Labour characteristics and uterine activity: misoprostol compared with oxytocin in women at term with prelabour rupture of the membranes


Correspondence: Dr S. W. Ngai, Department of Obstetrics and Gynaecology, University of Hong Kong, Hong Kong.


Objective To compare the labour pattern and uterine activity of oral misoprostol with oxytocin for labour induction in women presenting with prelabour rupture of membranes at term.

Design Prospective randomised study.

Setting Department of Obstetrics and Gynaecology, Queen Mary Hospital, Hong Kong.

Participants Eighty women presenting with prelabour rupture of membranes at term.

Methods The women were randomised to receive either 100 μg misoprostol orally every 4 hours to a maximum of three doses, or intravenous oxytocin infusion according to the hospital protocol. Intrauterine pressure transducers were inserted one hour before induction of labour in both groups of women. We compared the pattern of uterine activity, the induction-to-delivery interval, duration of labour, mode of delivery and neonatal outcome between the two groups.

Results Both oxytocin and oral misoprostol caused an increase in uterine activity within one hour of labour induction. Peak uterine activity was reached 6–8 h after oral misoprostol, with persistent effects, and 8–10 h after oxytocin, requiring continuous titration of medication. The duration of labour was significantly reduced in nulliparous women, but not in those who were multiparous in the misoprostol group. The induction-to-delivery interval, the mode of delivery and the perinatal outcome were similar for the two groups.

Conclusion Oral misoprostol caused earlier peak uterine activity, compared with oxytocin (6–8 h vs 8–10 h). Oral misoprostol was not only as effective as oxytocin in inducing labour in women at term with prelabour rupture of the membranes, but it reduced significantly the duration of labour in nulliparous women.


In the management of patients presenting with prelabour rupture of the membranes (PROM) at term, an active approach is desirable because a prolonged interval from the time of PROM to delivery is associated with increased incidence of histologic chorioamnionitis and funisitis1,2. Furthermore, maternal and neonatal morbidity is also increased3,4. However, many different regimens have been reported from different centres in the management of PROM at term. Titration with intravenous oxytocin is the commonest method used, but this is associated with restriction of patients and continuous assessment is required for dosage control. Since the early 1970s, prostaglandin preparations (PGE2 pessaries) have been used successfully for cervical ripening and induction of labour in women with PROM5–7. However, there is a risk of introducing ascending infection during the vaginal application of these preparations. Oral PGE2 tablets have been tried in other studies8–10, which, although effective, were associated with a significant number of gastrointestinal side effects9. Furthermore, the oral PGE2 tablets have not been shown to be superior to intravenous oxytocin8,9.

Misoprostol is a stable, orally active synthetic prostaglandin E1 analogue, used in prophylaxis of peptic ulcers. It is effective in priming of the cervix when administered either vaginally or orally11,12. We have reported previously that a single oral dose of 200 μg is effective for cervical priming in women at term with PROM and it may even have an effect on induction of labour12. We conducted this study to compare the pattern of labour and uterine activity in oral misoprostol with oxytocin for induction of labour in women at term presenting with PROM.


This randomised study was conducted in Tsan Yuk Hospital, Hong Kong from January 1996 to March 1997. During this time 4861 women were delivered in our department and 4930 infants were born. Eighty-six women presenting with PROM at term were recruited, counselled and informed consent obtained. The study was approved by the Ethics Committee, Faculty of Medicine, University of Hong Kong. The inclusion criteria were: the membranes had been ruptured for > 12 hours; there was no clinical evidence of infection; reactive nonstress test; cephalic presentation; and a singleton pregnancy. Women with a previous uterine scar, meconium stained amniotic fluid or a history of allergy to prostaglandins were excluded. Patients who were admitted for PROM occurring after 37 completed weeks of gestation were recruited into the study when no signs of labour were noted ≥ 12 hours following rupture of membranes.

A speculum examination was performed upon admission in every patient to confirm the diagnosis of PROM and a baseline Bishop score was obtained at the same time, and a high vaginal swab for bacteriological culture was taken.

