Open Access
CC BY-NC-ND 4.0 · Geburtshilfe Frauenheilkd 2025; 85(12): 1326-1333
DOI: 10.1055/a-2713-8946
GebFra Science
Original Article

Induction of Labor with Oral Misoprostol – Effectiveness and Safety of Low-dose vs. High-dose Regimens

Article in several languages: English | deutsch

Authors

  • Asal Fathi Roodsari

    1   Klinik für Frauenheilkunde und Geburtshilfe, Medizinische Hochschule Hannover, Hannover, Germany (Ringgold ID: RIN9177)
    2   Universitätsklinik für Kinder- und Jugendmedizin, Johannes Wesling Klinikum Minden, Universitätsklinikum der Ruhr-Universität Bochum, Minden, Germany (Ringgold ID: RIN39631)
  • Friederike Gebauer

    1   Klinik für Frauenheilkunde und Geburtshilfe, Medizinische Hochschule Hannover, Hannover, Germany (Ringgold ID: RIN9177)
  • Lars Brodowski

    1   Klinik für Frauenheilkunde und Geburtshilfe, Medizinische Hochschule Hannover, Hannover, Germany (Ringgold ID: RIN9177)
  • Vivien Less

    3   Institut für Statistik, Wirtschaftswissenschaftliche Fakultät, Leibniz Universität Hannover, Hannover, Germany (Ringgold ID: RIN232606)
  • Philipp Sibbertsen

    3   Institut für Statistik, Wirtschaftswissenschaftliche Fakultät, Leibniz Universität Hannover, Hannover, Germany (Ringgold ID: RIN232606)
  • Peter Hillemanns

    1   Klinik für Frauenheilkunde und Geburtshilfe, Medizinische Hochschule Hannover, Hannover, Germany (Ringgold ID: RIN9177)
  • Constantin Sylvius von Kaisenberg

    1   Klinik für Frauenheilkunde und Geburtshilfe, Medizinische Hochschule Hannover, Hannover, Germany (Ringgold ID: RIN9177)
  • Rüdiger Klapdor

    4   Gynäkologie und Geburtshilfe, Albertinen Krankenhaus Hamburg, Hamburg, Germany (Ringgold ID: RIN39611)
  • Vivien Dütemeyer

    1   Klinik für Frauenheilkunde und Geburtshilfe, Medizinische Hochschule Hannover, Hannover, Germany (Ringgold ID: RIN9177)
 

Abstract

Background

Before the market launch of low-dose misoprostol (LDM) in the form of 25 µg tablets, high-dose misoprostol (HDM) was used off-label for labor induction in Germany. The differences between the effectiveness and safety of both dosage forms were investigated.

Method

This retrospective cohort study included all pregnant women who were induced with misoprostol between 2019 and 2023. Until September 2021, misoprostol was available in the form of a 200 µg tablet which was divided manually and administered orally at an initial dose of 50 µg, followed by 100 µg every four hours. This was compared with the current administration regimen of an initial dose of 25 µg followed by 50 µg in terms of vaginal birth rate, duration of induction, and maternal and neonatal outcomes.

Results

A total of 1941 patients were analyzed; 1397 received HDM and 544 received LDM. Vaginal delivery was achieved with equal frequency (87% HDM vs. 88% LDM, p = 0.95). The duration of induction was shorter in the HDM group (20.7 hours vs. LDM 22.7 hours, p = 0.028), and opioid requirements were higher (HDM 74% vs. LDM 69%, p = 0.036). Maternal complications and neonatal outcomes did not differ. Regression analysis found that high BMI was an independent factor for a failed first induction attempt (B 0.002, SEM 0.0004, p < 0.001).

Conclusion

Both dosing regimens achieved high vaginal birth rates with comparable safety. Inductions with HDM were two hours shorter but were associated with higher opioid use.


Introduction

More than 20% of all births in Germany are induced [1]. According to the German guideline “Induction of Labor” published in December 2020 and updated in March 2021, active induction of labor is performed to achieve better maternal and perinatal outcomes than can be achieved with expectant management [2].

