Download PDF
Open Access
CC BY-NC-ND 4.0 · Geburtshilfe Frauenheilkd
DOI: 10.1055/a-2780-1607
GebFra Science
Original Article

Setting of Induction of Labour with Double-Balloon Catheter in Low-risk Pregnancies: Outpatient Versus Inpatient

Geburtseinleitung mit einem Doppelballonkatheter bei Niedrig-Risiko-Schwangerschaften: Vergleich von ambulantem und stationärem Setting

Authors

  • Sven Kehl

    1   Department of Obstetrics and Gynecology, LMU University Hospital, LMU Munich, Munich, Germany (Ringgold ID: RIN9183)

  • Anne Gernhardt

    2   Obstetrics and Gynecology, Erlangen University Hospital, Erlangen, Germany (Ringgold ID: RIN27168)

  • Christel Weiss

    3   Department of Medical Statistics and Biomathematics, University Medical Centre Mannheim, Mannheim, Germany (Ringgold ID: RIN36642)

  • Michael O. Schneider

    2   Obstetrics and Gynecology, Erlangen University Hospital, Erlangen, Germany (Ringgold ID: RIN27168)

  • Matthias W. Beckmann

    2   Obstetrics and Gynecology, Erlangen University Hospital, Erlangen, Germany (Ringgold ID: RIN27168)

  • Jutta Pretscher 

    2   Obstetrics and Gynecology, Erlangen University Hospital, Erlangen, Germany (Ringgold ID: RIN27168)
Further Information
Also available at
 
 PDF Download Permissions and Reprints

Abstract

Objectives

Induction of labour is one of the most frequently performed obstetric procedures. This analysis aimed to assess maternal and neonatal outcomes after labour induction with a double-balloon catheter in an outpatient versus inpatient setting in low-risk term pregnancies.

Methods

In this retrospective cohort study, 1010 women with singleton, low-risk term pregnancies undergoing labour induction with a double-balloon catheter were analysed. Of these, 394 received outpatient and 616 inpatient cervical ripening. The primary outcome was the caesarean section rate; secondary outcomes included the induction-to-delivery interval, vaginal delivery within 24 hours and neonatal parameters.

Results

Delivery modes were comparable between groups (p = 0.5524), but vaginal delivery within 24 hours occurred less frequently in the outpatient group (25.2% vs. 32.9%, p = 0.0201). Neonatal outcomes were significantly better in the outpatient cohort, with lower rates of arterial umbilical cord pH < 7.10 (2.0% vs. 5.5%, p = 0.0067) and NICU admissions (6.9% vs. 15.1%, p < 0.0001). Multiple logistic regression analysis identified nulliparity (OR = 10.545), maternal age (OR = 1.076), and birth weight (OR = 1.733) as significant predictors for caesarean section. Failure to achieve vaginal delivery within 24 hours was associated with higher BMI (OR = 1.080), nulliparity (OR = 4.191), and fetal macrosomia (OR = 3.675), while labour onset by balloon catheter alone reduced this risk (OR = 0.149). Outpatient induction was associated with a lower risk of arterial pH < 7.10 (OR = 0.352, p = 0.0088).

Conclusions

Outpatient cervical ripening with a double-balloon catheter is a safe and effective alternative to inpatient induction in low-risk term pregnancies, with comparable delivery outcomes and potential neonatal benefits.


Zusammenfassung

Zielsetzung

Die Geburtseinleitung stellt eines der am häufigsten durchgeführten Verfahren in der Geburtshilfe dar. Ziel dieser Analyse war es, die mütterlichen und neonatalen Outcomes nach einer Geburtseinleitung mit Doppelballonkatheter, die bei Niedrig-Risiko-Schwangerschaften in einem stationären oder ambulanten Setting durchgeführt wurde, zu evaluieren.

