.013 |
| Induction method (dinoprostone vs. balloon catheter) | 0.22 | 0.08–0.59 | 0.002 |
| Indication for labor induction (maternal vs. fetal) | 0.80 | 0.36–1.76 | 0.573 |
Note: Data are presented as ORs and 95% CIs. Reference categories are balloon catheter and fetal indication for labor induction.
Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio.
4. Discussion
4.1. Main Findings
Among 802 multiparous patients, the induction strategy initiated with dinoprostone was associated with a shorter cervical ripening‐to‐delivery interval and similar rates of delivery within 24 h compared with the strategy initiated with balloon catheter. Delivery within 12 h was more frequent following dinoprostone initiation, and a favorable post–cervical ripening Bishop score was achieved more frequently in this group. The dinoprostone‐initiated strategy was also associated with less frequent use of oxytocin and amniotomy, higher spontaneous vaginal delivery rates, and fewer cesarean deliveries.
4.2. Comment
The existing literature on cervical ripening methods presents conflicting results, with some studies favoring mechanical approaches and others pharmacological agents [9, 11, 13, 14, 16, 20, 21, 22, 23, 24]; however, few studies specifically address multiparous women and directly demonstrate differences in this population [12]. Interestingly, most previous studies [9, 10, 11] that stratified outcomes by parity have focused on nulliparous patients. For example, a study of 330 nulliparas [11] also reported a shorter cervical ripening to delivery interval with dinoprostone compared to balloon catheter ripening. The higher rate of delivery within 12 h observed following dinoprostone should be interpreted in the context of the institutional induction protocols. The balloon catheter was routinely maintained for up to 12 h, whereas dinoprostone could promote uterine activity and labor progression, leading to earlier removal and differences in the timing of subsequent interventions. Therefore, the observed differences reflect the overall induction strategies initiated with dinoprostone or balloon catheter rather than the isolated effects of the cervical ripening methods. Consistent with this consideration, rates of delivery within 24 h were similar between the two strategies (71.1% with balloon catheter vs. 75.7% with dinoprostone; p = 0.183), in line with previous reports in the literature [25]. We observed a longer cervical ripening‐to‐delivery interval and a higher overall cesarean delivery rate following the balloon catheter‐initiated strategy. The shorter interval observed following dinoprostone initiation may reflect both its pharmacological effects on cervical ripening and uterine contractility and differences in protocol‐driven induction management, including the timing of device removal and subsequent interventions. Therefore, these findings should be interpreted in the context of the overall induction strategies rather than as evidence of the isolated effects of either cervical ripening method.
Our results are further supported by a large randomized trial of 1220 patients undergoing IOL [24], which, similar to our study, demonstrated a shorter time to delivery with dinoprostone, and comparable cesarean rates for non‐reassuring fetal status between the groups. In that study [24] interaction analysis showed that parity did not affect the results. Previous meta‐analyses comparing balloon catheters and PGE2 did not account for parity and generally reported similar IOL outcomes [14, 16]. One meta‐analysis [14] of nine randomized controlled trials including 1866 patients compared balloon catheter with prostaglandin E2 for cervical ripening. Obstetric outcomes, including cesarean delivery rates (RR 0.92; 95% CI 0.79–1.07) and vaginal delivery within 24 h (RR 0.95; 95% CI 0.78–1.16), were comparable between groups. Neonatal outcomes were also similar, although the prostaglandin group demonstrated higher rates of uterine tachysystole (13.8%) and NICU admission (16.4%) [14]. In our study, the rate of tachysystole with dinoprostone was 9.3%, while NICU admission rates were low in both groups (3.9% with dinoprostone vs. 3.1% with balloon catheter; p = 0.541). In the dinoprostone group, the lower need for oxytocin augmentation, the higher rate of spontaneous membrane rupture, and the shorter interval from device removal to delivery were consistent with earlier progression to active labor. This observation is consistent with a randomized controlled trial [21] comparing outpatient balloon catheter with inpatient prostaglandin among a heterogeneous population of nulliparous and multiparous women, which similarly reported lower rates of oxytocin use and amniotomy in the dinoprostone group. The biological plausibility of this finding is supported by the central role of prostaglandins in the initiation and progression of labor, as they mediate both myometrial contractility and the biophysical changes associated with cervical ripening [26].
