472: Progestogen in Twin Pregnancies: an Individual Participant Data Meta-Analysis of Randomized Trials

Preterm birth is the principal factor contributing to adverse outcomes in multiple pregnancies. RCTs of progestogen treatment in twin pregnancies did not show benefits in preventing preterm births, but did not have adequate power to evaluate effects on perinatal outcome or effects among subgroups. We conducted an individual participant data meta-analysis of nine randomized, double-blind, placebo-controlled trials of progestogen treatment for prevention of preterm birth in twin pregnancy. The primary outcome was a composite of perinatal mortality and severe neonatal morbidity. Secondary outcomes included intrauterine death, a composite of intrauterine death or delivery before 37, 35, 32 and 28 weeks, and time-to-delivery or death. Prespecified subgroup analyses were performed for type of progestogen, monochorionicity, cervical length < 25 mm, a history of spontaneous preterm birth < 37 weeks, and ethnicity. We analyzed data from 3,304 women pregnant with 6,608 children. No statistically significant effect was seen from progestogens as compared to placebo for adverse perinatal outcome (RR 1.05; 95% CI 0.91-1.2) or any of the secondary outcomes. Separate comparisons of 17-alpha hydroxy progesterone caproate (17P) or vaginal progesterone to placebo did not show significant differences either. In women with a cervical length < 25 mm, the proportion of women with an adverse perinatal outcome was not significantly reduced: 15 (41%) versus 11 (61%) for progestogens versus placebo (RR 0.67; 95% CI 0.40-1.1). However, the numbers in this subgroup were too small to draw definitive conclusions. In women with a twin gestation, treatment with progestogens neither prolongs pregnancy nor improves neonatal outcome. The value of screening for short cervix in twin gestations and subsequent progestogen treatment should be explored in prospective studies.