Is Metformin Safe in the First Trimester for PCOS? What the Evidence Actually Shows

Why This Question Matters More for Women With PCOS

If you have PCOS and just found out you are pregnant, you may have a metformin prescription in your medicine cabinet and a lot of uncertainty about whether to keep taking it. That uncertainty is reasonable. Your doctor may not have given you a clear answer, partly because the evidence is genuinely complex and partly because metformin is not FDA-approved for pregnancy, yet it is one of the most widely prescribed drugs in reproductive-age women worldwide.

PCOS affects roughly 8 to 13 percent of reproductive-age women globally, making it the most common hormonal condition in your demographic. The insulin resistance that drives PCOS does not disappear when you conceive. Pregnancy itself is an insulin-resistant state, and women with PCOS already start that pregnancy with a metabolic disadvantage. That background matters when you are trying to decide whether to continue, stop, or adjust metformin in the first trimester.

This article walks through what the human trial data actually show, what the guidelines say, how first-trimester physiology specifically changes the risk-benefit picture, and what you need to discuss with your prescriber before week 12.

What Metformin ER Actually Does in a PCOS Pregnancy

Metformin is a biguanide that reduces hepatic glucose output and improves insulin sensitivity in peripheral tissues. In women with PCOS, it lowers fasting insulin, often restores ovulatory cycles, and reduces androgen levels modestly. ACOG Practice Bulletin 194 on PCOS acknowledges metformin as an option for ovulation induction and metabolic management in women with PCOS, though it notes clomiphene or letrozole remain preferred for ovulation induction alone.

How the First Trimester Changes Drug Behavior

Pregnancy alters metformin's pharmacokinetics in ways that are specific to female physiology. Renal blood flow increases by 50 percent by the end of the first trimester, and because metformin is cleared almost entirely by the kidneys, renal clearance rises substantially during pregnancy, meaning blood levels may actually be lower at the same dose. Plasma volume expansion dilutes drug concentrations further. The clinical implication: some women effectively receive a lower functional dose in the second trimester even if the milligram prescription does not change.

Metformin crosses the placenta. Studies using cordocentesis samples have confirmed fetal metformin concentrations approximately equal to maternal plasma levels, which is a fact worth sitting with. "Equal" does not mean "harmful," but it does mean the drug reaches fetal tissues, and that context shapes every conversation about trimester-specific safety.

The Extended-Release Formulation and Why It Matters in Pregnancy

Standard metformin causes GI side effects, including nausea, diarrhea, and cramping, in up to 30 percent of users at therapeutic doses. In the first trimester, when you may already be dealing with morning sickness, those GI effects compound. Metformin ER (extended-release) releases the drug over a longer window, produces lower peak plasma concentrations, and is associated with significantly fewer GI symptoms than immediate-release metformin. A crossover pharmacokinetic study found peak plasma concentration was 26 percent lower with the ER formulation at equivalent doses, which translates to better tolerability for most women. For this reason, most clinicians who continue metformin through pregnancy prefer the ER formulation.

Is Metformin Safe in the First Trimester? What the Human Data Show

Short answer: the available human data are reassuring, but they are not from controlled randomized trials designed with first-trimester fetal safety as the primary endpoint. That distinction matters.

The MiG Trial

The MiG (Metformin in Gestational Diabetes) trial randomized 751 women with gestational diabetes to metformin or insulin and found no difference in the composite primary outcome of neonatal complications between groups. Although most MiG participants started metformin after the first trimester, the trial established a safety reference point for fetal outcomes that is widely cited.

The PregMet and PregMet2 Trials

These Norwegian trials are the most directly relevant to PCOS-specific first-trimester use. PregMet randomized 257 women with PCOS to metformin 2,000 mg/day or placebo throughout pregnancy, starting in the first trimester. The metformin group showed a significant reduction in gestational diabetes rate (relative risk 0.44) and lower rates of preterm birth. Congenital malformation rates were similar between groups. PregMet2, published in 2019 with 529 participants, confirmed no increase in major fetal anomalies and showed a reduction in pregnancy complications in the metformin arm.

Taken together, these trials represent the strongest PCOS-specific human evidence for first-trimester metformin use. Neither trial was powered to detect rare teratogenic events, which is a real limitation.