In our cohort, intrapartum fever was more frequent following balloon catheter ripening compared with dinoprostone. This finding is consistent with data from a recent large randomized controlled trial of 1860 women [23], which demonstrated higher rates of suspected intrapartum and postpartum infection among patients induced with a balloon catheter compared to prostaglandins. Fever is an infrequent adverse effect of dinoprostone, reported in less than 2% of patients, with minimal difference from placebo in clinical trials [27]. Some studies, however, suggest that dinoprostone may be associated with increased cytokine release and an enhanced inflammatory response, which could contribute to the development of intrapartum fever [28].
A notable feature of dinoprostone use was the high rate of early insert removal (52.1%). The most frequent reason was labor progression (23.6%), followed by maternal pain (12.1%), tachysystole (9.3%), and fetal heart rate decelerations (7.1%). In contrast, early balloon catheter removal occurred in 9.0% of patients. These differences should be interpreted in the context of the distinct clinical effects and institutional protocols associated with the two cervical ripening strategies. Moreover, in this cohort, early device removal due to fetal heart rate decelerations was significantly more common in the dinoprostone group compared with the balloon catheter group (7.1% vs. 1.5%, p < 0.001). This finding is consistent with previous reports linking prostaglandins to uterine hyperstimulation and subsequent fetal heart rate changes. A recent individual patient data meta‐analysis [13], which included over 5400 women from 12 randomized trials, found that prostaglandins were associated with a higher risk of composite adverse perinatal outcomes. In this analysis, NICU admission rates and Apgar scores < 7 at 5 min were similar between groups. No statistically significant differences were observed in the measured neonatal outcomes; however, the study was not powered to detect differences in uncommon neonatal adverse outcomes.
4.3. Strengths and Limitations
This study has several strengths. First, it focuses exclusively on multiparous women with at least one prior vaginal birth. Second, detailed clinical data were available, including Bishop scores before and after ripening, device‐related outcomes, and maternal and neonatal complications. However, several limitations should be noted. The retrospective, single‐center design may limit generalizability and introduce potential selection bias. Because the cervical ripening method was not randomly assigned, confounding by indication cannot be excluded. Although we adjusted for measured baseline characteristics and key potential confounders in the multivariable analyses, residual confounding due to unmeasured factors may remain. We did not assess patient satisfaction and pain scores, which may have influenced cervical ripening method selection. Although device removal events were carefully documented, the retrospective design could still allow underreporting of certain complications. In addition, the study was not powered to detect differences in uncommon neonatal adverse outcomes; therefore, the absence of statistically significant differences in measured neonatal outcomes should not be interpreted as evidence of equivalent neonatal safety. Furthermore, because subsequent induction management was influenced by the initial cervical ripening method and labor progression, the observed outcomes represent the overall induction strategies and cannot be attributed solely to the isolated effects of dinoprostone or balloon catheter.
In multiparous women undergoing IOL, an induction strategy initiated with dinoprostone was associated with a shorter cervical ripening‐to‐delivery interval, higher spontaneous vaginal delivery rates, and fewer cesarean deliveries compared with a strategy initiated with balloon catheter. Delivery within 24 h was similar between the two strategies, and no significant differences were observed in the measured neonatal outcomes.