When choosing the induction method, the Bishop Score is used for orientation as it takes cervical dilation, the length, consistency, and position of the cervix as well as the level of the leading edge of the presenting part of the infant into account [3]. Assessment of cervical maturity is important when choosing the induction method because cervical maturity significantly affects the success of vaginal delivery [4]. Cervical ripening is decisive if the cervix is still immature [5].

Misoprostol has proven to be a very good active agent for cervical ripening, especially when compared to other methods of induction, and the World Health Organization has included it in its list of “Essential Drugs” [5] [6] [7]. It is a prostaglandin E1 analog which was originally developed for the prevention and treatment of gastroduodenal ulcers, but it is also used worldwide for the induction of labor due to its contractile uterine effect [6] [8]. In Germany, the use of misoprostol for induction of labor was off-label until September 2021, and every pregnant woman for whom misoprostol was considered had to be explicitly informed about its off-label use. Moreover, administration of misoprostol was problematic as the tablets were only available at dose of 200 µg whereas the WHO recommends lower doses for the induction of labor [9]. This resulted in the tablets being divided manually in many hospitals in Germany, which was associated with a risk of overdosing or underdosing. At the same time, there was no established dosing regimen with a defined maximum daily dose, which led to reports of undesirable side effects such as uterine overstimulation or even uterine rupture [10]. Although the majority of maternity hospitals in Germany use misoprostol, the Federal Institute for Drugs and Medical Devices issued a Red Hand Letter and warned against the off-label administration of misoprostol [11] [12].

With the recent market launch of low-dose misoprostol (LDM) in the form of 25 µg tablets, there is now an approved drug for the oral induction of labor which should ensure that it can be safely administered [13]. Hydrogenated castor oil, which was present in the previously used higher-dose form of misoprostol and which also induces labor in its pure form, is not part of the low-dose misoprostol tablets [8] [14]. The company distributing misoprostol has also issued clear recommendations regarding the induction regimen [15].

A question has arisen whether there are differences between LDM and high-dose misoprostol (HDM) in the context of labor induction, and if so, what they are. A comparison of the two dosing regimens with regards to their effectiveness, vaginal birth rate, duration of labor, and the maternal, fetal, and obstetric side effects was done to resolve the issue.


Method

Patient cohort

This is a retrospective cohort study which included all pregnant women between July 2019 and February 2023 treated with oral misoprostol for the induction of labor at Hanover Medical School. Exclusion criteria were patients with intrauterine fetal death, incomplete data, or patients with an unsuccessful initial attempt at induction who were treated with a different method after a break of at least 24 hours in a second induction cycle.


Induction method

During the first part of the study, misoprostol was only available in the form of 200 µg tablets. The tablets were halved or cut into quarters and patients initially received an oral dose of 50 µg followed, after an interval of four hours, by 100 µg. Since September 2021 misoprostol has been available at a dose of a 25 µg. In our department women initially receive a test dose of 25 µg, the recommended starting dose, to test their response and prevent uterine tachysystole [16], followed by oral administration of 50 µg at intervals of four hours in accordance with the administration guidelines for the preparation. The maximum daily dose for HDM was 300 µg [17], whereas it was 200 µg with the lower dose regimen [15]. CTG for at least 30 minutes was carried out prior to every administration of medication to monitor for infant distress and to exclude regular contractions. Monitoring was done on an inpatient basis and a repeat CTG was carried out in patients with painful contractions, rupture of membranes, or vaginal bleeding. If patients have a Bishop Score ≤ 6, our department initially places a double balloon catheter unless the patient explicitly rejects this approach or placement is unsuccessful. The double balloon catheter is removed after 6–12 hours after which induction is continued with medication. If there is no onset of labor after 2–3 days of one or the other of the misoprostol regimens, induction is continued with vaginal prostaglandin gel for a maximum of 2 more days. If this attempt to induce labor is also unsuccessful, the options of an amniotomy or – if the department permits – a 24 hour pause before starting a renewed cycle with misoprostol are discussed with the patient. If the patient explicitly refuses misoprostol, a second induction cycle is carried out with prostaglandin gel. If needed, an oxytocin infusion is administered during labor to support contractions.