Methoden

In dieser retrospektiven Kohortenstudie wurden 1010 Frauen mit Niedrig-Risiko-Einlingsschwangerschaften, die sich einer Geburtseinleitung mit einem Doppelballonkatheter unterzogen, analysiert. Bei 394 der Frauen wurde die zervikale Reifung ambulant durchgeführt, und bei 616 der Frauen wurde sie stationär durchgeführt. Das primäre Outcome war die Sectio-Rate; sekundäre Outcomes waren Dauer zwischen Einleitung und Geburt, vaginale Entbindung innerhalb von 24 Stunden und neonatale Parameter.

Ergebnisse

Die Entbindungsarten in den 2 Gruppen waren vergleichbar (p = 0,5524), aber bei der ambulanten Gruppe fand eine vaginale Entbindung innerhalb von 24 Stunden weniger oft statt als in der stationären Gruppe (25,2% vs. 32,9%, p = 0,0201). Die neonatalen Ergebnisse waren signifikant besser in der ambulanten Kohorte, mit niedrigeren Raten von Nabelschnur-pH-Werten < 7,10 (2,0% vs. 5,5%, p = 0,0067) und Aufnahmen in einer neonatalen Intensivstation (6,9% vs. 15,1%, p < 0,0001). Bei der mehrfachen logistischen Regressionsanalyse wurden Nulliparität (OR = 10,545), mütterliches Alter (OR = 1,076) und Geburtsgewicht (OR = 1,733) als signifikante Prädiktoren für eine Sectio ermittelt. Keine vaginale Entbindung innerhalb von 24 Stunden nach Einleitung war mit einem höheren BMI (OR = 1,080), Nulliparität (OR = 4,191) und fetaler Makrosomie (OR = 3,675) assoziiert, wohingegen dieses Risiko durch eine Geburtseinleitung nur mithilfe eines Ballonkatheters reduziert wurde (OR = 0,149). Die ambulante Einleitung war mit einem niedrigeren Risiko von einem Nabelschnur-pH-Wert < 7,10 assoziiert (OR = 0,352, p = 0,0088).

Schlussfolgerung

Die ambulante zervikale Reifung mit einem Doppelballonkatheter stellt eine sichere und effektive Alternative zur stationären Geburtseinleitung bei Frauen mit Niedrig-Risiko-Schwangerschaften dar. Die Geburtsergebnisse und potenziellen neonatalen Vorteile waren vergleichbar.


Keywords

delivery - induction of labor - balloon catheter - setting - outpatient - inpatient

Schlüsselwörter

Entbindung - Geburtseinleitung - Ballonkatheter - Setting - ambulant - stationär

Introduction

Induction of labour (IOL), the process of artificially initiating labour when the benefits of delivering the baby outweigh the risks of continuing the pregnancy, is one of the most frequently performed obstetric procedures, occurring in up to one out of every four pregnancies [1]. Globally, the rates of IOL are steadily increasing. In Germany, the rate of IOL in 2023 reached 22.6 % [2]. If the cervix is not favourable for birth, the chances of a successful IOL can be enhanced through cervical ripening, which can be achieved using mechanical or pharmacological methods. Balloon catheters, including both single and double balloons, are an effective option for promoting cervical ripening and initiating labour in cases of an unripe cervix [3]. Balloon catheters have lower rates of uterine overstimulation compared to prostaglandins [1] [4] [5] [6] [7]. A sequential induction approach, beginning with balloon catheters and followed by prostaglandins, has shown particular effectiveness for primiparous women with an unripe cervix. Moreover, combining a double-balloon catheter with oral prostaglandins further enhances IOL outcomes, especially in term pregnancies without premature rupture of membranes [3] [8] [9] [10] [11] [12]. Induction of labour in nulliparous women or when cervical priming is required can be a prolonged process, lasting several hours to days [13]. Outpatient cervical ripening has emerged as an appealing option for managing this process and the setting of induction, whether outpatient or inpatient, has become a topic of significant clinical interest. Outpatient IOL offers potential advantages, such as reduced antenatal hospital stays, lower hospital occupancy, improved maternal comfort and autonomy and potentially lower healthcare costs [3] [13] [14] [15] [16]. However, concerns about the safety and efficacy of outpatient IOL, particularly regarding caesarean section rates, remain and data comparing the sequential use of a double-balloon catheter for IOL in an inpatient versus in an outpatient setting are scarce [16] [17].