Author Contributions
Raneen Abu Shqara: conceptualization, investigation, visualization, validation, formal analysis, project administration, writing – review and editing, writing – original draft, methodology, data curation. Ruba Maklade: validation, methodology, software, writing – original draft, investigation. Lilian Haj: investigation, writing – original draft, methodology, validation, software. Nadir Ganem: supervision, writing – review and editing, writing – original draft, visualization, validation, conceptualization. Lior Lowenstein: conceptualization, writing – review and editing, supervision, project administration, data curation, visualization, validation. Maya Frank Wolf: conceptualization, writing – review and editing, writing – original draft, project administration, supervision.
Funding
The authors have nothing to report.
Disclosure
The authors have nothing to report.
Ethics Statement
This study was approved by the Institutional Review Board (Helsinki Committee) of Galilee Medical Center (Approval No. NHR‐0256‐24). The study was conducted in accordance with the Declaration of Helsinki and relevant national guidelines.
Consent
The requirement for informed consent was waived by the Institutional Review Board due to the retrospective nature of the study.
Conflicts of Interest
The authors declare no conflicts of interest.
Data Availability Statement
The datasets generated and analyzed during the current study are not publicly available because they were obtained from confidential patient medical records at Galilee Medical Center, Nahariya, Israel. Although the data were analyzed in a de‐identified format, the original datasets contain sensitive clinical and obstetric information. Public sharing of these data could potentially compromise patient confidentiality and would require additional institutional authorization beyond the scope of the original ethical approval (IRB Approval No. NHR‐0256‐24).
References
- 1. Swift E. M., Gunnarsdottir J., Zoega H., Bjarnadottir R. I., Steingrimsdottir T., and Einarsdottir K., “Trends in Labor Induction Indications: A 20‐Year Population‐Based Study,” Acta Obstetricia et Gynecologica Scandinavica 101 (2022): 1422–1430, 10.1111/aogs.14447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Lindblad Wollmann C., Ahlberg M., Petersson G., Wollmann C. L., Saltvedt S., and Stephansson O., “Time‐To‐Delivery and Delivery Outcomes Comparing Three Methods of Labor Induction in 7551 Nulliparous Women: A Population‐Based Cohort Study,” Journal of Perinatology 37 (2017): 1197–1203, 10.1038/jp.2017.122. [DOI] [PubMed] [Google Scholar]
- 3. Denona B., Foley M., Mahony R., and Robson M., “Discrimination by Parity Is a Prerequisite for Assessing Induction of Labour Outcome – Cross‐Sectional Study,” BMC Pregnancy and Childbirth 20 (2020): 709, 10.1186/s12884-020-03334-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Beckmann M., Gibbons K., Flenady V., and Kumar S., “Induction of Labour Using Prostaglandin E2 as an Inpatient Versus Balloon Catheter as an Outpatient: A Multicentre Randomised Controlled Trial,” BJOG: An International Journal of Obstetrics and Gynaecology 127 (2020): 571–579, 10.1111/1471-0528.16030. [DOI] [PubMed] [Google Scholar]
- 5. Mizrachi Y., Levy M., Bar J., and Kovo M., “Induction of Labor in Nulliparous Women With Unfavorable Cervix: A Comparison of Foley Catheter and Vaginal Prostaglandin E2,” Archives of Gynecology and Obstetrics 294 (2016): 725–730, 10.1007/s00404-016-4026-9. [DOI] [PubMed] [Google Scholar]
