Commentary on selected new evidence

With advice from topic experts we selected one study for further commentary.

Methods that are recommended for the induction of labour and that are not recommended for the induction of labour

We selected a systematic review, network meta-analysis and cost-effectiveness analysis by Alfirevic et al. 2016 for a full commentary because the results were found to have a potential impact on the guideline recommendations.

What the guideline recommends

The guideline recommends vaginal prostaglandin E2 (PGE2) as the preferred method for the induction of labour. Misoprostol and mifepristone are recommended only for the induction of labour in women who have intrauterine fetal death. The guideline does not recommend amniotomy, alone or with oxytocin, as a primary method of induction of labour, unless there are specific clinical reasons not to use vaginal PGE2.


A systematic review, network meta-analysis (NMA) and incremental cost-effectiveness analysis was undertaken.

The systematic review searched for all randomised control trials (RCTs) published before March 2014. All RCTs on mechanical, pharmacological and non-pharmacological methods, compared to placebo, no treatment or other interventions, to induce labour in women in their third trimester of pregnancy were included.

The systematic review searched for the following primary outcomes:

  • vaginal delivery not achieved within 24 hours (VD24)

  • uterine hyperstimulation with fetal heart rate changes

  • caesarean section

  • serious neonatal morbidity or death

  • serious maternal morbidity or death

  • instrumental delivery

  • maternal satisfaction with the method used

  • neonatal intensive care unit (NICU) admission

  • Apgar score at 5 minutes of <7.

An NMA was conducted for each of 6 outcomes where sufficient data were available. The NMA used both direct and indirect evidence and compared consistency of these in analysis to determine model fit.

A de novo decision-tree model was used for the cost-effectiveness analysis. The costs were taken from the manufacturers, NHS reference costs and the British National Formulary. Several outcomes were accounted for in the model.


The review included 611 studies on more than 100,000 women undergoing induction of labour using 31 interventions of different dosages and routes of administration.

The interventions included in the cost-effectiveness analysis and the systematic review were:

  • titrated (low dose) oral misoprostol solution

  • buccal/sublingual misoprostol

  • vaginal misoprostol <50 µg

  • vaginal misoprostol ≥50 µg

  • oral misoprostol tablet <50 µg

  • oral misoprostol tablet ≥50 µg

  • sustained release misoprostol vaginal pessary

  • vaginal PGE2 (gel)

  • vaginal PGE2 pessary (normal release)

  • intracervical PGE2

  • extra-amniotic PGE2

  • vaginal PGE2 pessary (slow release)

  • vaginal PGE2 (tablet)

  • intravenous oxytocin with amniotomy

  • intravenous oxytocin

  • mifepristone

  • double balloon or Cook's catheter

  • foley catheter

  • nitric oxide.

Evidence was also identified for the following interventions but these were not included in the cost-effectiveness analysis, due to limited reporting of outcomes in included studies:

  • corticosteroids

  • hyaluronidase

  • PGF2 gel

  • oral prostaglandins

  • intravenous prostaglandin

  • membrane sweeping

  • laminaria including dilapan

  • acupuncture

  • sexual intercourse

  • relaxin

  • amniotomy

  • estrogens.

The outcome VD24 includes 141 studies. The results showed that, for the interventions where VD24 was reported, the likelihood of VD24 increases in all interventions except for mifepristone and extra-amniotic PGE2. The lowest odds of not achieving VD24 were for:

  • intravenous oxytocin with amniotomy (odds ratio [OR] 0.05, 95% confidence intervals [CI] 0.01–0.14)

  • vaginal misoprostol ≥50 mcg (OR 0.09, 95% CI 0.06–0.24).

Caesarean section was an outcome in 307 included studies. When compared to placebo, titrated oral misoprostol had the lowest OR for caesarean section (OR 0.62, 95% CI 0.47–0.80). However, it is noted that there was uncertainty in the ranking of estimates.

