This protocol outlines a randomized trial of 630 women evaluating the use of micronized vaginal progesterone or pessary versus control (placebo) to prevent early preterm birth in women carrying twins and with a cervical length of less than 30 millimeters.

This protocol outlines a randomized trial of 630 women evaluating the use of micronized vaginal progesterone or pessary versus control (placebo) to prevent early preterm birth in women carrying twins and with a cervical length of less than 30 millimeters.

Multiple gestation increases the risk of preterm delivery. Babies born preterm have increased rates of neonatal mortality and long-term neurodevelopmental morbidities. Short cervical length is known to be an important risk factor for spontaneous preterm birth and to occur more frequently in women with a twin gestation. Although there is no evidence that progesterone reduces the risk of preterm birth in multifetal gestation, there is evidence that progesterone reduces the risk of prematurity in singleton gestations complicated with a short cervix. The Arabin pessary has also been shown to reduce the risk of preterm birth among singletons with a short cervix, and in a secondary subgroup analysis of a recent study of the use of pessary in multiple gestations, women with a cervical length < 25th percentile had a significantly reduced risk of the primary composite neonatal adverse outcome. Secondary analysis of studies of vaginal progesterone in multiple gestation with a short cervix also suggest a possible beneficial effect on preterm delivery.

Inclusion Criteria:

1. Twin gestation with cardiac activity in both fetuses. Higher order multifetal gestations reduced to twins, either spontaneously or therapeutically, are not eligible unless the reduction occurred by 13 weeks 6 days project gestational age.
2. Gestational age at randomization between 16 weeks 0 days and 23 weeks 6 days based on clinical information and evaluation of the earliest ultrasound.
3. Cervical length on transvaginal examination of less than 30 mm by a study certified sonographer.

Exclusion Criteria:

1. Cervical dilation (internal os) 3 cm or greater on digital examination or evidence of prolapsed membranes beyond the external cervical os either at the time of the qualifying cervical ultrasound examination or at a cervical exam immediately before randomization. There is no lower threshold of cervical length measurement threshold on ultrasound that is an exclusion criterion.
2. Monoamniotic gestation, due to increased risk of adverse pregnancy outcome
3. Twin-twin transfusion syndrome, due to increased risk of adverse pregnancy outcome
4. Evidence of severe IUGR (intrauterine growth restriction) (<5th percentile for gestational age) in either fetus
5. Fetal anomaly in either twin or imminent fetal demise. This includes lethal anomalies, or anomalies that may lead to early delivery or increased risk of neonatal death e.g., gastroschisis, spina bifida, serious karyotypic abnormalities). An ultrasound examination from 14 weeks 0 days to 23 weeks 6 days by project EDC (estimated date of conception) must be performed prior to randomization to evaluate the fetuses for anomalies.
6. Placenta previa, because of risk of bleeding and high potential for indicated preterm birth
7. Active vaginal bleeding greater than spotting at the time of randomization, because of potential exacerbation due to pessary placement.
8. Symptomatic, untreated vaginal or cervical infection, also because of potential exacerbation due to pessary placement. Patients may be treated and if subsequently asymptomatic, randomized.
9. Active, unhealed herpetic lesion on labia minora, vagina, or cervix due to the potential for significant patient discomfort or increasing genital tract viral spread. Once lesion(s) heal and the patient is asymptomatic, she may be randomized. History of herpes is not an exclusion.
10. Rupture of membranes due to likelihood of pregnancy loss and preterm delivery as well as the risk of ascending infection which could be increased with pessary placement
11. More than six contractions per hour reported or documented prior to randomization. It is not necessary to place the patient on a tocodynamometer
12. Known major Mullerian anomaly of the uterus (specifically bicornuate, unicornuate, or uterine septum not resected) due to increased risk of preterm delivery which is unlikely to be affected by progesterone
13. Any fetal/maternal condition which would require invasive in-utero assessment or treatment, for example significant red cell antigen sensitization or neonatal alloimmune thrombocytopenia
14. Major maternal medical illness associated with increased risk for adverse pregnancy outcome or indicated preterm birth (treated hypertension requiring more than one agent, pre-gestational treatment for diabetes prior to pregnancy, chronic renal insufficiency failure defined by creatinine >1.4 mg/dL, carcinoma of the breast, conditions treated with chronic oral glucocorticoid therapy. Specifically, patients with seizure disorders, HIV, and other medical conditions not specifically associated with an increased risk of indicated preterm birth are not excluded. Prior cervical cone/LOOP/LEEP is not an exclusion criterion.
15. Planned cerclage or cerclage already in place since it would preclude placement of a pessary
16. Planned indicated delivery prior to 35 weeks
17. Planned or actual progesterone treatment of any type or form after 15 weeks 6 days during the current pregnancy
18. Allergy to progesterone, silicone, or excipients in the study drug, including peanuts or peanut oil in the study drug or placebo
19. Known, suspected or history of breast cancer because breast cancer is a contraindication to the active study medication.
20. Known liver dysfunction or disease because liver disease is a contraindication to the active study medication.
21. Participation in another interventional study that influences gestational age at delivery or neonatal morbidity or mortality
22. Participation in this trial in a previous pregnancy. Patients who were screened in a previous pregnancy, but not randomized, do not have to be excluded.
23. Prenatal care or delivery planned elsewhere unless the study visits can be made as scheduled and complete outcome information can be obtained

The dataset will be shared per NIH policy after the completion and publication of the main analyses.

Participants and care providers will be blinded to active study drug vs. placebo.