Comparative study of oral and vaginal misoprostol for induction of labour, maternal and foetal outcome

The efficacy and safety of 25 μg or 50 μg oral misoprostol versus 25 μg vaginal misoprostol given at 4- or 6-hourly intervals for induction of labour in women at or beyond term with live singleton pregnancies: A systematic review and meta-analysis

BACKGROUND

Misoprostol is widely used for cervical ripening and labour induction as it is heat-stable and inexpensive. Oral misoprostol 25 μg given 2-hourly is recommended over vaginal misoprostol 25 μg given 6-hourly, but the need for 2-hourly fetal monitoring makes oral misoprostol impractical for routine use in high-volume obstetric units in resource-constrained settings.

OBJECTIVES

To compare the efficacy and safety of oral misoprostol initiated at 25 or 50 μg versus 25 μg vaginal misoprostol given at 4- to 6-hourly intervals for labor induction in women at or beyond term (≥ 37 weeks) with a single viable fetus and an unscarred uterus.

SEARCH STRATEGY

We identified eligible randomized, parallel-group, labor-induction trials from recent systematic reviews. We additionally searched PubMed, Cochrane CENTRAL, Epistemonikos, and clinical trials registries from February 1, 2020 to December 31, 2022 without language restrictions. Database-specific keywords for cervical priming, labor induction, and misoprostol were used.

SELECTION CRITERIA

We excluded labor-induction trials exclusively in women with ruptured membranes, in the third trimester, and those that initiated misoprostol at doses not specified in the review's objectives. The primary outcomes were vaginal birth within 24 h, cesarean section, perinatal mortality, neonatal morbidity, and maternal morbidity. The secondary outcomes were uterine hyperstimulation with fetal heart rate changes, and oxytocin augmentation.

DATA COLLECTION AND ANALYSIS

Two or more authors selected studies independently, assessed risk of bias, and extracted data. We derived pooled weighted risk ratios with 95% confidence intervals (CIs) for each outcome, subgrouping trials by the dose and frequency of misoprostol regimens. We used the I2 statistic to quantify heterogeneity and the random-effects model for meta-analysis when appropriate. We used the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) approach to assess certainty (confidence) in the effect estimates.

MAIN RESULTS

Thirteen trials, from Canada, India, Iran, and the US, randomizing 2941 women at ≥37 weeks of gestation with an unfavorable cervix (Bishop score <6), met the eligibility criteria. Five misoprostol regimens were compared: 25 μg oral versus 25 μg vaginal, 4-hourly (three trials); 50 μg oral versus 25 μg vaginal, 4-hourly (five trials); 50 μg followed by 100 μg oral versus 25 μg vaginal, 4-hourly (two trials); 50 μg oral, 4-hourly versus 25 μg vaginal, 6-hourly (one trial); and 50 μg oral versus 25 μg vaginal, 6-hourly (two trials). The overall certainty in the evidence ranged from moderate to very low, due to high risk of bias in 11/13 trials (affecting all outcomes), unexplained heterogeneity (1/7 outcomes), indirectness (1/7 outcomes), and imprecision (4/7 outcomes). Vaginal misoprostol probably increased vaginal deliveries within 24 h compared with oral misoprostol (risk ratio [RR] 0.82, 95% CI 0.70-0.96; 11 trials, 2721 mothers; moderate-certainty evidence); this was more likely with 4-hourly than with 6-hourly vaginal regimens. The risk of cesarean sections did not appreciably differ (RR 1.00, 95% CI 0.80-1.26; 13 trials, 2941 mothers; very low-certainty evidence), although oral misoprostol 25 μg 4-hourly probably increased this risk compared with 25 μg vaginal misoprostol 4-hourly (RR 1.69, 95% CI 1.21-2.36; three trials, 515 mothers). The risk of perinatal mortality (RR 0.67, 95% CI 0.11-3.90; one trial, 196 participants; very low-certainty evidence), neonatal morbidity (RR 0.84, 95% CI 0.67-1.06; 13 trials, 2941 mothers; low-certainty evidence), and maternal morbidity (RR 0.83, 95% CI 0.48-1.44; 6 trials; 1945 mothers; moderate-certainty evidence) did not differ appreciably. The risk of uterine hyperstimulation with fetal heart rate changes may be lower with oral misoprostol (RR 0.70, 95% CI 0.52-0.95; 10 trials, 2565 mothers; low-certainty evidence). Oxytocin augmentation was probably more frequent with oral compared with vaginal misoprostol (RR 1.29, 95% CI 1.10-1.51; 13 trials, 2941 mothers; moderate-certainty evidence).