The aim of this study was therefore to evaluate whether IOL with a double-balloon catheter in an outpatient setting leads to higher caesarean section rates compared to the inpatient setting, while also assessing other key maternal and neonatal outcome parameters.


Material and Methods

This historical cohort study was conducted at a single tertiary referral centre and included women with singleton, low-risk pregnancies in cephalic presentation at ≥ 37 + 0 weeks of gestation who underwent induction of labour (IOL) between 2015 and 2022. To improve clarity, all inclusion and exclusion criteria have been summarised in a dedicated table ([Table 1]). In brief, exclusion criteria comprised gestational diabetes, intrahepatic cholestasis of pregnancy, hypertensive disorders of pregnancy, fetal abnormalities on cardiotocography or Doppler ultrasound, oligohydramnios, fetal growth restriction, intrauterine fetal death, structural or chromosomal fetal anomalies, and placenta previa. Women with a prior caesarean section were also excluded.

Table 1 Inclusion and exclusion criteria of the study cohort.

Category

Inclusion criteria

Exclusion criteria

Pregnancy characteristics

Singleton pregnancy

Cephalic presentation

≥ 37 + 0 weeks of gestation

Planned induction of labour (2015–2022)

Gestational diabetes

Intrahepatic cholestasis of pregnancy

Hypertensive disorders of pregnancy

Oligohydramnios

Fetal growth restriction

Placenta previa

Prior caesarean section

Fetal findings

Normal cardiotocography

Normal Doppler

No structural findings

No chromosomal findings

Abnormal cardiotocography

Abnormal Doppler

Structural anomalies

Chromosomal anomalies

Intrauterine fetal death

Data were collected as part of routine clinical care and were fully anonymized for analysis. As a retrospective investigation, the study did not require informed consent, and ethical approval was obtained from the Clinical Ethics Committee (Ref. No. 23–480-Br).

Gestational age was determined according to current clinical guidelines, calculated from the first day of the last menstrual period and adjusted in cases of a discrepancy greater than seven days using the crown–rump length measured in the first-trimester ultrasound. Upon hospital admission for IOL, a comprehensive initial assessment was performed, including medical history, fetal ultrasound (for estimated fetal weight, assessment of amniotic fluid via the single deepest pocket method, and umbilical artery Doppler), and a digital cervical examination to assign the Bishop score [3].

In cases of an unfavourable cervix (Bishop score < 6), the use and procedure of double-balloon catheter placement were explained in detail, and patients were scheduled to return between 6:00 and 8:00 PM for insertion. Prior to placement, cardiotocography (CTG) was performed [18]. The double-balloon catheter (Cervical Ripening Balloon, CRB, Cook Medical) was inserted under direct visualization via sterile speculum examination, and 80 ml of saline was gradually infused into each balloon per manufacturer instructions. The catheter was secured to the inner thigh without traction. A second CTG was routinely performed after insertion.

Patients in the outpatient group received thorough written and verbal instructions before discharge, including 24-hour telephone access to medical personnel. They were instructed to contact the hospital immediately in the event of heavy vaginal bleeding, rupture of membranes, painful contractions occurring at intervals of five minutes or less, significant abdominal discomfort, or reduced fetal movements. Information regarding possible catheter expulsion at home was also provided, including reassurance that expulsion was expected and non-alarming if no symptoms were present. Patients were instructed to return to the hospital at 8:00 AM the following day for evaluation.

After a minimum dwell time of 12 hours, the CRB was removed, and cervical status was reassessed. If the cervix remained unfavourable, labour induction continued with oral misoprostol for up to three days. In the event of a favourable cervix (Bishop score > 6), intravenous oxytocin was administered under continuous CTG monitoring. Induction was continued until the onset of regular uterine contractions. Routine amniotomy was not performed.

Following prostaglandin administration, CTG was performed for 60 minutes and repeated two hours later.