Following recruitment, the women were allocated randomly to one of the two treatment groups (misoprostol or oxytocin), following the randomisation schedule described by Meinert13. The random number was sealed in an envelope which would be unsealed after recruitment. Multiparous and nulliparous women were randomised separately. In the misoprostol group the induction regimen was oral misoprostol 100 μg, followed by repeated doses 100 μg every 4 hours for a maximum of three doses. Augmentation with oxytocin infusion was given if active labour did not begin with oral misoprostol alone. The dosage of misoprostol used in this study was based on our previous study in which a single dose of 200 μg oral misoprostol was given for cervical priming and 82% of the women went into labour12. In this study we planned to administer repeated doses, therefore we decided to decrease the dosage to 100 μg in order to assess how well the patients responded. Also, we wanted to test for the lowest effective dosage.

In the oxytocin group induction of labour was performed with oxytocin administered intravenously according to our established hospital protocol. The starting dose was 1 mIU/min. If optimal uterine contractions (moderate to strong in character, at intervals of 3–5 minutes) did not occur, the dosage of oxytocin infusion was doubled every 15 minutes up to a maximal dosage of 32 mIU/min. Maternal vital signs were evaluated every 15 minutes, and fetal heart rate and uterine activity was monitored continuously.

Every patient was connected to a continuous fetal heart rate monitor, intrauterine pressure catheters were inserted in both groups of patients, and the progress of labour was reassessed by the same investigator. Onset of labour was diagnosed by either regular uterine contractions or progressive cervical dilatation. Failed induction was defined as no appreciable change in the cervix after 12 hours of adequate uterine contractions or as no progressive increase in cervical dilatation after more than 2 hours in the active phase of labour. Patients in the misoprostol group were given oxytocin if active labour was not established after three doses of oral misoprostol.

Uterine activity was monitored for 1 hour before instillation of either oxytocin or misoprostol. Intrauterine pressure transducer (Sonacaid, Utah, USA) was inserted prior to induction of labour. Monitoring data from each case were stored in the recording unit. The printout on the strip (paper speed 1 cm/min) gave starting time and duration of monitoring, presence or absence of monitoring gaps, and warnings about any artifacts, poor connection, and end of transmission.

Two interpreters were used to assure reliability of interpretation. Uterine activity patterns were analysed as described by Sheerer et al.14. These patterns were divided into two groups according to the amplitude of ≤ 5 mm on the tocograph tracing; those which occurred at 1 to 2 minute intervals represented a low amplitude, high frequency pattern. A second group of contractions consisted of those with a high amplitude (≥ 6 mm), lasting at least 30 s and occurring less frequently. We also sought evidence for uterine hyperstimulation, described as a contraction frequency of > 5 in 10 minutes or contractions exceeding a 2 minute duration. For each case, the proportion of time occupied by any contractions and the number of high amplitude contractions was recorded for each hourly period.

Maternal characteristics including parity, gestational age, baseline Bishop score, maternal vital signs, incidence of side effects, PROM-to-delivery interval, induction-to-delivery interval, duration of labour, mode of delivery, analgesic requirement, maternal and fetal morbidity and cord pH were compared.

The calculation of sample size was based on the expected difference of the induction-to-delivery interval between the two groups. Based on our previous study, the established protocol led to a mean (SD) vaginal delivery in 450 (360) minutes. A 240-minute difference was chosen as clinically significant. A sample size of 40 women per group would be required to show this difference with a two-tailed alpha of 0.05 and power of 80%.

The nulliparous and multiparous women were analysed separately. The differences in discontinuous variables were analysed by χ2 test and Fisher's exact test. The difference in continuous variables was analysed by the Student's t test for normally distributed data and Mann-Whitney U test for the skewed data.