Statistical analysis

Primary endpoint was the vaginal birth rate, with spontaneous delivery and instrumental vaginal birth grouped together. The secondary endpoint to test for effectiveness was the interval between the first administration of misoprostol and delivery as well as the additional use of oxytocin to stimulate contractions. The safety assessment was done based on the following factors: tocolytic therapy with intravenous fenoterol in patients with uterine tachysystole or permanent contractions related to the administration of misoprostol, meconium-stained amniotic fluid, opioid requirements (tramadol, meptazinol, nalbuphine) and administration of an epidural anesthesia (EDA) as well as maternal morbidity (perineal tear, postpartum hemorrhage [PPH] defined as a blood loss ≥ 500 ml after vaginal delivery or ≥ 1000 ml after caesarean section [18], chorioamnionitis [Triple I]) and neonatal outcome (APGAR score, cord blood pH, respiratory support, reanimation, transfer to pediatric department).

Possible influencing variables such as maternal body mass index (BMI), parity, gestational age delivery in weeks of gestation (GW), twin pregnancy (the delivery and parameters of the first delivered neonate were used), breech presentation, birth weight, cumulative dose of misoprostol, previous balloon catheter placement, or other forms of induction using prostaglandin gels were recorded. If induction was unsuccessful, a second induction cycle was carried out in some patients after a pause of at least 24 hours. In these cases, the second induction cycle was evaluated.

The study was approved by the local ethics committee (No. 10860_BO_K_2023) and was carried out in accordance with the Helsinki Declaration [19].

Statistical analysis of the data was done with R version 4.4.3. Metric variables were presented as means and standard deviation (SD) and categorical variables as frequency rates. Shapiro-Wilk test was carried out to test for normal distribution. Mann-Whitney U-test and chi-square test were used to compare two factors. Multivariate binary logistic regression analysis or simple linear regression tested the factors for independence in terms of vaginal birth or duration of induction and number of induction cycles. The level of significance was set at p ≤ 0.05.



Results

Patient cohort

A total of 2827 patients had an induction during the study period. After taking the inclusion criteria into account, a total of 1941 cases remained for evaluation. Of these, 1397 were induced with HDM and, later in the study, 544 were induced with LDM ([Fig. 1]).

Zoom
Fig. 1 Flow chart of the study population (Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 2009; 6: e1000097. DOI: 10.1371/journal.pmed1000097).

The mean gestational age was significantly higher in the HDM group (39.8 vs. 39.6 GW, p = 0.012) as was the birth weight (3450 g vs. 3380 g, p = 0.01) and fewer pregnant women were nulliparous (70% vs. 81%, p < 0.001) or had hypertension (4% vs. 8%, p = 0.003) compared to the LDM group. Fewer pregnant women induced with HDM underwent prior balloon catheter placement (45% vs. 53%, p = 0.002), had an amniotomy during induction (10% vs. 14%, p = 0.019) or required prostaglandin gel for a further attempt at induction of labor (6% vs. 14%, p < 0.001). There were significantly more cases with breech presentation in the group induced with LDM (5%) compared to the group treated with the HDM regimen (1%, p < 0.001). The cumulative dose was significantly higher for patients treated with HDM (231 µg) compared to 165 µg for the group induced with LDM (p < 0.001). The duration of induction was also significantly shorter (HDM 20.7 h vs. LDM 22.7 h, p = 0.028) and opioid requirements were higher (HDM 74% vs. LDM 69%, p = 0.036). The remaining factors including maternal characteristics, obstetric complications or neonatal outcomes did not differ between groups ([Table 1]).

Table 1 Characteristics of the study population.

HDM (n = 1397)

LDM (n = 544)

p value

Data are presented as rates for categorical variables (absolute number = n) or means (standard deviation) for metric variables.