The primary outcome parameter was the rate of caesarean section. Secondary outcomes included mode of vaginal delivery (spontaneous or operative), induction-to-delivery interval, vaginal delivery within 24, 48, and 72 hours, and onset of labour induced by balloon catheter alone. Neonatal outcome measures included arterial umbilical cord pH < 7.10, base excess < −12, 5-minute Apgar score < 7, combined endpoint of pH < 7.00 and base excess ≤ −12, abnormal CTG, performance and result of fetal blood sampling, shoulder dystocia, admission to a neonatal intensive care unit (NICU), and neonatal death.


Statistical Analysis

All statistical calculations have been done with SAS softer, release 9.4 (SAS Institute Inc., Cary, North Carolina, USA). For qualitative factors, absolute frequencies and percentages are given. For quantitative variables, approximately normally distributed, mean values and standard deviations have been calculated. For ordinally scaled variables median together with minimum and maximum are given. In order to compare two groups regarding a qualitative factor, Chi2 test has been performed. In the cases where the conditions of the Chi2 test have not been fulfilled, Fisher’s exact test has been used instead. In order to compare two mean values, a 2 sample t test has been conducted. For ordinally scaled variables, Wilcoxon 2 sample test has been performed.

In order to analyse the impact of several factors on a binary outcome (i.e. caesarean section), logistic regression analysis have been performed and odds ratios have been assessed. Furthermore, a multiple logistic regression analysis has been performed in order to analyse several factors simultaneously. For the multivariable models, AUC has been calculated as a measure of the goodness of the model.

In general, the result of a statistical test has been considered as statistically significant for p less than 0.05.


Results

During the study period, a total of 20531 deliveries were recorded. Following application of inclusion and exclusion criteria, 1010 women undergoing induction of labour (IOL) were included in the analysis. Of these, 616 underwent IOL in an inpatient setting, while 394 received cervical ripening in an outpatient setting.

In the demographic analysis ([Table 2]), maternal age and body mass index (BMI) were comparable between the outpatient and inpatient groups, with no statistically significant differences. Gestational age at delivery was higher in the outpatient cohort (287.8 ± 5.0 days vs. 284.1 ± 8.0 days, p < 0.0001), and mean birth weight was slightly higher (3719.7 ± 402.4 g vs. 3645.6 ± 437.8 g). Parity was higher in the inpatient group (0.7 vs. 0.5, p = 0.0115). Nulliparous women were more frequently represented in the outpatient group (61.9% vs. 55.4%, p = 0.0391). Among these parameters, the proportion of nulliparous women appears to be the most clinically relevant difference.

Table 2 Demographic parameters.

Outpatient setting (n = 394)

Inpatient setting (n = 616)

P value

Maternal age (years)

31.4 ± 4.6

31.1 ± 5.4

0.3538

Body mass index

29.2 ± 6.4
(n = 370)

29.3 ± 5.9
(n = 578)

0.8585

Parity (mean, median, range)

0.5
0 (0–4)

0.7
0 (0–6)

0.0115

Nulliparous (n, %)

244 (61.9%)

341 (55.4%)

0.0391

Gestational age (days)

287.8 ± 5.0

284.1 ± 8.0

< 0.0001

Birth weight (grams)

3719.7 ± 402.4
(n = 393)

3645.6 ± 437.8

0.0070

Bishop score (median, range)

1.4
1 (0–5)
(n = 364)

1.5
1 (0–5)
(n = 563)

0.2879

Large for gestational age (n, %)

20 (5.1%)

49 (8.0%)

0.0770

Indications for IOL differed notably between the two settings ([Table 3]). Late-term pregnancy was the most frequent indication in the outpatient group (76.4% vs. 51.3%), whereas induction on maternal request (i.e., without medical indication) was more common in the inpatient group (34.7% vs. 19.5%). IOL due to suspected fetal macrosomia was rare in the outpatient group (0.5%) compared to the inpatient cohort (4.2%).

Table 3 Indication for induction of labour.