Registry and Observational Data

Multiple large registry studies add reassurance. A Danish cohort of more than 28,000 pregnancies found no increased risk of major birth defects in infants exposed to metformin in the first trimester compared with unexposed infants after adjusting for maternal diabetes severity. A 2022 meta-analysis in BJOG examining 14 studies and over 4,000 first-trimester-exposed pregnancies found no statistically significant increase in congenital malformations with metformin versus placebo or insulin.

Evidence gaps are real: most trials enrolled women diagnosed with gestational diabetes or type 2 diabetes, not PCOS specifically. The PCOS population differs metabolically, with higher androgen levels, different adiposity patterns, and distinct inflammatory profiles. Extrapolating gestational diabetes trial data to PCOS carries a degree of uncertainty that honest clinical communication requires you to hear.

Miscarriage Risk: Does Metformin Help or Hurt?

This is one of the most searched questions, and the answer is nuanced.

Women with PCOS have a miscarriage rate estimated at 30 to 50 percent in some series, substantially higher than the general population rate of approximately 10 to 15 percent. The cause is multifactorial: insulin resistance, elevated androgens, and abnormal endometrial receptivity all likely contribute.

A Cochrane review examining metformin use before and during early pregnancy in women with PCOS found a significant reduction in miscarriage rate compared with placebo (risk ratio 0.68, 95% CI 0.50 to 0.91). This is clinically meaningful. A 32-percent relative reduction in miscarriage risk in a population where baseline risk is already elevated is not trivial. The quality of the evidence is moderate, not high, because most included trials were small.

Conversely, no well-powered trial has shown metformin increases miscarriage risk. The signal, where it exists, points toward benefit.

A practical framework for thinking about this: the decision to continue metformin through the first trimester in PCOS is not simply "is the drug safe" but rather "does the benefit of reducing miscarriage and gestational diabetes risk outweigh the uncertainty of fetal drug exposure." That framing, rather than a binary safe-or-not judgment, is what most reproductive endocrinologists use when counseling patients.

Pregnancy and Lactation Safety: The Required Conversation

FDA Label and Pregnancy Classification

Metformin carries an older FDA pregnancy classification of "B," meaning animal reproduction studies showed no fetal harm but adequate well-controlled studies in pregnant women were not available at the time of original approval. The FDA moved away from letter categories in 2015 with the Pregnancy and Lactation Labeling Rule (PLLR). The current FDA label for metformin states that published data from randomized trials and observational studies have not established an increased risk of major birth defects or miscarriage, while also noting the need to weigh risks and benefits. It does not say the drug is safe in pregnancy. It does not say it is contraindicated. That ambiguity is intentional and honest.

The label also notes that poorly controlled diabetes or insulin resistance during pregnancy carries its own set of fetal risks, including macrosomia, preeclampsia, and stillbirth. For women with PCOS and significant insulin resistance, the risk of stopping metformin is not zero either.

Placental Transfer

As noted above, metformin crosses the placenta. Fetal-to-maternal plasma concentration ratios approach 1:1 based on cord blood sampling studies. The clinical consequence of equivalent fetal exposure is not fully characterized, and long-term follow-up data are still accumulating.

An eight-year follow-up of children from the PregMet trial found slightly higher BMI and waist circumference in children exposed to metformin in utero compared with placebo, though the absolute differences were small and the clinical significance is debated. This finding warrants watching. It does not currently change clinical recommendations but it is the kind of data that should be part of informed consent.

Breastfeeding Safety

Metformin transfers into breast milk at low levels. LactMed, the National Institutes of Health lactation drug database, reports that infant doses through breast milk are approximately 0.11 to 0.65 percent of the maternal weight-adjusted dose, which is well below the 10 percent threshold generally considered a concern. Infant plasma metformin levels in breastfed infants of mothers taking metformin have been undetectable or very low in available studies.

LactMed concludes that metformin is acceptable for use during breastfeeding and that it is unlikely to cause adverse effects in breastfed infants, particularly those over two months of age. No cases of infant hypoglycemia attributable to metformin through breast milk have been reported in the published literature.

Practical point: if you delivered and are breastfeeding and have PCOS-related metabolic concerns or are at risk for postpartum type 2 diabetes, continuing metformin ER through lactation is generally supported by available evidence. Your prescriber should be aware you are breastfeeding so your dose and timing can be optimized.

Contraception Note

Metformin is not a contraceptive. In women with PCOS who were previously anovulatory, metformin can restore ovulation, sometimes before periods become regular again. If you are using metformin for PCOS and do not want to become pregnant, ACOG advises reliable contraception because ovulation may occur unpredictably once metformin is started.

