Caroline Helwick

February 09, 2016

ATLANTA — In women with high-risk pregnancies, including those with a short cervix, vaginal progesterone does not protect against preterm birth, according to disconcerting results from the OPPTIMUM study.

"We saw no significant effect of vaginal progesterone on obstetrical, neonatal, or childhood outcomes," reported Jane Norman, MD, from the University of Edinburgh in the United Kingdom.

This was true even for women with a cervix of 25 mm or less, and "we found no evidence of benefit in any identifiable subgroups," she said here at the Society for Maternal-Fetal Medicine 2016 Pregnancy Meeting.

However, "we did see infrequent but significantly increased events of renal, gastrointestinal, and respiratory harm that deserve consideration in other studies," said Dr Norman.

"There is a remaining unmet need for a safe and effective agent to prevent preterm birth," she said.

A Cochrane Meta-analysis

The use of progesterone — both the vaginal form and 17-hydroxyprogesterone, the intramuscular form — was assessed in a Cochrane meta-analysis of women who had previously delivered a preterm child (Cochrane Database Syst Rev. 2013;7:CD004947). Progesterone was shown to prevent preterm births, reduce perinatal mortality, reduce the incidence of birthweight below 2500 g, and reduce neonatal deaths. In women with a short cervix, progesterone was shown to reduce preterm births, but did not affect the other parameters.

"Progesterone administration has been shown to reduce the risk of preterm birth and neonatal morbidity in women at high risk, but there has been little information about the long-term risk this may have on children," Dr Norman explained.

"We felt we needed to look at the long-term effects. We cannot assume that things with short-term benefit will not have long-term harm," she added.

The OPPTIMUM Study

The double-blind, placebo-controlled, randomized OPPTIMUM study — the largest study to date of this intervention — involved 1228 women with singleton pregnancies at risk for preterm birth because of a positive fetal fibronectin test, a history of spontaneous preterm birth at 34 weeks of gestation or earlier, or a cervical length 25 mm or less.

Women were randomized to use vaginal progesterone 200 mg daily from 22 to 24 weeks through to 34 weeks of gestation or placebo.

Median age of the women was 31 years, and median body mass index was 27 kg/m². Approximately 75% of the women had a history of preterm birth and about 35% had a cervical length of 25 mm or less.

The primary outcomes were preterm birth or fetal death before 34 weeks, neonatal death or major morbidity (brain injury or bronchopulmonary dysplasia), and Bayley III cognitive scores in 2-year-old children.

Data were available for 97.4% of the obstetric outcomes, 95.8% of the neonatal outcomes, and 70.8% of the childhood outcomes.

Progesterone had no significant effect on either the obstetric or childhood outcomes. It also had no significant effect on any components of the primary outcome, including fetal death or live-born delivery before 34 weeks.

Table 1. Primary Outcomes

Outcome Progesterone Placebo Odds Ratio P Value
Fetal death or delivery before 34 weeks 16.0% 18.1% 0.86 .67
Neonatal morbidity or death 6.6% 10.2% 0.62 .07
Cognitive composite score in 2-year-old children 97.3 97.7 .68

 

"We were really surprised that we didn't show that progesterone prevented preterm birth, and we became concerned that perhaps our cutoff of 34 weeks was the wrong choice," said Dr Norman. But this concern was not borne out in a post hoc survival analysis of the trajectory to delivery for all women in the study.

A per protocol analysis of women compliant with therapy, after adjustment for multiple comparisons, showed that, again, progesterone had no effect on the primary outcome.

When the components of the neonatal primary outcome were examined, a small but significant reduction was seen for neonatal death (1.0% vs 0.2%; P = .01) and brain injury (5.9% vs 3.1%; P = .01). However, these are secondary outcomes and should be viewed with caution, Dr Norman pointed out.

In a subgroup analysis of women with a short cervix, progesterone had no significant benefit on obstetric, neonatal, or childhood outcomes. This was true for women with a cervix of 25 mm or less and for those with a cervix of 15 mm or less.

The only differences observed in the study were in the individual components of childhood disability.

Table 2. Secondary Outcomes of Childhood Disability

Outcome Progesterone, % Placebo, % Odds Ratio P Value
Cognitive function 4.1 4.0 1.03 .92
Respiratory event 1.7 0.7 3.03 .0011
Gastrointestinal event 2.2 0.9 2.67 .004
Renal event 0.7 0.2 3.65 <.001

 

"We did find differences in respiratory, renal, and gastrointestinal disabilities, which were significantly increased in the progesterone group," Dr Norman reported. However, she added, these are secondary outcomes.

"There may be subgroups that can benefit, but this study did not identify them," she pointed out.

After the presentation, attendees took to the microphone to call the study "fascinating", "amazing", and "outstanding".

Provocative Findings

One member of the audience commented on the 48% increase in moderate to severe neurodevelopment impairment at 2 years (95% confidence interval, 0.98 - 2.33; P = .87). "It was not significant, but the lower limit of the confidence interval was 0.98. Do you think progesterone is safe?" he asked.

Dr Norman explained that "for our primary safety outcome, there was no difference. The safety signals are in the secondary outcomes and have to be taken with caution." However, "I wouldn't advise my daughter to take progesterone," she acknowledged.

Another audience member said he "retains some optimism of the possibility that progesterone has a positive effect on gestational age for women with a very short cervix," and noted that some studies have shown a benefit.

"We were not powered to look at subgroups, so anything we say about them has to be said with caution," Dr Norman responded. However, "we showed absolutely no evidence it is more effective in the short-cervix group."

In a previous presentation at the meeting, it was reported that "just 4% of spontaneous preterm births have a cervical length that is short," said Michal Elovitz, MD, director of maternal child health research at the University of Pennsylvania in Philadelphia.

"In the United States, most of us do universal cervical length. Do you argue that we shouldn't do this anymore, based on the results of your study?" she asked.

"If we assume our study is correct, there is no benefit," Dr Norman said. "However, there are other things for short cervix, such as cerclage and cervical pessary. Maybe it's worth screening for short cervix, but I'm not sure progesterone is the answer."

 
This certainly gives pause around vaginal progesterone.
 

"This certainly gives pause around vaginal progesterone," said Mary D'Alton, MD, the William C. Rappleye Professor of Obstetrics and Gynecology at the Columbia University Medical Center in New York City. But she said she would not advise a change in practice before seeing the final published results.

Dr D'Alton noted that the results should not be extrapolated to 17-hydroxyprogesterone, which was not part of this analysis.

The US Food and Drug Administration has not approved vaginal progesterone for the prevention of preterm birth in women with a short cervix.

Besins Healthcare donated the active drug but was not involved in the study design or analysis. Dr Norman, Dr Elovitz, and Dr D'Alton have disclosed no relevant financial relationships.

Society for Maternal-Fetal Medicine (SMFM) 2016 Pregnancy Meeting: Abstract LBA2. Presented February 5, 2016.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
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.
Post as:

processing....