Outpatient setting (n = 394)

Inpatient setting (n = 616)

P value

On request (n, %)

77 (19.5%)

214 (34.7%)

< 0.0001

Suspected fetal macrosomia (n, %)

2 (0.5%)

26 (4.2%)

Late term (n, %)

301 (76.4%)

316 (51.3%)

Others (n, %)

1 (0.3%)

26 (4.2%)

Large for gestational age (n, %)

13 (3.3%)

34 (5.5%)

Outcome parameters are summarized in [Table 4]. The overall mode of delivery (vaginal, operative vaginal, and caesarean section) did not differ significantly between the groups (p = 0.5524). However, the rate of vaginal deliveries within 24 hours was lower in the outpatient group (25.2% vs. 32.9%, p = 0.0201). No significant differences were observed for vaginal deliveries within 48 hours (p = 0.2819) or 72 hours (p = 0.7869), nor for the induction-to-delivery interval (p = 0.2277). Adverse neonatal outcomes, including arterial umbilical cord pH < 7.10 (2.0% vs. 5.5%, p = 0.0067) and transfer to a neonatal intensive care unit (6.9% vs. 15.1%, p < 0.0001), were more frequent in the inpatient group. Fetal blood sampling was more commonly performed in the outpatient setting (6.9% vs. 3.7%, p = 0.0258).

Table 4 Outcome parameter.

Outcome parameter

Outpatient setting (n = 394)

Inpatient setting (n = 616)

P values

* Caesarean sections were excluded, n. c. = not calculable

Normal delivery

307 (77.9%)

463 (75.2%)

0.5524

Vaginal-operative delivery

57 (14.5%)

96 (15.6%)

Caesarean section

30 (7.6%)

57 (9.3%)

Induction to delivery
interval (min)*
median (range)

1788
(369–7685)
(n = 314)

1707
(294–9008)
(n = 487)

0.2277

Vaginal delivery within 24 hours (n, %)*

79 (25.2%)
(n = 314)

160 (32.9%)
(n = 487)

0.0201

Vaginal delivery within 48 hours (n, %)*

171 (54.5%)
(n = 314)

284 (58.3%)
(n = 487)

0.2819

Vaginal delivery after 72 hours (n, %)*

21 (6.7%)
(n = 314)

35 (7.2%)
(n = 487)

0.7869

Onset of labour only by double-balloon catheter (n, %)

54 (13.7%)

94 (15.3%)

0.4957

Arterial umbilical cord pH < 7.10 (n, %)

8 (2.0%)

34 (5.5%)

0.0067

Arterial umbilical cord base excess < −12 (n, %)

8 (2.0%)

29 (4.7%)

0.0272

Apgar after 5 minutes < 7 (n, %)

4 (1.0%)

14 (2.3%)

0.1406

pH < 7.00 and base excess < 12 and Apgar after 5 minutes < 7 (n, %)

1 (0.3%)

1 (0.2%)

1.0000

Abnormal CTG (n, %)

53 (13.5%)

86 (14.0%)

0.8188

Fetal blood analysis (n, %)

27 (6.9%)

23 (3.7%)

0.0258

Abnormal fetal blood analysis (n, %)

4 (1.0%)

4 (0.6%)

0.7184

Meconium stained liquor (n, %)

72 (18.3%)

112 (18.2%)

0.9704

Suspected Triple I (n, %)

3 (0.8%)

5 (0.8%)

1.0000

Uterine rupture (n, %)

0

0

n. c.

Shoulder dystocia (n, %)

2 (0.5%)

7 (1.1%)

0.4951

Transfer to neonatal (intensive) care unit (n, %)

27 (6.9%)

93 (15.1%)

< 0.0001

Neonatal death (n, %)

0

1 (0.2%)

1.0000

Factors associated with caesarean section were explored using regression analysis ([Table 5]). Nulliparity was the strongest predictor of caesarean delivery (odds ratio [OR] 10.545, p < 0.0001). Maternal age (OR 1.076, p = 0.0001) and birth weight (OR 1.733, p = 0.0163) were also identified as significant predictors.

Table 5 Univariable and multiple regression analysis of the primary outcome parameter “caesarean section” (yes [n = 153] versus no [n = 857]).