Eighty-six women were recruited and randomised, but six were excluded from the final analysis because they did not fulfill the study criteria. One presented with undiagnosed breech during first stage of labour in the oxytocin group. An emergency lower segment caesarean section was performed for breech and unfavourable pelvis. Five patients did not have baseline cervical score recorded (two and three in the misoprostol and oxytocin groups, respectively). All of them had normal vaginal delivery. Sixty-six nulliparous and 14 multiparous women were recruited and distributed equally between the misoprostol and oxytocin groups. The two groups were comparable with respect to maternal age, gestational age, initial Bishop scores, and duration of PROM prior to recruitment (Table 1). Three women in the oxytocin group and two in the misoprostol group were given epidural analgesia, while others used intramuscular pethidine or inhaled Entonox. The type of analgesia used was comparable between the two groups and probably had no effect on the pattern of labour.

Table 1.  Clinical characteristics of women in the misoprostol and oxytocin groups. Values are given as mean (SD), unless otherwise indicated. ROM = rupture of membranes.
 Misoprostol (n= 40)Sxytocin (n= 40)
Age (years)28.9 (4.0)30.7 (3.0)
Gestation at ROM (weeks)38.39 (1.7)39.4 (1.5)
Initial Bishop score5.8 (1.2)6.0 (1.4)
Median [range]5.5[3–8]5.6 [3–9]
ROM to recruitment interval(h)21.8 (6.7)21.3 (7.4)
Percentage of primigravida {n}82.5 {33/40}82.5 {33/40}

Eighty women had the intrauterine pressure transducer inserted. Five were excluded due to incomplete data. Figure 1 shows the mean hourly high amplitude contraction frequencies and the proportion of time of any uterine activity for the two treatment groups. Both groups showed statistically significant increase in uterine activity after dosing (Friedman analysis of variance, P < 0.001). All women had increased uterine activity within the first hour. The misoprostol group showed a higher percentage time of total uterine contraction in the first 4 hours.

Figure 1.

Mean hourly high amplitude contraction frequency and percent time of total uterine contractions for oral misoprostol (□) and oxytocin. (×).

For those treated with oral misoprostol, mean uterine activity reached a peak within 6–8 hours after the medication. Increased in high amplitude uterine activity was observed in the first 1 to 2 hours and persisted for the rest of labour. Only 40% of women required a second dose of misoprostol. In the oxytocin group, the mean uterine activity rosed steadiy and reached the peak at 10–12 hours after treatment. Continuous infusion of oxytocin was required until delivery.

The outcome of labour is shown in Table 2. In the misoprostol group 24 women (60%) required a single dose (100 μg), while the remaining 16 (40%) required a second dose (200 μg) only. No women required oxytocin for augmentation of labour. The duration of labour in nulliparous and multiparous women was analysed separately. In the nulliparous women the overall duration of labour was significantly shorter in the misoprostol group (5.3 h vs 10.1 h, P= 0.004). A significant difference was seen both for the first and second stages of labour (5.0 h vs 9.6 h, P= 0.03 and 0.6 h vs 1.2 h, P= 0.01, respectively). Although the induction-to-delivery interval in the nulliparous women was also shorter in the misoprostol group, this difference did not reach statistical significance. In the multiparous women the first, second and third stages of labour and the overall duration of labour were similar between the two groups. The induction-to-delivery interval just reached statistical significance (P= 0.05). However, the number of multiparous women recruited was inadequate to make a conclusion here. Thirteen women (32.5%) in the misoprostol group and 10 (25.0%) in the oxytocin group showed tachysystole. No women in either group suffered from uterine hyperstimulation.

Table 2.  Outcome of labour in the misoprostol and oxytocin groups. Values are given as mean (SD).
 Misoprostol (n= 40)Oxytocin (n= 40)P
Nulliparous(n= 33)(n= 33) 
1st stage (h)5.0 (2.9)9.6 (5.0)0.03*
2nd stage (h)0.6 (0.3)1.2 (0.8)0.01*
3rd stage (min)2.2 (3.7)3.5 (4.0)0.67
Duration of labour (h)5.3 (2.5)10.1 (5.6)0.004*
Induction to delivery interval (h)7.3 (3.1)11.1 (4.9)0.24
Multiparous(n= 7)(n= 7) 
1st stage (h)2.3 (1.5)4.3 (4.5)0.1
2nd stage (h)0.4 (0.4)0.3 (0.2)0.2
3rd stage (min)0.9 (2.0)94 (5.2)0.06
Duration of labour (h)3.0 (1.7)5.3 (4.1)0.3
Induction to delivery interval (h)3.7 (1.2)7.5 (6.5)0.05