- 6. Laughon S. K., Zhang J., Grewal J., Sundaram R., Beaver J., and Reddy U. M., “Induction of Labor in a Contemporary Obstetric Cohort,” American Journal of Obstetrics and Gynecology 206 (2012): 486.e1–486.e9, 10.1016/J.AJOG.2012.03.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Nicholson S. M., Hatt S., Kent E. M., et al., “The Rising Tide: Trends in Induction of Labor at Term Over a 5‐Year Period at a Single Centre,” International Journal of Gynecology & Obstetrics 169 (2025): 383–390, 10.1002/ijgo.16054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Lee N., Ballard E., and Humphrey T., “Trends in Induction of Labour and Associated Co‐Morbidities and Demographics in Queensland, Australia From 2001 to 2020: A Population‐Based Study,” BMC Pregnancy and Childbirth 25 (2025): 354, 10.1186/s12884-025-07379-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Tan P. C., Valiapan S. D., Tay P. Y. S., and Omar S. Z., “Concurrent Oxytocin With Dinoprostone Pessary Versus Dinoprostone Pessary in Labour Induction of Nulliparas With an Unfavourable Cervix: A Randomised Placebo‐Controlled Trial,” BJOG 114 (2007): 824–832, 10.1111/j.1471-0528.2007.01384.x. [DOI] [PubMed] [Google Scholar]
- 10. Facchinetti F., Venturini P., Verocchi G., and Volpe A., “Comparison of Two Preparations of Dinoprostone for Pre‐Induction of Labour in Nulliparous Women With Very Unfavourable Cervical Condition: A Randomised Clinical Trial,” European Journal of Obstetrics and Gynecology and Reproductive Biology 119 (2005): 189–193, 10.1016/j.ejogrb.2004.06.039. [DOI] [PubMed] [Google Scholar]
- 11. Pennell C. E., Henderson J. J., O'Neill M. J., et al., “Induction of Labour in Nulliparous Women With an Unfavourable Cervix: A Randomised Controlled Trial Comparing Double and Single Balloon Catheters and PGE2 Gel,” BJOG: An International Journal of Obstetrics and Gynaecology 116, no. 11 (2009): 1443–1452, 10.1111/j.1471-0528.2009.02279.x. [DOI] [PubMed] [Google Scholar]
- 12. Manly E., Hiersch L., Moloney A., et al., “Comparing Foley Catheter to Prostaglandins for Cervical Ripening in Multiparous Women,” Journal of Obstetrics and Gynaecology Canada 42 (2020): 853–860, 10.1016/j.jogc.2019.11.001. [DOI] [PubMed] [Google Scholar]
- 13. Jones M. N., Palmer K. R., Pathirana M. M., et al., “Balloon Catheters Versus Vaginal Prostaglandins for Labour Induction (CPI Collaborative): An Individual Participant Data Meta‐Analysis of Randomised Controlled Trials,” Lancet 400 (2022): 1681–1692, 10.1016/S0140-6736(22)01845-1. [DOI] [PubMed] [Google Scholar]
- 14. Du Y. M., Zhu L. Y., Cui L. N., et al., “Double‐Balloon Catheter Versus Prostaglandin E2 for Cervical Ripening and Labour Induction: A Systematic Review and Meta‐Analysis of Randomised Controlled Trials,” BJOG: An International Journal of Obstetrics and Gynaecology 124 (2017): 891–899, 10.1111/1471-0528.14256. [DOI] [PubMed] [Google Scholar]
- 15. de Vaan M. D. T., ten Eikelder M. L. G., Jozwiak M., et al., “Mechanical Methods for Induction of Labour,” Cochrane Database of Systematic Reviews 3, no. 3 (2023): CD001233, 10.1002/14651858.CD001233.pub4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Liu Y. R., Pu C. X., Wang X. Y., and Wang X. Y., “Double‐Balloon Catheter Versus Dinoprostone Insert for Labour Induction: A Meta‐Analysis,” Archives of Gynecology and Obstetrics 299 (2019): 7–12, 10.1007/s00404-018-4929-8. [DOI] [PubMed] [Google Scholar]
- 17. Schoen C. N., Saccone G., Backley S., et al., “Increased Single‐Balloon Foley Catheter Volume for Induction of Labor and Time to Delivery: A Systematic Review and Meta‐Analysis,” Acta Obstetricia et Gynecologica Scandinavica 97 (2018): 1051–1060, 10.1111/aogs.13353. [DOI] [PubMed] [Google Scholar]