Univariable analysis, odds ratio

Univariable p values

Multiple analysis, odds ratio

Multiple analysis, p values (significant factors)

AUC = 0.744 (multiple model)

Setting (outpatient versus inpatient)

0.916

0.6920

Onset of labour only by double-balloon catheter (yes versus no)

0.642

0.1111

Body mass index (kg/m2)

1.011

0.4488

Maternal age (years)

1.027

0.1280

1.076

0.0001

Nulliparous (yes versus no)

7.817

< 0.0001

10.545

< 0.0001

Bishop score (grade)

0.816

0.0027

Gestational age (weeks)

1.316

0.0066

Birth weight (kg)

1.241

0.2950

1.733

0.0163

Large for gestational age (yes versus no)

1.469

0.2172

Multivariable regression analysis of the primary outcome parameter “no vaginal delivery within 24 hours” revealed several significant predictors ([Table 6]). The likelihood of not achieving vaginal delivery within 24 hours was significantly lower when labour was induced using a double-balloon catheter alone (OR 0.149, p < 0.0001). Conversely, higher BMI (OR 1.080, p < 0.0001), nulliparity (OR 4.191, p < 0.0001), and presence of a large-for-gestational-age fetus (OR 3.675, p = 0.0086) were associated with increased odds of prolonged induction.

Table 6 Univariable and multiple regression analysis of the primary outcome parameter “Vaginal delivery within 24 hours” (no [n = 562] versus yes [n = 239]).

Univariable analysis, odds ratio

Univariable p values

Multiple analysis, odds ratio

Multiple analysis, p values (significant factors)

AUC = 0.757 (multiple model)

Setting (outpatient versus inpatient)

1.456

0.0201

Onset of labour only by double-balloon catheter (yes versus no)

0.188

< 0.0001

0.149

< 0.0001

Body mass index (kg/m2)

1.043

0.0034

1.080

< 0.0001

Maternal age (years)

0.970

0.0460

Nulliparous (yes versus no)

3.384

< 0.0001

4.191

< 0.0001

Bishop score (grade)

0.849

0.0018

Gestational age (weeks)

0.953

0.5250

Birth weight (kg)

1.561

0.0158

Large for gestational age (yes versus no)

4.074

0.0015

3.675

0.0086

For the outcome “arterial umbilical cord pH < 7.10”, outpatient setting was the only significant impact factor ([Table 7]). Neither BMI nor nulliparity had a statistically significant effect. Developing a multiple model was only possible when assuming a significance level of 0.10. Doing so, outpatient setting and maternal age revealed to be significant. Outpatient IOL was associated with a reduced risk for an arterial pH < 7.10 (OR 0.352, p = 0.0088).

Table 7 Univariable and multiple regression analysis of the primary outcome parameter “Arterial umbilical cord pH < 7.10” (yes [n = 42] versus no [n = 968]).

Univariable analysis, odds ratio

Univariable p values

Multiple analysis, odds ratio

Multiple analysis, p values (significant factors)

AUC = 0.651 (multiple model)

Setting (outpatient versus inpatient)

0.355

0.0067

0.352

0.0088

Onset of labour only by double-balloon catheter (yes versus no)

0.437

0.1597

Body mass index (kg/m2)

1.011

0.6743

Maternal age (years)

1.059

0.0712

1.059

0.0661

Nulliparous (yes versus no)

1.189

0.5932

Bishop score (grade)

0.936

0.5500

Gestational age (weeks)

1.118

0.5019

Birth weight (kg)

0.876

0.7202

Large for gestational age (yes versus no)

0.672

1.0000

Discussion

The findings of this study provide valuable insights into the effectiveness and safety of outpatient IOL using the double-balloon catheter in low-risk pregnancies. By comparing outpatient and inpatient settings, the results expand upon the existing body of literature.

In the present study, no significant difference was observed in the caesarean section rate between outpatient and inpatient induction of labour using a double-balloon catheter (7.6% vs. 9.3%, p = 0.5524). Regression analysis also confirmed that the setting (outpatient vs. inpatient) was not a significant predictor for caesarean delivery. Instead, key influencing factors included nulliparity (OR: 10.545; p < 0.0001), maternal age (OR: 1.076; p = 0.0001), and birth weight (OR: 1.733; p = 0.0163). A higher Bishop score was associated with a reduced risk of caesarean section, underscoring the clinical relevance of cervical favourability at the time of induction.

