Metformin May Prevent Pregnancy Complications in PCOS

Miriam E Tucker

March 22, 2018

CHICAGO, Illinois — Metformin treatment may improve pregnancy outcomes for women with polycystic ovary syndrome (PCOS), but it doesn't prevent gestational diabetes and its long-term effects on the offspring are unclear, new research indicates.

Tone Shetelig Løvvik, MD, of the Norwegian University of Science and Technology, Trondheim, presented findings from the randomized controlled PregMet 2 trial on March 20 here at ENDO 2018: The Endocrine Society Annual Meeting.

Women with PCOS are at increased risk of pregnancy complications, and metformin has increasingly been used off-label during pregnancy to prevent or treat those complications. Previous studies investigating this have been underpowered, Løvvik explained.

In PregMet 2, women with PCOS randomized to metformin beginning in the first trimester — and who actually took the drug throughout pregnancy — had lower risks for late miscarriage and preterm birth compared with women given placebo. They also gained less weight during pregnancy.

However, metformin did not prevent the development of gestational diabetes. Also, newborns of the women taking metformin had larger head circumferences, albeit still in the normal range.

Previous research by the same group has shown that in a follow-up of two cohorts, children of women with PCOS who took metformin during pregnancy had a higher average weight at 4 years of age than children not exposed to metformin in utero, as reported by Medscape Medical News.

"I think that together with the woman we need to discuss the pros and cons, and decide whether she should take metformin or not," principal investigator of PregMet 2, Eszter Vanky, MD, PhD, also of Norwegian University of Science and Technology, told Medscape Medical News.

Session moderator Corrine K Welt, MD, a reproductive endocrinologist at the University of Utah, Salt Lake City, said: "This Norwegian group has been doing very large studies and looking at big numbers, so they are quite well-powered to look at these endpoints."

She added that the offspring weight issue "is a little bit worrisome" and urged cautious optimism. "I don't think these results are final. It's hopeful metformin might be a treatment for prevention of preterm birth and late miscarriage, but we probably need more data." 

Metformin Compliance Halved Rate of Miscarriage, Preterm Birth

In PregMet 2, 487 pregnant women with PCOS from 14 centers in Norway, Sweden, and Iceland were randomized to metformin (500 mg twice daily the first week, 1000 mg twice daily until delivery) or placebo. At baseline, women were an average age of 30 years and had a mean BMI of about 28 kg/m2

In the intent-to-treat analysis of the entire cohort, the composite incidence of late miscarriage and preterm birth was 5% with metformin vs 10% with placebo, giving an nonsignificant odds ratio (OR) of 1.99 (P = .08). 

However, a per-protocol analysis excluding 46 women who dropped out yielded a significant OR of 2.55 (P = .03) in favor of metformin. A second per-protocol analysis further excluding 134 women with less than 90% adherence to study medication also produced significantly different rates of the two adverse outcomes (OR, 2.76; P = 0.05).

The number needed to treat was 22 to prevent one miscarriage or preterm birth. 

For comparison, the overall rate of miscarriage and preterm births in the Norwegian population is 5.2%, similar to other Nordic countries.

In the United States, that rate is 10%, according to the Centers for Disease Control and Prevention.

"This indicates that our treatment group at least doesn't have more preterm births than the general population," Løvvik noted. 

Does Metformin Work in Pregnant Women?

The "most surprising and striking" secondary outcome was the lack of significant effect on gestational diabetes, with rates of 25% in the metformin group vs 24% with placebo (P = .75), Løvvik said. 

And, in response to an audience member's question, Løvvik noted that her group had previously found the same thing.

"We have done extensive analysis on glucose homeostasis...and there was no effect whatsoever of metformin, no matter which method we used to measure it, whether by fasting blood glucose, 2-hour oral glucose tolerance test, HOMA, HOMA-IR, or insulin levels. Nothing."

Speaking with Medscape Medical News, both Løvvik and Vanky pointed out that this has obvious implications for the widespread use of metformin in treating gestational diabetes, which is currently recommended by professional society guidelines.

Vanky speculated that there might be something about the state of pregnancy that negates the glucose-lowering effect of metformin. 

"We don't know, but there must be a physiologic reason," she said. 

Less Maternal Weight Gain, Higher Head Circumference

There were no differences in hypertension or pre-eclampsia, mode of delivery, or postpartum hemorrhage between those taking metformin or placebo. 

Weight gain among women was lower with metformin than placebo (8.7 vs 11.5 kg; P < .001).

There were no differences in weight or length of offspring at birth, just slightly larger head circumference, although within the normal range, of babies born to mothers who took metformin (35.4 vs 35.0 cm; P = .006). 

Serious adverse events were similar between the two groups. Malformations were also equally distributed.

There were five intrauterine deaths, 3 with metformin and 2 with placebo — a high rate overall, but more likely related to PCOS than the medication, Løvvik concluded.

The study was funded by the Norwegian Research Council and Novo Nordisk Foundation. Løvvik and Vanky have reported no relevant financial relationships. Welt writes for UpToDate and consults for Novartis.

ENDO 2018. March 20, 2018; Chicago, Illinois. Abstract OR33-4

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.


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.