CONCLUSIONS

Low-dose, 4- to 6-hourly vaginal misoprostol regimens probably result in more vaginal births within 24 h and less frequent oxytocin use compared with low-dose, 4- to 6-hourly, oral misoprostol regimens. Vaginal misoprostol may increase the risk of uterine hyperstimulation with fetal heart changes compared with oral misoprostol, without increasing the risk of perinatal mortality, neonatal morbidity, or maternal morbidity. Indirect evidence indicates that 25 μg vaginal misoprostol 4-hourly may be more effective and as safe as the recommended 6-hourly vaginal regimen. This evidence could inform clinical decisions in high-volume obstetric units in resource-constrained settings.

Low-dose oral misoprostol for induction of labour

BACKGROUND

Misoprostol given orally is a commonly used labour induction method. Our Cochrane Review is restricted to studies with low-dose misoprostol (initially ≤ 50 µg), as higher doses pose unacceptably high risks of uterine hyperstimulation.

OBJECTIVES

To assess the efficacy and safety of low-dose oral misoprostol for labour induction in women with a viable fetus in the third trimester of pregnancy.

SEARCH METHODS

We searched Cochrane Pregnancy and Childbirth's Trials Register, ClinicalTrials.gov,  the WHO International Clinical Trials Registry Platform (14 February 2021) and reference lists of retrieved studies.

SELECTION CRITERIA

Randomised trials comparing low-dose oral misoprostol (initial dose ≤ 50 µg) versus placebo, vaginal dinoprostone, vaginal misoprostol, oxytocin, or mechanical methods; or comparing oral misoprostol protocols (one- to two-hourly versus four- to six-hourly; 20 µg to 25 µg versus 50 µg; or 20 µg hourly titrated versus 25 µg two-hourly static).

DATA COLLECTION AND ANALYSIS

Using Covidence, two review authors independently screened reports, extracted trial data, and performed quality assessments. Our primary outcomes were vaginal birth within 24 hours, caesarean section, and hyperstimulation with foetal heart changes.