These findings are consistent with the current evidence base. Randomized controlled trials and meta-analyses have consistently shown that caesarean section rates are comparable between outpatient and inpatient cervical ripening, particularly in nulliparous women or when cervical ripening is required. A meta-analysis by Dong et al., including 2615 women, found no significant difference in caesarean rates between outpatient and inpatient protocols (RR 0.89; 95% CI 0.74–1.07). However, a subgroup analysis demonstrated that mechanical cervical ripening using a balloon catheter in the outpatient setting was associated with a significantly lower caesarean section rate (RR 0.52; 95% CI 0.30–0.90) [13].

The Cochrane review by Alfirevic et al. supports this conclusion. When using prostaglandin E2 or a balloon catheter, no significant differences in caesarean section rates were found between outpatient and inpatient settings (e.g., balloon/Foley catheter: RR 0.64; 95% CI 0.41–1.01), although the quality of evidence was rated as very low [16].

Similarly, a systematic review by Sangskär et al. focusing on outpatient inductions in low-risk pregnancies concluded that caesarean section rates were comparable between settings. The authors emphasized, however, that the overall quality of evidence was very low and affected by methodological limitations, including heterogeneous definitions of outpatient protocols and early randomization [19].

Major limitations of the available evidence include small sample sizes, heterogeneity of induction methods, inconsistent definitions of outpatient care, early timing of randomization, and lack of blinding. Most trials were also underpowered to detect rare but clinically relevant safety outcomes, and generalizability to high-risk populations is limited [13] [16] [19].

In the present study, vaginal delivery within 24 hours after induction was less frequently achieved in the outpatient setting compared to the inpatient setting (25.2% vs. 32.9%, p = 0.0201). Univariable analysis showed that the outpatient setting was significantly associated with a higher likelihood of delayed vaginal delivery beyond 24 hours (OR: 1.456; p = 0.0201). This finding can be attributed in part to the higher proportion of nulliparous women in the outpatient group (61.9% vs. 55.4%; p = 0.0391). Nulliparity emerged as the strongest predictor for not achieving a vaginal delivery within 24 hours (OR: 3.384; p < 0.0001). A higher body mass index (BMI) was also significantly associated with delayed vaginal delivery (OR: 1.043; p = 0.0034), while a higher Bishop score significantly reduced this risk (OR: 0.849; p = 0.0018). In multivariable regression analysis, use of the double-balloon catheter as the sole induction method markedly decreased the odds of no vaginal delivery within 24 hours (OR: 0.149; p < 0.0001). Nulliparity (adjusted OR: 4.191; p < 0.0001) and elevated BMI (adjusted OR: 1.080; p < 0.0001) remained significant predictors after adjustment for confounders.

These findings underscore the clinical relevance of individual patient characteristics – particularly parity, BMI, and cervical status – for the timing of vaginal delivery following mechanical cervical ripening with a double-balloon catheter.

Current evidence from randomized controlled trials and meta-analyses supports the notion that the rate of vaginal delivery within 24 hours is generally comparable between outpatient and inpatient induction settings, especially among nulliparous women or those with an unfavourable cervix. For example, a systematic review and meta-analysis by Dong et al. found no significant difference in the rate of vaginal birth within 24 hours between outpatient and inpatient induction approaches across various methods [13]. Similarly, Sangskär et al. reported comparable outcomes regarding the 24-hour delivery rate, while highlighting shorter hospital stays and similar safety profiles in the outpatient setting [19]. A retrospective study by Hallén et al. also found no significant difference in the timing of vaginal delivery when oral misoprostol was used in outpatient versus inpatient settings [20].

A randomized trial using a synthetic osmotic dilator found a non-significant trend toward more frequent vaginal deliveries within 24 hours in the outpatient group, but overall delivery rates remained similar [21]. Mazzoli and O’Malley, in a systematic review of slow-release dinoprostone inserts, also observed no significant differences in the timing of vaginal delivery between outpatient and inpatient administration [22]. Finally, a meta-analysis by Pierce-Williams et al. focused specifically on balloon catheter use and found no significant difference in 24-hour vaginal delivery rates, while confirming reduced hospital stays and a lower caesarean section rate in the outpatient group [23].

