We set out to find out which progestogen treatment is most effective, safe, and has fewer side-effects for preventing miscarriage in women with threatened and with recurrent miscarriage, using evidence from randomised controlled trials. We looked at the number of women who went on to have a live birth, or miscarriage.
What is the issue?
Miscarriage is the most common cause of early pregnancy loss in the first 24 weeks and one of the most common complications in early pregnancy. An estimated 15% to 20% of pregnancies will end in a miscarriage, with 25% of women experiencing a miscarriage in their lifetime. Women can be at risk of a miscarriage if they experience early pregnancy bleeding, or if they have a history of previous miscarriages.
Why is this important?
Progesterone is an important pregnancy hormone that helps to maintain a pregnancy. A variety of different progesterone-like treatments (known as progestogens) have been used to treat women with early pregnancy bleeding. They are also used to prevent miscarriage in women with a history of previous miscarriages. There is uncertainty about the effectiveness, safety, and side-effects of the available progestogens for preventing miscarriage in these different groups of women. We wanted to find out which, if any, of the treatments is the most effective and safest. We collected and analysed all the relevant studies to answer this question.
What evidence did we find?
We searched for evidence in December 2020 and identified seven studies involving 5,682 women. All women were managed in hospitals. Women were diagnosed with early pregnancy bleeding (known as threatened miscarriage), or had a history of three or more previous miscarriages (known as recurrent miscarriage). Four different progestogen treatments were used: vaginal micronized progesterone, oral dydrogesterone, oral micronized progesterone and 17-α-hydroxyprogesterone injected into muscle. In six of the studies the treatments were compared to inactive placebo.
Three studies involved 4496 women with threatened miscarriage, some of whom had previously experienced a miscarriage. Overall, vaginal micronized progesterone (high-quality evidence) and oral dydrogesterone (moderate-quality evidence) made little difference to the number of women who went on to have a live birth when compared with placebo. We further studied the women who had experienced a previous miscarriage, were now presenting with a threatened miscarriage, and were given vaginal micronized progesterone or placebo. For women with one or more previous miscarriages, vaginal micronized progesterone increased the live birth rate compared to placebo (high-quality evidence). Those women who had no previous miscarriages, but were now presenting with early pregnancy bleeding showed no improvement in live birth rate (high-certainty evidence).
For women with recurrent miscarriage, we based our findings on one study involving 826 women. Overall, vaginal micronized progesterone made little difference to the live birth rate when compared with placebo. The evidence for dydrogesterone compared with placebo for women with recurrent miscarriage is of very low-certainty evidence, therefore the effects remain unclear. No data are available to assess the effectiveness of 17-α-hydroxyprogesterone or oral micronized progesterone for the outcome of live birth in women with recurrent miscarriage.
From the available data, there are likely no differences in adverse events associated with vaginal micronized progesterone. There was no difference in birth defects and side effects with vaginal micronized progesterone when compared with placebo. There was not enough information about safety and birth defects for us to analyse for all the other treatments.
What does this mean?
The overall available evidence suggests that progestogens probably make little or no difference to live birth rate for women with threatened or recurrent miscarriage. Vaginal micronized progesterone may increase the live birth rate for women who are experiencing early pregnancy bleeding and have a history of one or more previous miscarriages, with likely no difference in adverse events. There is still uncertainty over the effectiveness and safety of alternative progestogen treatments for threatened and recurrent miscarriage.
The overall available evidence suggests that progestogens probably make little or no difference to live birth rate for women with threatened or recurrent miscarriage. However, vaginal micronized progesterone may increase the live birth rate for women with a history of one or more previous miscarriages and early pregnancy bleeding, with likely no difference in adverse events. There is still uncertainty over the effectiveness and safety of alternative progestogen treatments for threatened and recurrent miscarriage.
Miscarriage, defined as the spontaneous loss of a pregnancy before 24 weeks’ gestation, is common with approximately 25% of women experiencing a miscarriage in their lifetime, and 15% to 20% of pregnancies ending in a miscarriage. Progesterone has an important role in maintaining a pregnancy, and supplementation with different progestogens in early pregnancy has been attempted to rescue a pregnancy in women with early pregnancy bleeding (threatened miscarriage), and to prevent miscarriages in asymptomatic women who have a history of three or more previous miscarriages (recurrent miscarriage).
To estimate the relative effectiveness and safety profiles for the different progestogen treatments for threatened and recurrent miscarriage, and provide rankings of the available treatments according to their effectiveness, safety, and side-effect profile.