- 18. Kim S. W., Nasioudis D., and Levine L. D., “Role of Early Amniotomy With Induced Labor: A Systematic Review of Literature and Meta‐Analysis,” American Journal of Obstetrics & Gynecology MFM 1 (2019): 100052, 10.1016/j.ajogmf.2019.100052. [DOI] [PubMed] [Google Scholar]
- 19. Berghella V., Bellussi F., and Schoen C. N., “Evidence‐Based Labor Management: Induction of Labor (Part 2),” American Journal of Obstetrics & Gynecology MFM 2 (2020): 100136, 10.1016/j.ajogmf.2020.100136. [DOI] [PubMed] [Google Scholar]
- 20. Rodríguez‐Zurita A., Álvarez S. C., García C. C., et al., “Effectiveness and Safety of the Double Intracervical Balloon vs Dinoprostone in Patients With Previous Cesarean Section,” Reproductive Sciences 31 (2024): 3148–3158, 10.1007/s43032-024-01617-5. [DOI] [PubMed] [Google Scholar]
- 21. Wise M. R., Thompson J. M. D., Battin M., et al., “Outpatient Balloon Catheter vs Inpatient Prostaglandin for Induction of Labor: A Randomized Trial,” American Journal of Obstetrics & Gynecology MFM 5 (2023): 100958, 10.1016/j.ajogmf.2023.100958. [DOI] [PubMed] [Google Scholar]
- 22. Wise M. R., Marriott J., Battin M., Thompson J. M. D., Stitely M., and Sadler L., “Outpatient Balloon Catheter vs Inpatient Prostaglandin for Induction of Labour (OBLIGE): A Randomised Controlled Trial,” Trials 21 (2020): 190, 10.1186/s13063-020-4061-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Liu X., Huang D., Liu Y., et al., “Vaginal Dinoprostone vs Foley Catheter for Induction of Labor at Term With an Unfavorable Cervix: An Open‐Label Randomized Controlled Trial,” Am J Obstet Gynecol MFM 6 (2024): 101436, 10.1016/j.ajogmf.2024.101436. [DOI] [PubMed] [Google Scholar]
- 24. Diguisto C., Le Gouge A., Arthuis C., et al., “Cervical Ripening in Prolonged Pregnancies by Silicone Double Balloon Catheter Versus Vaginal Dinoprostone Slow Release System: The MAGPOP Randomised Controlled Trial,” PLoS Medicine 18 (2021): e1003448, 10.1371/JOURNAL.PMED.1003448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Abu‐Zaid A., Khadawardi K., Al‐Matary A., et al., “Isosorbide Mononitrate Versus Dinoprostone for Cervical Ripening During Labor Induction: A Systematic Review and Meta‐Analysis of Randomized Controlled Trials,” Reproductive Sciences 30, no. 2 (2023): 380–389. [DOI] [PubMed] [Google Scholar]
- 26. O'Brien W. F., “The Role of Prostaglandins in Labor and Delivery,” Clinics in Perinatology 22 (1995): 973–984, 10.1016/S0095-5108(18)30265-3. [DOI] [PubMed] [Google Scholar]
- 27. National Library of Medicine , “Label: CERVIDIL‐Dinoprostone Insert,” DailyMed, June 2, 2023, https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=f3bc9031‐0620‐4a50‐b440‐f661cadfeb1d.
- 28. Burgess A. P. H., Katz J. E., Moretti M., and Lakhi N., “Risk Factors for Intrapartum Fever in Term Gestations and Associated Maternal and Neonatal Sequelae,” Gynecologic and Obstetric Investigation 82 (2017): 508–516, 10.1159/000453611. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and analyzed during the current study are not publicly available because they were obtained from confidential patient medical records at Galilee Medical Center, Nahariya, Israel. Although the data were analyzed in a de‐identified format, the original datasets contain sensitive clinical and obstetric information. Public sharing of these data could potentially compromise patient confidentiality and would require additional institutional authorization beyond the scope of the original ethical approval (IRB Approval No. NHR‐0256‐24).