MAIN RESULTS

We included 61 trials involving 20,026 women. GRADE assessments ranged from moderate- to very low-certainty evidence, with downgrading decisions based on imprecision, inconsistency, and study limitations. Oral misoprostol versus placebo/no treatment (four trials; 594 women) Oral misoprostol may make little to no difference in the rate of caesarean section (risk ratio (RR) 0.81, 95% confidence interval (CI) 0.59 to 1.11; 4 trials; 594 women; moderate-certainty evidence), while its effect on uterine hyperstimulation with foetal heart rate changes is uncertain (RR 5.15, 95% CI 0.25 to 105.31; 3 trials; 495 women; very low-certainty evidence). Vaginal births within 24 hours was not reported. In all trials, oxytocin could be commenced after 12 to 24 hours and all women had pre-labour ruptured membranes. Oral misoprostol versus vaginal dinoprostone (13 trials; 9676 women) Oral misoprostol probably results in fewer caesarean sections (RR 0.84, 95% CI 0.78 to 0.90; 13 trials, 9676 women; moderate-certainty evidence). Subgroup analysis indicated that 10 µg to 25 µg (RR 0.80, 95% CI 0.74 to 0.87; 9 trials; 8652 women) may differ from 50 µg (RR 1.10, 95% CI 0.91 to 1.34; 4 trials; 1024 women) for caesarean section. Oral misoprostol may decrease vaginal births within 24 hours (RR 0.93, 95% CI 0.87 to 1.00; 10 trials; 8983 women; low-certainty evidence) and hyperstimulation with foetal heart rate changes (RR 0.49, 95% CI 0.40 to 0.59; 11 trials; 9084 women; low-certainty evidence). Oral misoprostol versus vaginal misoprostol (33 trials; 6110 women) Oral use may result in fewer vaginal births within 24 hours (average RR 0.81, 95% CI 0.68 to 0.95; 16 trials, 3451 women; low-certainty evidence), and less hyperstimulation with foetal heart rate changes (RR 0.69, 95% CI 0.53 to 0.92, 25 trials, 4857 women, low-certainty evidence), with subgroup analysis suggesting that 10 µg to 25 µg orally (RR 0.28, 95% CI 0.14 to 0.57; 6 trials, 957 women) may be superior to 50 µg orally (RR 0.82, 95% CI 0.61 to 1.11; 19 trials; 3900 women). Oral misoprostol probably does not increase caesarean sections overall (average RR 1.00, 95% CI 0.86 to 1.16; 32 trials; 5914 women; low-certainty evidence) but likely results in fewer caesareans for foetal distress (RR 0.74, 95% CI 0.55 to 0.99; 24 trials, 4775 women). Oral misoprostol versus intravenous oxytocin (6 trials; 737 women, 200 with ruptured membranes) Misoprostol may make little or no difference to vaginal births within 24 hours (RR 1.12, 95% CI 0.95 to 1.33; 3 trials; 466 women; low-certainty evidence), but probably results in fewer caesarean sections (RR 0.67, 95% CI 0.50 to 0.90; 6 trials; 737 women; moderate-certainty evidence). The effect on hyperstimulation with foetal heart rate changes is uncertain (RR 0.66, 95% CI 0.19 to 2.26; 3 trials, 331 women; very low-certainty evidence). Oral misoprostol versus mechanical methods (6 trials; 2993 women) Six trials compared oral misoprostol to transcervical Foley catheter. Misoprostol may increase vaginal birth within 24 hours (RR 1.32, 95% CI 0.98 to 1.79; 4 trials; 1044 women; low-certainty evidence), and probably reduces the risk of caesarean section (RR 0.84, 95% CI 0.75 to 0.95; 6 trials; 2993 women; moderate-certainty evidence). There may be little or no difference in hyperstimulation with foetal heart rate changes (RR 1.31, 95% CI 0.78 to 2.21; 4 trials; 2828 women; low-certainty evidence). Oral misoprostol one- to two-hourly versus four- to six-hourly (1 trial; 64 women) The evidence on hourly titration was very uncertain due to the low numbers reported. Oral misoprostol 20 µg hourly titrated versus 25 µg two-hourly static (2 trials; 296 women) The difference in regimen may have little or no effect on the rate of vaginal births in 24 hours (RR 0.97, 95% CI 0.80 to 1.16; low-certainty evidence). The evidence is of very low certainty for all other reported outcomes.

AUTHORS' CONCLUSIONS

Low-dose oral misoprostol is probably associated with fewer caesarean sections (and therefore more vaginal births) than vaginal dinoprostone, and lower rates of hyperstimulation with foetal heart rate changes. However, time to birth may be increased, as seen by a reduced number of vaginal births within 24 hours. Compared to transcervical Foley catheter, low-dose oral misoprostol is associated with fewer caesarean sections, but equivalent rates of hyperstimulation. Low-dose misoprostol given orally rather than vaginally is probably associated with similar rates of vaginal birth, although rates may be lower within the first 24 hours. However, there is likely less hyperstimulation with foetal heart changes, and fewer caesarean sections performed due to foetal distress. The best available evidence suggests that low-dose oral misoprostol probably has many benefits over other methods for labour induction. This review supports the use of low-dose oral misoprostol for induction of labour, and demonstrates the lower risks of hyperstimulation than when misoprostol is given vaginally. More trials are needed to establish the optimum oral misoprostol regimen, but these findings suggest that a starting dose of 25 µg may offer a good balance of efficacy and safety.

