Rationale and impact
These sections briefly explain why the committee made the recommendations and how they might affect practice.
There was good evidence that vaginal progesterone reduced the risk of preterm birth before 34 weeks in women with a previous history of preterm birth, and in women with a short cervix (25 mm or less). The committee were aware that these groups overlapped, as some women with a previous history of preterm birth will also have a short cervix. Therefore, they adopted the recommendation from the previous guideline to offer vaginal progesterone to women with a previous history of preterm birth and a short cervix. The committee concluded that, as in the previous guideline, progesterone should be offered as an equal option with cervical cerclage (for which no new evidence review had been done), as there is no evidence to determine which of these options is more effective.
As the treatment options are very different (regular use of vaginal progesterone pessaries throughout pregnancy, compared with a single operative procedure), the committee highlighted that the choice of treatment should be made after discussion of the risks and benefits of the 2 treatments.
The committee were aware that there is uncertainty about which risk factors should be used to identify women at risk of preterm birth (cervical length measurements, previous history of preterm birth, previous cervical surgery). There is also variation in practice across the country about which women are offered cervical length scanning. Cervical scanning is currently offered when there is clinical concern about the risk of preterm birth, rather than as a routine part of antenatal care. Also, vaginal progesterone may be effective at reducing preterm birth for women with some risk factors, but not others.
Identifying specific groups of women who would benefit from treatment with progesterone was difficult because of the overlap in risk factors for an individual woman: some women with a previous history of preterm birth also have a cervical length of 25 mm or less, and some women with a cervical length of 25 mm or less also have a previous history of preterm birth. Therefore, it was hard to determine which of these 2 factors could identify women at high risk of preterm birth who would definitely benefit from treatment with vaginal progesterone. Consequently, the committee agreed that treatment with progesterone should be considered for women with either of these risk factors (cervical length of 25 mm or less, or a previous history of preterm birth). Because of the uncertainty over the benefits of progesterone in women who have risk factors for a preterm birth but do not have a cervical length of 25 mm or less, and women who have a cervical length of 25 mm or less but do not have a history of preterm birth, the committee made recommendations for research on this topic.
The timing of progesterone administration varied between the studies. However, most trials started treatment between 16+0 and 24+0 weeks. This was in keeping with the experience of the committee members, therefore they made a recommendation to start treatment at any suitable time during that range of gestational age. There was no evidence on when progesterone should be stopped, but the committee's experience was that it should be continued until at least 34 weeks. As there was uncertainty about these timings, the committee also made a recommendation for research on the optimal timing of treatment.
The recommendation on ensuring a plan is in place for removal of the suture when prophylactic cervical cerclage is used was made in response to an NHS England safety report, which highlighted some instances when removal did not happen.
Vaginal progesterone is a relatively inexpensive and commonly used treatment for women at risk of preterm birth, so the recommendations are unlikely to significantly alter practice. As vaginal progesterone should now be considered for women with a history of preterm birth (with an unknown cervical length or a cervical length greater than 25 mm on scan), this might increase the use of progesterone, but the benefits of reduced numbers of preterm births are likely to lead to cost savings overall.
The recommendation on planning for removal of the suture when prophylactic cervical cerclage is used is not expected to affect practice.
The recommendation on ensuring a plan is in place for removal of the suture when emergency cervical cerclage is used was made in response to an NHS England safety report, which highlighted some instances when removal did not happen.
The recommendation is not expected to affect practice.
There was some evidence that repeat doses of maternal corticosteroids reduce birthweight, but the absolute reductions in birthweight were small, with a mean difference in birthweight of 114 g between women receiving repeat courses and women receiving a single course. Subgroup analyses showed reductions were seen when corticosteroids were administered at lower gestational ages (below 30 weeks), when administered at intervals of less than 7 days, and when higher doses of more than 24 mg (total dose of repeat course) were administered. There was also a significant trend for reducing birthweight as the number of repeat courses increased. There was no evidence of benefit of maternal corticosteroids on chronic lung disease, but the committee were aware of a benefit seen with the need for respiratory support in neonates, although this outcome had not been prioritised for inclusion in the review. There was good evidence that repeat courses of maternal corticosteroids had no effect or beneficial effects on perinatal mortality, neonatal admission, intraventricular haemorrhage, growth at 2 years and neurodevelopmental delay. The committee agreed that a single repeat course may be beneficial in certain circumstances, when the previous course had been given more than 7 days previously and preterm birth was imminent, but that with multiple repeat courses the effects on birthweight may outweigh the benefits. However, the committee agreed that corticosteroids administered for other reasons during pregnancy would not count towards this total of 2 courses, and so clarified in their recommendation that only courses administered for preterm labour should be counted.
The committee were concerned with the lack of evidence for longer-term neurodevelopmental and growth outcomes beyond 2 years and lack of evidence on the optimal dose and interval for the repeat corticosteroids and so made a recommendation for research.
The recommendations provide guidance on when a single repeat course of maternal corticosteroids may be used, and so may reduce variation in practice. This may increase the number of women who receive a single repeat course, and may reduce the number of multiple (more than 2) courses of maternal corticosteroids given. The cost impact is therefore likely to be minimal considering the low cost of a course of maternal corticosteroids and the relatively small population of women for whom this will be considered.