Abbreviations: BMI = body mass index; EDA = epidural anesthesia; GDM = gestational diabetes; h = hours; HDM = high-dose misoprostol; LDM = low-dose misoprostol

BMI, kg/m2

26.41 (0.48)

26.68 (0.49)

0.452

Maternal age, years

31.8 (0.44)

31.9 (0.46)

0.717

Nicotine abuse

2% (23)

2% (12)

0.521

Diabetes mellitus/GDM

17% (235)

15% (82)

0.386

Hypertension

4% (57)

8% (41)

0.003

Nulliparity

70% (969)

81% (440)

< 0.001

Gestational age at delivery, weeks

39.83 (0.24)

39.64 (0.35)

0.012

Twin pregnancy

1.03 (0.65)

1.03 (0.58)

0.951

Breech presentation

1% (20)

5% (25)

< 0.001

Balloon catheter

45% (625)

53% (286)

0.002

Oxytocin

36% (502)

39% (213)

0.205

Amniotomy

10% (137)

14% (74)

0.019

Prostaglandin gel

6% (88)

14% (76)

< 0.001

Tocolytic therapy

11% (152)

13% (71)

0.208

Meconium-stained amniotic fluid

7% (91)

8% (41)

0.482

Chorioamnionitis

1% (14)

2% (10)

0.205

Opioids

74% (1031)

69% (375)

0.036

EDA

16% (208)

17% (89)

0.863

Cumulative dose, micrograms

231 (206)

165 (118)

< 0.001

Duration of induction, hours

20.7 (0.0)

22.7 (0.0)

0.028

Vaginal delivery

87% (1219)

88% (476)

0.946

Vacuum extraction

8% (98)

8% (41)

0.909

Birth weight, grams

3449.5 (0.37)

3379.5 (0.37)

0.01

5-minute APGAR score

9.8 (0.29)

9.83 (0.3)

0.322

Arterial cord blood pH

7.25 (0.3)

7.24 (0.34)

0.056

Perineal tear °III

2% (28)

2% (12)

0.918

Postpartum hemorrhage

9% (106)

10% (42)

0.64

Neonatal reanimation

0%

0%

1.0

Mask ventilation

3% (47)

2% (11)

0.161

Transfer to pediatric department

5% (69)

5% (27)

1.0


Primary endpoint

Vaginal delivery after induction with HDM was achieved in 87% of cases, which is comparable with the rate of 88% for the cases in the LDM group (p = 0.95).

The goal of a vaginal birth was achieved less often if tocolytic therapy (OR 0.8, 95% CI: 0.7–0.8, p < 0.001), oxytocin (OR 0.92, 95% CI: 0.9–1.0, p = 0.05) or prostaglandin gel (OR 0.89, 95% CI: 0.8–0.9, p < 0.001) was used or the patient had an EDA (OR 0.94, 95% CI: 0.9–1.0, p = 0.01). Opioid administration during labor (OR 1.04, 95% CI: 1.0–1.1, p = 0.05) resulted more commonly in a vaginal birth ([Table 2]).

Table 2 Linear regression for vaginal delivery.

OR (95% CI)

p value

Tocolytic therapy

0.8 (0.7–0.8)

< 0.001

Oxytocin

0.92 (0.9–1.0)

0.05

Prostaglandin gel

0.89 (0.8–0.9)

< 0.001

Epidural anesthesia

0.94 (0.9–1.0)

0.01

Opioids

1.04 (1.0–1.1)

0.05


Secondary endpoint

Regression analysis showed that a higher cumulative dose (B 0.095, SEM 0.002, p = 0.001), nulliparity (B −0.523, SEM 0.234, p = 0.05), use of opioids (B 1.661, SEM 0.58, p = 0.01), oxytocin (B 5.651, SEM 0.587, p = 0.001) or EDA (B 1.735, SEM 0.745, p = 0.05) and chronic hypertension (B 2.607, SEM 1.132, p = 0.05) were associated with a longer induction duration. The most important factor for a longer duration was the use of LDM, with a regression coefficient of 6.27 (SEM 0.569, p < 0.001), and the subsequent use of prostaglandin gel (B 31.25, SEM 1.017, p < 0.001) ([Table 3]).