Neonatal outcome parameters in our study revealed several significant differences in favour of the outpatient setting. The incidence of arterial umbilical cord pH < 7.10 was significantly lower in the outpatient group (2.0%) compared to the inpatient group (5.5%; p = 0.0067). This protective effect persisted in the multivariable analysis (OR 0.352; p = 0.0088). Likewise, arterial umbilical cord base excess < −12 occurred less frequently in the outpatient group (2.0% vs. 4.7%; p = 0.0272). Additionally, fewer neonates required transfer to a neonatal intensive care unit (NICU) in the outpatient setting (6.9% vs. 15.1%; p < 0.0001). This may reflect reduced procedural intervention or improved maternal well-being in the outpatient environment.

These findings align with the meta-analysis by Dong et al., which discussed the potential benefits of outpatient labour induction in reducing stress-related complications for both mother and fetus [13]. Sutton et al. also reported psychological benefits from outpatient cervical ripening, with most women expressing a preference for undergoing this procedure again in the future [24].

A recent meta-analysis by Pierce-Williams et al. demonstrated that the use of outpatient balloon catheters does not increase the incidence of adverse maternal or neonatal outcomes, including arterial umbilical cord pH < 7.10 or NICU admissions [23]. However, this meta-analysis included trials comparing balloon catheters in outpatient settings with various inpatient methods, including prostaglandin E2 (PGE2), which limits direct comparability with our results.

Other systematic reviews and randomized trials confirm that NICU admission rates and other neonatal outcomes are comparable between outpatient and inpatient labour induction in well-selected low-risk women [16] [19] [25]. Nevertheless, the overall quality of evidence is considered low, as most studies are underpowered to detect rare neonatal complications and event rates remain generally low.

This study provides robust real-world data supporting the safety and effectiveness of outpatient induction of labour using a double-balloon catheter in a well-defined low-risk population. The lack of difference in caesarean section rates between outpatient and inpatient settings aligns with previous trials and meta-analyses, reinforcing the role of patient characteristics – especially nulliparity, BMI, maternal age and birth weight – as key predictors of delivery outcomes, regardless of the care setting. The outpatient cohort demonstrated a lower incidence of adverse neonatal parameters such as umbilical cord pH < 7.10 and NICU admissions, suggesting potential benefits in neonatal wellbeing. The observed delay in vaginal delivery within 24 hours in the outpatient group can be largely explained by a higher proportion of nulliparous women, as confirmed in regression analyses. Although the study is limited by its retrospective design and single-centre setting, the large sample size, strict inclusion criteria, and uniform adherence to standardized protocols enhance the validity and clinical relevance of the findings.


Conclusions

Outpatient cervical ripening with a double-balloon catheter represents a safe and effective alternative to inpatient induction for carefully selected low-risk pregnancies. Caesarean section rates and overall delivery outcomes were comparable between groups. Importantly, the outpatient setting was associated with improved neonatal outcomes, including a lower incidence of arterial cord pH < 7.10 and fewer NICU admissions. These findings underscore the importance of individualized patient selection, particularly with respect to parity and BMI, when planning labour induction. Outpatient IOL may contribute to maternal satisfaction and healthcare resource optimization without compromising safety. Future prospective, ideally randomized, trials are warranted to confirm these findings and to further refine outpatient induction protocols.



Conflict of Interest

The authors declare that they have no conflict of interest.


Correspondence

Prof. Sven Kehl
Department of Obstetrics and Gynecology, LMU University Hospital, LMU Munich
Marchioninistraße 15
81377 Munich
Germany   

Publication History

Received: 25 October 2025

Accepted after revision: 25 December 2025

Article published online:
29 January 2026

© 2026. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial-License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://creativecommons.org/licenses/by-nc-nd/4.0/).

Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany


  Permissions and Reprints