The use of bed rest has remained controversial. It is commonly prescribed to prevent preterm birth in women with a twin gestation and a short cervix, despite the fact that we have no published data demonstrating its effectiveness in prolonging pregnancy. Personally, I believe that bed rest can reduce the frequency of uterine contractions and help protect the cervix from the weight of the pregnancy.
Interest in the treatment of twin pregnancies with progesterone had waned after 2007, when a National Institutes of Health multicenter, randomized trial of injectable progesterone reported that 17-alpha-hydroxyprogesterone caproate did not reduce the rate of twin preterm birth (N. Engl. J. Med. 2007;357:454-61).
However, progesterone recently has reemerged as a treatment for women who are at higher risk of preterm birth, based on the publication this year of Dr. Roberto Romero’s review and meta-analysis of vaginal progesterone for women with a sonographic short cervix (Am. J. Obstet. Gynecol. 2012;206:124.e1-19). Indeed, this review provides the first evidence of a beneficial effect in women with a short cervix who are carrying a twin gestation. The meta-analysis covered five prospective, randomized trials of vaginal progesterone with a total of 775 women who had a sonographically confirmed short cervix (25 mm or less) in the midtrimester. The vast majority of women in these studies carried singleton pregnancies, but two of the studies enrolled twin as well as singleton gestations, and one of the five studies – albeit a small one – focused solely on twin pregnancies.
Overall, treatment with vaginal progesterone was associated with a highly significant 42% reduction in the rate of preterm birth at less than 33 weeks. Among twin gestations specifically, the reduction in preterm birth was 30% – a meaningful trend, but not statistically significant. When it came to neonatal morbidity and mortality, however, the effect of vaginal progesterone among twin gestations was far more striking: The group that received vaginal progesterone had a 48% reduction in the risk of composite neonatal morbidity and mortality.
Admittedly, a primary randomized, controlled trial in twin gestations is still needed. In the meantime, however, given the tremendous risk faced by women with a twin gestation and a short cervix, and the lack of any other proven treatment, I believe that vaginal progesterone (Prometrium, in either a 200-mg suppository or 90-mg gel) is an appropriate treatment for this condition.
In our practice, we routinely perform transvaginal cervical-length measurements with our twin pregnancies at the time of their anatomical survey at 18-20 weeks, and then every 2-4 weeks (depending on how short the cervix is) up to 26-28 weeks. If a short cervix is diagnosed, we restrict activity and start vaginal progesterone.
Surveillance, Delivery
Careful surveillance during the late gestational period is critical, as twins – particularly monochorionic twins – are at increased risk of growth restriction or growth discordancy, and have an increased risk of developing abnormalities in amniotic fluid volume. Compared with singletons, twins also are at increased risk of stillbirth in the third trimester; this risk, again, appears to be higher for monochorionic gestations.
We perform routine ultrasound evaluations every 3 weeks for our monochorionic twins and every 4 weeks for our dichorionic twins, in the absence of any abnormalities. If abnormalities in growth are suspected with standard ultrasound evaluation, we add umbilical artery Doppler studies to further assess well-being. We also routinely institute fetal nonstress testing at 32 weeks’ gestation for monochorionic twins and at 34 weeks for our dichorionic twins. Additional strategies are employed as necessary.
The overall risk of stillbirth for twin gestations is 0.2%-0.4% per week after 32 weeks’ gestation, and rises further beyond 38 weeks – a risk that makes surveillance critical.
Some investigators, however, have recently reported higher-than-expected stillbirth rates for "apparently uncomplicated" monochorionic twin gestations. This risk has ranged from 1% to 4% at 32 and 38 weeks’ gestation in various reports.
These studies were debated as part of a workshop held in 2011 by the National Institutes of Health and the Society for Maternal-Fetal Medicine on the "Timing of Indicated Late Preterm and Early Term Births." The findings of the higher stillbirth rates remain controversial, but on the basis of these concerns, it was recommended that monochorionic twins – even "uncomplicated" cases – should be offered elective delivery at 34-37 weeks’ gestation, with decisions made after careful discussion and informed consent.
Uncomplicated dichorionic twins, on the other hand, appear to have optimal outcomes when delivered at 38 weeks’ gestation (Obstet. Gynecol. 2011;118:323-3; Semin. Perinatol. 2011;35:277-85).
Recommended surveillance and delivery of monoamniotic twins – a rare but serious type of monochorionic twin gestation – has evolved recently in favor of intensive inpatient monitoring.