Table 3 Logistic regression analysis of duration of induction.

B (SEM)

p value

Low-dose misoprostol

6.27 (0.569)

< 0.001

Prostaglandin gel

31.25 (1.017)

< 0.001

Cumulative dose

0.095 (0.002)

0.001

Nulliparity

−0.523 (0.234)

0.05

Opioids

1.661 (0.58)

0.01

Oxytocin

5.651 (0.587)

0.001

Epidural anesthesia

1.735 (0.745)

0.05

Chronic hypertension

2.607 (1.132)

0.05

The only independent factor associated with the need for several induction cycles was a high BMI (B 0.002, SEM 0.0004, p < 0.001).



Discussion

Low-dose and high-dose misoprostol were equally successful in inducing labor and had similar side-effects profiles. The duration of induction was around two hours shorter with HDM but the opioid requirements were higher.

The older gestational age in the group induced with HDM was undoubtedly related to the higher birth weight in this cohort. It could be a coincidence but could possibly also be due to the fact that in recent years, more patients with pathologies are being induced with LDM who do not go on to give birth via primary caesarean section. This includes, for example, pregnant women with arterial hypertension who may be offered the option of giving birth from week 37 of gestation in accordance with guideline recommendations [20]. The increasing numbers of early inductions recorded in recent years explains the higher rate of balloon catheter placements in the LDM group as an early week of gestation is more likely to be associated with an immature cervix. What is certain is that the number of breech presentation pregnancies in our department has increased and the rate of planned caesarean sections for breech presentation has decreased [21].

Some factors are associated with a longer duration of induction or a longer interval between induction and delivery but should not be considered as causative. For example, prostaglandin gel is only applied when the administration of misoprostol did not induce labor. Oxytocin does not result in longer labor times but is administered when the progress of labor is protracted due to weak contractions. Pregnant women often request an EDA if the first stage of labor is protracted and it is sometimes also recommended in cases where the fetal position is unfavorable and could delay the progression of labor; however, an EDA may also reduce the intensity and frequency of contractions and thereby prolong the time between induction and delivery.

Previous studies have already shown that obesity increases the risk of longer induction times and may even result in induction being unsuccessful [22] [23]. Our study also found that pregnant women whose first induction cycle was unsuccessful were more likely to have a higher BMI. A randomized study showed no benefit from higher doses of misoprostol administered vaginally [24] and it therefore appears that the first stage of labor may generally be longer in obese patients, even in women with spontaneous onset of labor, and that obese women are more likely to have a caesarean section in this phase compared to women of normal weight [25] [26].

When choosing the induction method, it is essential that it is not just considered to be effective but has also been assessed as safe. Initial placement of a balloon catheter when the cervix is still immature, especially in primigravidae, followed by oral misoprostol appears to be more beneficial compared to direct induction using misoprostol alone, and this approach is standard in our department [27]. It has also been found that oral intake of misoprostol should be preferred to vaginal administration [28].

The provider of the 25 µg misoprostol tablets has proposed two different dosing regimens [15]. The proscribing information cites studies which looked at higher doses of misoprostol or misoprostol in soluble form [29] [30] [31]. The drug was tested in 29000 pregnant women in Denmark, Norway, and Finland in the context of a compassionate use program and was found to be low risk [15]. A regimen of 25 µg misoprostol every two hours is also recommended by the WHO [32]. The other induction regimen, which is used in our department, consists of the administration of 50 µg every four hours. The advantage of this approach is that it is more convenient, reduces the workload of the midwives and also offers pregnant women more rest times. Whether a more frequent administration of just 25 µg is more likely to result in a vaginal birth is still controversial but it does appear to be associated with a lower risk of transfer of the neonate to a pediatric department [30] [31] [32] [33] [34] [35].

