Preterm birth is a complex phenotype with varied mechanisms of parturition based on risk factor. Multiple gestations account for a significant and disproportionate percentage of the neonatal morbidity and mortality rates that are associated with prematurity. As clinicians, we encounter more women with twin pregnancies than ever before and are faced with making decisions about which, if any, therapies to offer this select group of women who are at increased risk of preterm birth.

For years, the mechanisms of preterm birth have eluded both the researcher and the clinician. Widely accepted risk factors for preterm birth include a history of preterm birth, multiple gestations, and a short cervix. Targeted counseling of women can be onerous and quite complicated because we often encounter those women with multiple risk factors, some of which, in combination, may not have been addressed adequately by the published studies.

Therapeutic modalities such as cervical cerclage and home uterine activity monitoring have fallen short in reducing the rate of preterm birth in twin gestations. In fact, cervical cerclage has been shown to increase adverse outcomes.

One of the most promising therapies in recent years is the introduction of weekly 17 alpha hydroxyprogesterone caproate (OHPC) 250 mg injections. Since the publication of the National Institute of Child Health and Human Development-Maternal Fetal Medicine Unit trial in 2003 that found a 30% reduction in recurrent preterm birth in women with a singleton gestation and a history of spontaneous preterm who were treated with weekly 17 OHPC injections, investigators have attempted to use this therapy for both multiple gestations and nulliparous women with a short cervix without success. These findings illustrate that the mechanism of preterm birth is complex and multifaceted and that the application of a proven therapy for one at-risk group may not be effective for all.

Sonographic measurement of the cervix has also been used to identify women with twin gestations who are at increased risk for preterm birth. Cervical shortening in multiple gestations has been shown to occur earlier than in singleton gestations, which has led investigators to question whether twin gestations with early second-trimester cervical shortening have an inflammation-mediated component of preterm parturition that may be responsive to 17 OHPC injections. In a subanalysis of the STTARS (Seventeen Alpha-Hydroxyprogesterone Caproate in Twins and Triplets) trial, weekly 17 OHPC injections did not reduce the rate of preterm birth at <35 weeks’ gestation in those women who were identified with a short cervix before the initiation of treatment. Similar findings were reported by da Fonseca et al in a small group of women with twins and a short cervix who received vaginal progesterone. Some of the proposed theories about the reason that progesterone supplementation has been ineffective in this subset of women include that the mechanism of preterm birth in multiple gestations primarily may be related to uterine distention and that dosing may need to be adjusted because of the expanded plasma volume in twins.

In this issue of the Journal, Senat et al report a multicenter randomized controlled trial of asymptomatic women with a twin gestation who were identified to have a short cervix between 24 ⁰ -31 ⁶ weeks’ gestation and who were assigned randomly to receive either 500 mg of intramuscular 17 OHPC twice weekly from diagnosis until 36 weeks’ gestation vs no treatment. The investigators found no prolongation of pregnancy using this higher progesterone dosing regimen. An unanticipated finding was the association between progesterone supplementation and a doubling of the preterm delivery rate at <32 weeks’ gestation (29% vs 12%). In a recent pharmacodynamic study that explored the relationship between 17 OHPC concentrations and gestational age at delivery in twin gestations, women with higher plasma concentrations of 17 OHPC delivered at an earlier gestational age, despite similar corticotrophin-releasing hormone levels. These findings should give us pause for thought. In singleton gestations, genetic variations have been shown to contribute to the clinical response that has been seen with 17 OHPC treatment. The genetic contribution to clinical response has yet to be explored in multiple gestations. Further investigation is needed to better elucidate the relationship among genetic susceptibility, progesterone dosing, and plasma concentrations in these unique populations independently. Only with an understanding of the complex mechanisms that surround the variability in clinical response and the identification of biologic targets will the identification of those women who will benefit from 17 OHPC and those who potentially could be at more risk related to the treatment be possible.

Until more targeted studies are performed, this leaves the clinician to ponder 3 questions: (1) Should I recommend 17 OHPC treatment to women with a history of a twin spontaneous preterm birth who are now carrying a singleton pregnancy? These women are likely have an increased risk of preterm birth, but treatment with 17 OHPC has not been studied in this population. Given the most likely contributor to her risk was the multiple gestation, no treatment is warranted. (2) Should I recommend 17 OHPC in women carrying a twin gestation and a history of a singleton spontaneous preterm birth? Published studies have failed to show a benefit with treatment of these women, despite their qualifying singleton delivery. Given the uncertainty of benefit and recent findings that raise a concern for potential worse outcomes, women should be counseled thoroughly regarding the uncertainty of benefit and the potential for harm if 17 OHPC is prescribed. (3) Should I recommend 17 OHPC treatment to women with a twin gestation without a history of preterm birth and a short cervix? Without an effective treatment to offer these women, one would question the utility of cervical screening in this population because no intervention has been proved to prolong pregnancy or alter outcome.

Clinicians are often compelled to offer some therapy to a patient because the act of doing something is perceived to be better than doing nothing. But in the case of 17 OHPC treatment in twin gestations, too many questions remain unanswered, and no clear benefit has been demonstrated.