Metformin may reduce miscarriage risk in women with polycystic ovary syndrome (PCOS), but the evidence is not yet definitive. The strongest signal comes from women who start metformin before conception and continue it through the first trimester. In that group, a 2025 meta-analysis of 12 studies found a 36% lower odds of miscarriage compared to placebo, though the result didn’t reach statistical significance. The picture is clearer on one point: stopping metformin abruptly at a positive pregnancy test appears to be worse than either continuing it or never taking it at all.
Why Metformin Might Help
PCOS is closely tied to insulin resistance, a state where your body produces excess insulin to compensate for cells that don’t respond well to it. That surplus insulin drives up androgen levels and creates a hormonal environment that can interfere with implantation, placental development, and early pregnancy maintenance. Metformin works by making your cells more responsive to insulin, which lowers both circulating insulin and the androgens it stimulates. It also reduces glucose production in the liver and helps muscles absorb more glucose from the bloodstream.
The theory behind using metformin in pregnancy is straightforward: correct the insulin resistance, and you correct the downstream hormonal chaos that raises miscarriage risk. This is why its potential benefits are specific to women with PCOS or insulin resistance. There’s no evidence it prevents miscarriage in women without these underlying metabolic issues.
What the Clinical Trials Show
The largest randomized trial on this question, published in the Journal of Clinical Endocrinology & Metabolism, compared metformin to placebo in women with PCOS undergoing standard fertility treatment. The miscarriage rates were nearly identical: 15.2% in the metformin group versus 17.8% in the placebo group. That difference was not statistically meaningful. However, live birth rates were significantly higher with metformin (41.9% versus 28.8%), and women taking metformin were 1.6 times more likely to become pregnant in the first place. So while that particular trial didn’t show a miscarriage benefit, more women ended up with babies.
Smaller studies have been more encouraging. One trial followed 75 pregnant women with PCOS divided into three groups based on how long they continued metformin (stopped at 5 to 6 weeks, continued to 8 weeks, or continued to 12 weeks of pregnancy). The women who continued through 8 weeks saw their miscarriage rate drop from 40% in previous pregnancies to 8%, and those who continued through 12 weeks saw a decline from 32% to 4%. The group that stopped earliest also saw a reduction, from 20% to 4%, but that drop wasn’t statistically significant given the small sample size.
Timing Matters More Than You’d Expect
A 2025 systematic review and meta-analysis pooling data from 1,708 women found that when you stop metformin may be just as important as whether you take it. Women who started metformin before conception and continued through the first trimester had 36% lower odds of miscarriage compared to those receiving placebo or no treatment. That’s a meaningful trend, even though the confidence interval crossed the threshold for statistical significance.
The more striking finding was what happened when women stopped metformin as soon as they became pregnant. That group actually showed a 46% increase in miscarriage odds compared to placebo. This suggests that withdrawing metformin at the moment of a positive pregnancy test, when the metabolic benefits are still needed, could be counterproductive. When researchers indirectly compared the two approaches (continuing versus stopping), continuing metformin through the first trimester was associated with 56% lower odds of miscarriage.
The practical takeaway: if you’re already on metformin for PCOS and become pregnant, the decision about when to stop should involve your provider, and the evidence leans toward not stopping immediately.
Is It Safe During Pregnancy?
Metformin crosses the placenta, which raises understandable concern. But decades of use and multiple meta-analyses have found no increased risk of major birth defects. A 2023 systematic review looking at first-trimester metformin exposure in women with PCOS found no significant increase in congenital malformations in either randomized trials or observational studies. A large European registry study examining over 50,000 infants with congenital anomalies, including 168 exposed to metformin in the first trimester, similarly found no overall increased risk.
The open question is about longer-term effects on children. Some follow-up studies of children exposed to metformin in the womb have found slightly higher rates of childhood obesity and markers of metabolic problems, while other studies have found no such effect. This inconsistency means it’s too early to draw firm conclusions, but it’s worth noting that the short-term safety profile during pregnancy is reassuring.
Typical Dosing in Pregnancy Studies
Most clinical trials have used 1,500 mg per day, split into three doses of 500 mg. This is the same dose commonly prescribed for PCOS outside of pregnancy. In practice, many providers start at a lower dose and increase gradually to minimize the gastrointestinal side effects (nausea, diarrhea, bloating) that metformin is known for. The studies showing the strongest miscarriage reduction started metformin before conception, not after a positive test, and continued it through at least 8 to 12 weeks of pregnancy.
Who Benefits Most
The evidence for metformin and miscarriage prevention applies specifically to women with PCOS, particularly those with clear signs of insulin resistance such as elevated fasting insulin, difficulty managing blood sugar, or a history of recurrent early pregnancy loss. Women without PCOS or insulin resistance have not been shown to benefit.
Even within the PCOS population, the evidence is described as low to moderate quality. The meta-analyses are based on relatively small trials, and the confidence intervals are wide enough that the true effect could range from a substantial benefit to no benefit at all. What the data do suggest consistently is that metformin improves live birth rates and pregnancy rates in PCOS, and that continuing it through the first trimester is preferable to stopping it cold at conception.