There was no significant difference in the mode of delivery between the two groups (Table 3). The caesarean section rate was similar between the two groups. The indication for caesarean delivery in the misoprostol group was cephalopelvic disproportion (n= 2). The indications for caesarean delivery in oxytocin group were cephalopelvic disproportion (n= 2) and failed induction (n= 1).

Table 3.  Mode of delivery in the misocprotol and oxytocin groups. Values are given as n (%).
 Misoprostol (n= 40)Oxytocin (n= 40)
Spontaneous28 (70)27 (67.5)
Vacuum extraction8 (20)8 (20)
Low forceps2 (5)2 (5)
Caesarean section2 (5)3 (7.5)

The neonatal outcome in both groups was comparable (Table 4). No babies had an Apgar score of < 7 at 5 minutes of life. The mean birthweight and umbilical artery pH were similar in both groups. There were no significant differences in the number of admissions to the special care neonatal unit.

Table 4.  Neonatal outcome in misoprostol and oxytocin groups. Values are given as n, or mean (SD).
 Misoprostol (n= 40)Oxytocin (n= 40)
Apgar < 7 at 5 min00
Birthweight (g)3206 (413)3286 (491)
Neonatal infection11
Umbilical artery pH7.3 (0.2)7.3 (0.2)
Admission to special-care neonatal unit34

The high vaginal swab showed positive culture of Group B streptococcal infection in one and three women in the misoprostol and oxytocin groups, respectively. Patients were given antibiotic treatment accordingly.


Prelabour rupture of membranes is a common indication for labour induction. Oxytocin and prostaglandins are the agents most frequently used for this purpose. Although oxytocin is widely accepted as a safe and effective initiator of uterine contractions, its success is dependent on the condition of the cervix at the beginning of induction. Thus, women who present with PROM and an unfavorable cervix may have a higher chance of caesarean section if labour is induced by oxytocin. Induction of labour with prostaglandins offers the advantage of promoting both cervical ripening and myometrial contractility in this group of women. A drawback of prostaglandins is their ability to induce excessive uterine contractility, which can lead to perinatal and maternal morbidity14–16. Also, frequent vaginal administration of prostaglandin pessary may increase the risk of ascending infection in PROM patients.

Vaginal misoprostol has been shown to be effective in labour induction at doses of 50–100 μg. There are few data in the literature regarding the ideal dose in tetric patients. In one report 6000 μg of misoprostol was taken at 31 weeks of gestation and resulted in hypertonic uterine contractions which led to fetal death and spontaneous delivery afterwards, although all the systemic side effects subsided within 12 hours17. In our previous study to investigate the effectiveness of single dose of misoprostol for cervical priming in PROM patients, the results showed that misoprostol 200 μg given orally was effective in improving the Bishop score and led to a reduction in the incidence of oxytocin infusion for labour induction, together with a reduction in leaking-to-delivery interval. The incidence (87%) of patients in the misoprostol group who went into labour within 12 hours of medication was comparable to the incidence of successful induction of labour using vaginal preparations, as reported in the literature. Thus, we conducted this study to compare oral misoprostol with oxytocin for the effectiveness of labour induction in PROM patients.

Windrim et al.18 recently reported a randomised trial comparing oral misoprostol versus vaginal dinoprostone for labour induction in patients with intact membranes. They used 50 μg every 4 hours for two doses. After two 50 μg doses, if a satisfactory contraction pattern with cervical softening, effacement, and dilation had not occurred, misoprostol was increased to 100 μg orally at 4-hourly intervals. The control groups were managed by 0.5 mg dinoprostone intracervically every 6 hours or with dilute oxytocin infusion, as considered appropriate by the attending physician. There was no statistically significant difference in the induction-to-delivery interval between the two groups. The secondary outcome measures (e.g. caesarean rate, frequency of epidural use, perineal trauma, or manual removal of the placenta) were also similar in both groups. They suggested that oral misoprostol may be a new option for labour induction.