Another study from Denmark reported that induction performed on an outpatient basis is possible with a low-dose regimen. The pregnant woman only receives the first dose in hospital where she is monitored, and she then subsequently takes the tablets herself at home every two hours [36]. Kandahari et al. reported that in 2012 they already started to induce patients with misoprostol on an outpatient basis in California under study conditions. They found that the time between admission of the pregnant woman and delivery of the neonate was more than six hours shorter compared to pregnant women who were induced in hospital, while neonatal and maternal outcomes and the rates of caesarean sections were comparable [37]. Future prospective randomized studies are needed to determine to what extent such a form of induction would also be possible in Germany and/or whether it could be implemented with a dosing regimen of 50 µg and whether it would be safe. In view of the costs involved, this could be a useful measure.

Compared to other studies, the secondary caesarean section rate in our study population was low (13% vs. 12% in the HDM vs. LDM cohort). In a meta-analysis which included a total of 21030 women around the time of delivery, the secondary caesarean section rate was recorded as 18.6% in the group of women with spontaneous onset of labor and 16.7% for the group of women who were induced. The authors concluded that induction of labor in a low-risk cohort at the end of a pregnancy could potentially be associated with a lower caesarean section rate (OR 0.90, 95% CI: 0.85–0.95) [38]. Fonseca et al. compared women over the age of 35 years who were induced with women who had a spontaneous onset of labor and came to similar conclusions as they found no differences in secondary caesarean section rates or rates of postpartum bleeding [39].

Our evaluation showed that tocolytic therapy for uterine tachysystole or abnormal CTG was administered in 11% and 13% of cases (HDM and LDM, respectively). A recent retrospective study from Cologne of 3005 women who were past their due date comparing the induction of labor with natural onset of labor found that tocolytic therapy was administered more often in the induction group (10.7% vs. 6.5%, p < 0.001). However, the fetal acidosis rates did not differ between the two groups [40].

This retrospective cohort study from a university hospital, level I perinatal center of the effectiveness and safety of low-dose and high-dose misoprostol regimens contributes additional data to the existing literature. The approval of this induction medication which our results found to be effective with few side effects is useful for obstetric care. It qualifies the Red Hand Letter and provides a defense against possible lawsuits and shows that, despite manual division of the tablets as was previously done without being able to ensure that the planned dose was achieved, the outcomes were comparable with those reported for the medication recently approved for induction [12]. Possible differences in the bioavailability of misoprostol do not appear to have had a negative impact [41]. Even if a precise cost analysis has not yet been done, it can be assumed that induction with the LDM regimen is more expensive. First and foremost, the approved medication is more expensive, the induction time is somewhat longer and, on average, prostaglandin gels are used more often over the course of the induction. Whether this justifies producing the medication oneself or whether there should be a change in the induction regimen remains to be elucidated [42].

Because of the retrospective nature of the study, only written data recorded during treatment were analyzed (attrition bias). The opioid requirements permit only limited and indirect conclusions to be made about the level of pain experienced; patients were not surveyed with a pain scale to determine their levels of pain.

The study population was very homogeneous. The few inhomogeneities are potentially due to the constantly updated treatment protocols. The risk of confounders was addressed statistically in our study which limited the selection bias.


Summary

In summary, both dosing regimens achieve high vaginal birth rates. Our study found no differences relating to side effects from oral misoprostol administration between the regimen of 50 µg and the regimen of 100 µg every four hours with the exception of higher opioid requirements.



Conflict of Interest

The authors declare that they have no conflict of interest.


Correspondence

Dr. Vivien Dütemeyer, PhD
Klinik für Frauenheilkunde und Geburtshilfe, Medizinische Hochschule Hannover
Carl-Neuberg-Straße 1
30625 Hannover
Germany   

Publication History

Received: 20 May 2025

Accepted after revision: 27 September 2025

Article published online:
04 November 2025

© 2025. 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


Zoom
Fig. 1 Flow chart of the study population (Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 2009; 6: e1000097. DOI: 10.1371/journal.pmed1000097).
Zoom
Abb. 1 Flowchart der Studienpopulation (Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 2009; 6: e1000097. DOI: 10.1371/journal.pmed1000097).