Uterine hyperstimulation with fetal heart rate changes was an outcome included in 180 studies. Compared to placebo, the highest OR for this outcome were for misoprostol:

  • sustained release vaginal pessary (OR 5.58, 95% CI 1.58–14.57)

  • vaginal tablet ≥50 mcg (OR 4.40, 95% CI 2.22–7.94)

  • buccal/sublingual (OR 4.25, 95% CI 1.71–9.02)

  • oral tablet ≥50 mcg (OR 2.85, 95% CI 1.41–5.20).

NICU admissions was an outcome in 204 included studies. Compared to placebo, reduction in NICU admissions was reported for extra-amniotic PGE2 (OR 0.4, 95% CI 0.16–0.82). However, uncertainty in the effect estimates meant that no rank order was reported.

Instrumental delivery was an outcome in 299 included studies, these were interventions compared to placebo for reduction in instrumental delivery and reached statistical significance for vaginal PGE2 pessary (slow release) (OR 0.72, 95% CI 0.50–0.99) and Foley catheter (OR 0.68, 95% CI 0.50–0.91).

A significant reduction in occurrence of an Apgar score of ≤7, when interventions were compared with placebo, occurred for nitric oxide (OR 0.49, 95% CI 0.20–0.95) and buccal/sublingual misoprostol (OR 0.41, 95% CI 0.15–0.99).

There was insufficient reporting of serious neonatal and maternal morbidity and mortality for an NMA to be conducted for this outcome. For the outcome of maternal satisfaction there was insufficient reporting to indicate a preference.

Subgroup analysis was conducted for women with:

  • ruptured or unruptured membranes

  • with or without previous caesarean section

  • cervical assessment using Bishop score ≤6 or >6

  • differing gestational ages.

The results were consistent with the main results however for VD24 oxytocin with amniotomy was mostly conducted in women with a favourable Bishop score >6.

The cost-effectiveness analysis found that buccal/sublingual misoprostol and intravenous oxytocin with amniotomy have the lowest expected total cost. At a willingness-to-pay threshold of £20,000 the expected net benefit was highest for buccal/sublingual misoprostol (£14,669), then titrated (low dose) oral misoprostol solution (£14,658) and intravenous oxytocin with amniotomy (£14,652). It was the lowest for intracervical PGE2 (£10,617). Titrated low-dose oral misoprostol solution had the highest expected utility but with an ICER >£20,000. The ICER is the expected extra cost for each extra unit gain in utility compared with the prior non-dominated intervention.

Subgroup analyses were conducted for women with intact membranes and women with an unfavourable cervix. In women with intact membranes the intervention with highest net benefit was intravenous oxytocin with amniotomy. In women with an unfavourable cervix the most cost-effective interventions were buccal/sublingual misoprostol or titrated low-dose oral misoprostol solution.

Strengths and limitations


The study included a number of strengths:

  • There is a detailed protocol available.

  • Two reviewers independently assessed studies and extracted data from the studies.

  • Cochrane tools were used to assess risk of bias. Sensitivity analysis was conducted for high risk of bias studies.

  • To account for studies at high or unclear risk of bias for allocation, for caesarean section the model only included studies at low risk of bias.

  • Consistency was tested for between direct and indirect evidence.

  • It is very applicable to the guideline. Efforts were made in the cost-effectiveness analysis to make it as applicable as possible to UK NHS setting. The review also includes an interpretation of the implications for the guideline.


The authors note throughout the study that there is considerable uncertainty in the effect estimates and in the findings from the cost-effectiveness analysis. Additionally, the review is limited by the lack of reporting of outcomes of included studies. Only 20 interventions could be compared in the cost-effectiveness analysis for VD24. Half of the included studies also had a high or unclear risk of bias for allocation. Another limitation is that the search was until March 2014 and there may now be more evidence published that may affect the conclusions.

Impact on guideline

The study found that intravenous oxytocin with amniotomy, and misoprostol ≥50 mcg were the most likely interventions for VD24 for women undergoing induction of labour. However, there is considerable uncertainty in the effect estimates and in the findings from the cost-effectiveness analysis. The guideline does not recommend amniotomy alone or with oxytocin for the induction of labour. It only recommends misoprostol should be offered for the induction of labour in intrauterine fetal death. There is a need to review the recommendations on methods for the induction of labour.

This page was last updated: 09 January 2017