The oral misoprostol protocol in our study differs from that of Windrim et al.18 in that we used 100 μg instead of 50 μg, and even though at a higher dosage, no women suffered from uterine hyperstimulation. Furthermore, the women in our study were confined to those presenting with ruptured membranes. Their response to exogenous prostaglandins may be different from those with intact membranes.

This is the first study to compare the pattern of uterine activity in patients treated by misoprostol versus oxytocin for labour induction. Low amplitude, high frequency, and high amplitude patterns of contractions were measured with intrauterine pressure transducers. Significant increases in total uterine activity were found with both misoprostol and oxytocin. Misoprostol leads to an increase in uterine activity reaching a maximum value within 5 hours after the induction of labour. In the oxytocin group the maximum uterine activity was attained at about 10 hours after induction of labour. Zie-man et al.19 compared the absorption kinetics between oral and vaginal administration of misoprostol. They showed that in women receiving oral misoprostol, the plasma concentration of misoprostol rose quickly, peaked between 12.5 and 60 minutes after administration, fell steeply by 120 minutes, and remained low for the duration of the study. These findings explain the rapid increased in uterine activity after oral medication in our study, but they cannot not explain the persistency of the uterine activity after a single dose of oral misoprostol. We postulate that misoprostol may initiate the endogenous prostanglandins secretion which simulates spontaneous onset of labour. However, this has to be proven by further studies.

The incidence of misoprostol-associated tachysystole was increased significantly when compared with oxytocin20,21. We demonstrated that the uterine activity reached the peak 5 hours after medication. Therefore, to repeat the second dose at a 6-hour interval, instead of every 4 hours, may avoid excessive uterine activity. Alternatively, a lowering of second dose may be useful to decrease the incidence of tachysystole.

The caesarean section rate in our hospital was 16% in 1997. The use of misoprostol did not lead to a further reduction of caesarean section rate, which was quite low in both the misoprostol (5%) and oxytocin (7.5%) groups. The size of the study might not have been adequate to show any significant difference, if any. Nonetheless, although the caesarean rate was not different, the use of misoprostol reduced the duration of first and second stages of labour in nulliparous women. No patient in the study group required oxytocin for augmentation of labour. This would have reduced the stress to the patients and the risk of maternal and neonatal infections.

Finally, the cost of various induction regimens must be considered. Misoprostol is inexpensive when the drug and staff costs are considered. It is cheap when compared with prostin (50 pence for 400 μg misoprostol vs£24 for 3 mg vaginal prostin). It is less expensive when compared with intravenous oxytocin infusion since a midwife or doctor is required in the later for continuous titration and dosage control.

We conclude that oral misoprostol at a dose of 100 μg every 4 hours not only is as successful as oxytocin for labour induction in women presenting with prelabour rupture of the membranes at term, but also reduced the duration of labour in nulliparous women. Moreover, the recent report by Seaward et al.22 has clearly demonstrated an association between chorioamnionitis with nulliparous women, the duration of active labour, and a dose-response type of relation with the number of vaginal examination in patients with PROM at term. Therefore, the use of misoprostol for labour induction in nulliparous patients with PROM at term could provide the additional benefits of reducing the risk of chorioamnionitis. Observations from the current study provide more precise knowledge about patterns of uterine activity with oral misoprostol for labour induction. Uterine activity may be initiated by either oxytocin or misoprostol, with activity persisting more frequently in the misoprostol group. Our data adds further information to the formulation of induction regimen using misoprostol. We suggest that the second dose should be given at an interval of 6 hours, instead of 4 hours, to avoid excessive uterine activity. However, our experience in multiparous women is limited and further studies are required to define the optimal regimen for labour induction in this group of women. Further evaluation on the incidence of side effects, maternal and neonatal safety is required.