Too Soon to Adopt Progesterone for the Prevention of Preterm Delivery

Peter S. Bernstein, MD, MPH


May 16, 2003


Two recent studies have renewed interest in the usefulness of progesterone supplementation during pregnancy for the prevention of preterm delivery.[1,2] The studies are based on the theory that parturition is initiated by the withdrawal of progesterone.

The first of these studies to be published was authored by da Fonseca and colleagues.[1] It included 142 singleton pregnancies at high risk for preterm delivery. In this randomized, placebo-controlled, double-blind trial, subjects were given 100 mg progesterone vaginal suppositories daily or an identical-looking placebo. Patients were instructed to insert the suppository every night during the period from 24 to 34 weeks of gestation. Participants were deemed at high risk for preterm delivery on the basis of at least 1 of the following factors: history of at least 1 spontaneous preterm delivery, presence of a prophylactic cervical cerclage, or the presence of a uterine malformation. All patients underwent uterine contraction monitoring once a week in the morning for 1 hour during the period from 24 to 34 weeks of gestation.

The study was conducted at a single tertiary medical center in Brazil from February 1996 through March 2001. Of the 157 asymptomatic high-risk patients enrolled in the protocol, 142 were able to complete it. The authors noted a significant reduction in the number of preterm births (defined as delivery before 37 weeks of gestation) in the patients treated with progesterone suppositories. In the progesterone group, the incidence of preterm delivery was 13.8% (10/72) compared with 28.5% (20/70) in the control group (P = .03). Delivery at less than 34 weeks of gestation was also significantly reduced, from 18.6% to 2.8%, in the progesterone group. Finally, the authors noted that the mean number of contractions for each gestational week studied was significantly lower in the progesterone group compared with the placebo group. Although this study was able to show a significant decrease in the rate of preterm delivery (an intermediate variable), it was not large enough to address the more important question of whether progesterone supplementation can reduce perinatal morbidity and mortality.

At first glance, this study appears impressive, but I have concerns about its design that make me reluctant to adopt its conclusions. Most of my concerns relate to the makeup of the population studied. At a tertiary medical center, it took 4 years to recruit 157 patients for this study. Can this really be a reliable sample of patients from this center at risk for preterm delivery -- especially given the very high rate of preterm delivery in the placebo group? The composition of the study population suggests that the patients included were at extremely high risk for preterm delivery, and thus the results of this study may not be applicable to a patient with a single risk factor for preterm delivery, for example a single prior preterm delivery at 35 weeks of gestation. In short, not nearly enough information is provided to describe the population enrolled or about how they were selected from the larger population of women seeking care at this medical center. This significantly weakens the extrapolation of the conclusions of this study to other populations of pregnant women.

Additionally, even though the majority of patients studied had a prior preterm birth as one of the risk factors that made them eligible for inclusion, the risk factors of incompetent cervix and uterine malformations suggest mechanisms that lead to preterm birth of a very different sort than a prior preterm birth. I am not sure that I would have included them in the design of the study from the start. How are we to know which patients are really the best candidates for this therapy?

The second study,[2] conducted by the National Institute of Child Health and Human Development Maternal Fetal Medicine Units (MFMU) Network, was presented at this year's Society of Maternal-Fetal Medicine's annual meeting. This study was performed at the 19 institutions across the United States that comprise the MFMU Network and entailed weekly injections of 17 alpha hydroxyprogesterone (17P). Patients in this study all had a history of preterm delivery. None had cerclages. Participants were randomized to 250 mg of 17P or placebo at 15 to 20 weeks of gestation, which they continued until 37 weeks of gestation or delivery, whichever came first.

In this study, 1039 women were eligible for participation, but only 463 agreed to participate and were randomized. The study was stopped after 351 participants had delivered, when a planned interim analysis revealed a significant positive effect of the study medication. In the placebo group, nearly 55% of the women delivered at less than 37 weeks of gestation compared with 36% in the group that received weekly 17P. Even more importantly, the rate of neonatal mortality fell from 5.9% in the placebo group to 2.6% in the study group, although this finding did not reach statistical significance (P = .08). The rates of neonates with birth weight less than 2500 g and less than 1500 g were also significantly reduced in the 17P group. The authors of the study hypothesized that only 5 to 6 women with prior preterm births would need to be treated to prevent 1 recurrent preterm birth.

But exactly which women should be treated? Again, more information is needed about the population of women enrolled in this study. Typically, the rate of preterm delivery after a single prior preterm birth is quoted as approximately 18% to 20%. In the placebo group in this study, the rate of preterm birth was nearly 55%. This is clearly a unique population of women. The average gestational age at delivery among these women at the time of their prior preterm delivery was 30.7 weeks. Approximately 30% of the participants had had more than 1 prior premature delivery. I am concerned that this therapy may not be appropriate for a more typical patient who has a prior preterm birth.

I am also concerned about the generalizability of these findings to other populations of women, given that less than 50% of the women eligible for the study actually participated. The study population may not be representative of women at risk for preterm delivery. Nevertheless, I eagerly await the publication of this study in order to learn more about how it was conducted and who exactly participated.

These studies will undoubtedly encourage the obstetric community to take another look at usefulness of progesterone in preventing preterm birth. Still, it is too soon to adopt this therapy. The problem of premature delivery has multiple causes. More work will need to be done to elucidate which women will benefit from this promising intervention.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.