We searched the following databases up to 15 December 2020: Cochrane Central Register of Controlled Trials, Ovid MEDLINE(R), ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform (ICTRP), and reference lists of retrieved studies.
We included all randomised controlled trials assessing the effectiveness or safety of progestogen treatment for the prevention of miscarriage. Cluster-randomised trials were eligible for inclusion. Randomised trials published only as abstracts were eligible if sufficient information could be retrieved. We excluded quasi- and non-randomised trials.
At least two review authors independently assessed the trials for inclusion and risk of bias, extracted data and checked them for accuracy. We performed pairwise meta-analyses and indirect comparisons, where possible, to determine the relative effects of all available treatments, but due to the limited number of included studies only direct or indirect comparisons were possible. We estimated the relative effects for the primary outcome of live birth and the secondary outcomes including miscarriage (< 24 weeks of gestation), preterm birth (< 37 weeks of gestation), stillbirth, ectopic pregnancy, congenital abnormalities, and adverse drug events. Relative effects for all outcomes are reported separately by the type of miscarriage (threatened and recurrent miscarriage). We used the GRADE approach to assess the certainty of evidence.
Our meta-analysis included seven randomised trials involving 5,682 women, and all provided data for meta-analysis. All trials were conducted in hospital settings. Across seven trials (14 treatment arms), the following treatments were used: three arms (21%) used vaginal micronized progesterone; three arms (21%) used dydrogesterone; one arm (7%) used oral micronized progesterone; one arm (7%) used 17-α-hydroxyprogesterone, and six arms (43%) used placebo.
Women with threatened miscarriage
Based on the relative effects from the pairwise meta-analysis, vaginal micronized progesterone (two trials, 4090 women, risk ratio (RR) 1.03, 95% confidence interval (CI) 1.00 to 1.07, high-certainty evidence), and dydrogesterone (one trial, 406 women, RR 0.98, 95% CI 0.89 to 1.07, moderate-certainty evidence) probably make little or no difference to the live birth rate when compared with placebo for women with threatened miscarriage. No data are available to assess the effectiveness of 17-α-hydroxyprogesterone or oral micronized progesterone for the outcome of live birth in women with threatened miscarriage.
The pre-specified subgroup analysis by number of previous miscarriages is only possible for vaginal micronized progesterone in women with threatened miscarriage. In women with no previous miscarriages and early pregnancy bleeding, there is probably little or no improvement in the live birth rate (RR 0.99, 95% CI 0.95 to 1.04, high-certainty evidence) when treated with vaginal micronized progesterone compared to placebo. However, for women with one or more previous miscarriages and early pregnancy bleeding, vaginal micronized progesterone increases the live birth rate compared to placebo (RR 1.08, 95% CI 1.02 to 1.15, high-certainty evidence).
Women with recurrent miscarriage
Based on the results from one trial (826 women) vaginal micronized progesterone (RR 1.04, 95% CI 0.95 to 1.15, high-certainty evidence) probably makes little or no difference to the live birth rate when compared with placebo for women with recurrent miscarriage. The evidence for dydrogesterone compared with placebo for women with recurrent miscarriage is of very low-certainty evidence, therefore the effects remain unclear. No data are available to assess the effectiveness of 17-α-hydroxyprogesterone or oral micronized progesterone for the outcome of live birth in women with recurrent miscarriage.
Additional outcomes
All progestogen treatments have a wide range of effects on the other pre-specified outcomes (miscarriage (< 24 weeks of gestation), preterm birth (< 37 weeks of gestation), stillbirth, ectopic pregnancy) in comparison to placebo for both threatened and recurrent miscarriage. Moderate- and low-certainty evidence with a wide range of effects suggests that there is probably no difference in congenital abnormalities and adverse drug events with vaginal micronized progesterone for threatened (congenital abnormalities RR 1.00, 95% CI 0.68 to 1.46, moderate-certainty evidence; adverse drug events RR 1.07 95% CI 0.81 to 1.39, moderate-certainty evidence) or recurrent miscarriage (congenital abnormalities 0.75, 95% CI 0.31 to 1.85, low-certainty evidence; adverse drug events RR 1.46, 95% CI 0.93 to 2.29, moderate-certainty evidence) compared with placebo. There are limited data and very low-certainty evidence on congenital abnormalities and adverse drug events for the other progestogens.