Progestogens for preventing miscarriage: a network meta-analysis.
Level 1 - systematic review of randomized trials
Systematic review and network meta-analysis of randomized controlled trials
PubMed 33872382 · doi:10.1002/14651858.CD013792.pub2
What was done
This Cochrane systematic review and network meta-analysis searched major databases up to December 2020 for randomized controlled trials evaluating progestogens (vaginal or oral micronized progesterone, dydrogesterone, 17-α-hydroxyprogesterone) against placebo or each other in women with threatened or recurrent miscarriage. Two review authors independently selected studies, extracted data, assessed risk of bias, and evaluated evidence certainty using GRADE. The primary outcome was live birth rate; secondary outcomes included miscarriage under 24 weeks, preterm birth, stillbirth, ectopic pregnancy, congenital abnormalities, and adverse drug events.
What was found
Seven randomized trials involving 5,682 women were included. For threatened miscarriage: - Vaginal micronized progesterone vs placebo showed little or no overall difference in live birth rate (2 trials, 4,090 women, RR 1.03, 95% CI 1.00 to 1.07; high-certainty evidence). - Dydrogesterone vs placebo showed little or no difference in live birth rate (1 trial, 406 women, RR 0.98, 95% CI 0.89 to 1.07; moderate-certainty evidence). - In subgroup analysis, vaginal micronized progesterone did not improve live birth in women with early bleeding and no previous miscarriages (RR 0.99, 95% CI 0.95 to 1.04; high certainty), but increased live birth in women with early bleeding and one or more previous miscarriages (RR 1.08, 95% CI 1.02 to 1.15; high certainty). For recurrent miscarriage: - Vaginal micronized progesterone vs placebo showed little or no difference in live birth rate (1 trial, 826 women, RR 1.04, 95% CI 0.95 to 1.15; high-certainty evidence). - Evidence for dydrogesterone was very low certainty. No live birth data existed for oral micronized progesterone or 17-α-hydroxyprogesterone. Vaginal micronized progesterone showed no statistically significant increase in congenital abnormalities (threatened RR 1.00, 95% CI 0.68 to 1.46; recurrent RR 0.75, 95% CI 0.31 to 1.85) or adverse drug events (threatened RR 1.07, 95% CI 0.81 to 1.39; recurrent RR 1.46, 95% CI 0.93 to 2.29) compared to placebo.
Why it matters
Progestogens provide no general benefit across all women with threatened or recurrent miscarriage, but vaginal micronized progesterone specifically increases live birth rates in women experiencing early pregnancy bleeding who have had at least one prior miscarriage.
Limits
The analysis is restricted to seven trials, preventing full network treatment rankings. Data for recurrent miscarriage and non-vaginal progestogen formulations were sparse or entirely absent for live birth outcomes.
Cited by
- context Low progesterone levels are a primary cause of miscarriage.