Which method is best for the induction of labour? A systematic review, network meta-analysis and cost-effectiveness analysis

BACKGROUND

More than 150,000 pregnant women in England and Wales have their labour induced each year. Multiple pharmacological, mechanical and complementary methods are available to induce labour.

OBJECTIVE

To assess the relative effectiveness, safety and cost-effectiveness of labour induction methods and, data permitting, effects in different clinical subgroups.

METHODS

We carried out a systematic review using Cochrane methods. The Cochrane Pregnancy and Childbirth Group's Trials Register was searched (March 2014). This contains over 22,000 reports of controlled trials (published from 1923 onwards) retrieved from weekly searches of OVID MEDLINE (1966 to current); Cochrane Central Register of Controlled Trials (The Cochrane Library); EMBASE (1982 to current); Cumulative Index to Nursing and Allied Health Literature (1984 to current); ClinicalTrials.gov; the World Health Organization International Clinical Trials Registry Portal; and hand-searching of relevant conference proceedings and journals. We included randomised controlled trials examining interventions to induce labour compared with placebo, no treatment or other interventions in women eligible for third-trimester induction. We included outcomes relating to efficacy, safety and acceptability to women. In addition, for the economic analysis we searched the Database of Abstracts of Reviews of Effects, and Economic Evaluations Databases, NHS Economic Evaluation Database and the Health Technology Assessment database. We carried out a network meta-analysis (NMA) using all of the available evidence, both direct and indirect, to produce estimates of the relative effects of each treatment compared with others in a network. We developed a de novo decision tree model to estimate the cost-effectiveness of various methods. The costs included were the intervention and other hospital costs incurred (price year 2012-13). We reviewed the literature to identify preference-based utilities for the health-related outcomes in the model. We calculated incremental cost-effectiveness ratios, expected costs, utilities and net benefit. We represent uncertainty in the optimal intervention using cost-effectiveness acceptability curves.

RESULTS

We identified 1190 studies; 611 were eligible for inclusion. The interventions most likely to achieve vaginal delivery (VD) within 24 hours were intravenous oxytocin with amniotomy [posterior rank 2; 95% credible intervals (CrIs) 1 to 9] and higher-dose (≥ 50 µg) vaginal misoprostol (rank 3; 95% CrI 1 to 6). Compared with placebo, several treatments reduced the odds of caesarean section, but we observed considerable uncertainty in treatment rankings. For uterine hyperstimulation, double-balloon catheter had the highest probability of being among the best three treatments, whereas vaginal misoprostol (≥ 50 µg) was most likely to increase the odds of excessive uterine activity. For other safety outcomes there were insufficient data or there was too much uncertainty to identify which treatments performed 'best'. Few studies collected information on women's views. Owing to incomplete reporting of the VD within 24 hours outcome, the cost-effectiveness analysis could compare only 20 interventions. The analysis suggested that most interventions have similar utility and differ mainly in cost. With a caveat of considerable uncertainty, titrated (low-dose) misoprostol solution and buccal/sublingual misoprostol had the highest likelihood of being cost-effective.

LIMITATIONS

There was considerable uncertainty in findings and there were insufficient data for some planned subgroup analyses.

CONCLUSIONS

Overall, misoprostol and oxytocin with amniotomy (for women with favourable cervix) is more successful than other agents in achieving VD within 24 hours. The ranking according to safety of different methods was less clear. The cost-effectiveness analysis suggested that titrated (low-dose) oral misoprostol solution resulted in the highest utility, whereas buccal/sublingual misoprostol had the lowest cost. There was a high degree of uncertainty as to the most cost-effective intervention.

FUTURE WORK

Future trials should be powered to detect a method that is more cost-effective than misoprostol solution and report outcomes included in this NMA.

STUDY REGISTRATION

This study is registered as PROSPERO CRD42013005116.

FUNDING

The National Institute for Health Research Health Technology Assessment programme.