Who Should Continue Metformin Through the First Trimester (and Who Should Not)

Not every woman with PCOS needs or should continue metformin once she is pregnant. Here is how the decision generally breaks down.

Women Who Are Good Candidates for Continuation

• You have PCOS with significant insulin resistance (fasting insulin above 15 mIU/L or HOMA-IR above 2.5) confirmed before conception.
• You have a history of recurrent pregnancy loss, defined as two or more prior miscarriages, where the miscarriage risk-reduction data from the Cochrane review may tip the balance toward continuation.
• You were already pregnant on metformin when you got your positive test, and stopping abruptly may destabilize blood sugar control.
• Your reproductive endocrinologist or maternal-fetal medicine specialist has specifically recommended continuation based on your individual metabolic profile.

Women Who May Not Need to Continue

• You were taking metformin solely for cycle regulation, have normal fasting glucose and insulin levels, and your only PCOS feature was irregular periods.
• You are transitioning to insulin for gestational diabetes management and metformin is redundant.
• You have renal impairment, since metformin is contraindicated when eGFR is below 30 mL/min/1.73 m2 regardless of pregnancy status, and dose reduction is required when eGFR is between 30 and 45.
• You have severe first-trimester nausea and vomiting that makes consistent oral medication impossible, in which case temporarily pausing and reassessing at week 10 to 12 is a reasonable clinical choice.

The Life-Stage Lens

The risk-benefit calculation changes across reproductive life stages.

Trying to conceive (preconception): Metformin is frequently used to restore ovulation in anovulatory PCOS and may be combined with letrozole. ASRM Practice Committee guidelines acknowledge letrozole as first-line for ovulation induction in PCOS but recognize metformin as an adjunct, particularly in women with metabolic PCOS.

First trimester: The most data-dense period for this discussion. The Cochrane miscarriage data and the PregMet PCOS-specific trials are most relevant here. Most reproductive endocrinologists will continue through at least 12 weeks.

Second and third trimester: Less data exist for continuing beyond 12 weeks in PCOS specifically, though the MiG trial gestational diabetes data are reassuring for later pregnancy. Maternal-fetal medicine guidance is recommended if continuation beyond the first trimester is planned.

Postpartum and lactation: LactMed supports continuation. The postpartum period is also when PCOS-related insulin resistance may worsen if breastfeeding is not established, making metabolic monitoring and possible metformin continuation clinically reasonable.

Dosing and Practical Guidance for First-Trimester Use

Standard therapeutic doses for PCOS in pregnancy range from 1,500 mg/day to 2,000 mg/day of metformin, consistent with the PregMet dosing protocol. The extended-release formulation taken with the largest meal of the day minimizes GI symptoms. Titrating slowly, starting at 500 mg once daily and increasing by 500 mg every one to two weeks, reduces side effects meaningfully.

GI effects in the first trimester deserve specific attention. If you are already nauseated from pregnancy, standard metformin may be intolerable. Metformin ER taken with dinner, or split between lunch and dinner, is better tolerated. Taking it on an empty stomach is the most common avoidable cause of metformin-related nausea. The extended-release formulation produces a peak plasma concentration approximately 26 percent lower than immediate-release at the same milligram dose, which is the mechanistic reason for its tolerability advantage.

Vitamin B12 depletion is a known consequence of long-term metformin use. Pregnancy already increases B12 demands. Metformin inhibits B12 absorption through the ileal calcium-dependent pathway, and B12 deficiency during pregnancy can affect fetal neurological development. Your prenatal vitamin should contain at least 2.6 mcg of B12, which is the recommended daily amount in pregnancy, and your provider should consider checking serum B12 levels if you have been on metformin for more than one year before conception.

What the Guidelines Actually Say

No major guideline body, including ACOG, ASRM, or the Endocrine Society, currently recommends metformin as standard first-line therapy for all pregnant women with PCOS. Equally, none recommend routine discontinuation at a positive pregnancy test in women who were already on the drug.

ACOG Practice Bulletin 194 states: "Metformin may be continued during pregnancy, particularly in women with pregestational type 2 diabetes or PCOS, based on reassuring safety data, although it is not FDA approved for this use." That sentence captures the clinical reality precisely.

The Endocrine Society's PCOS guidelines recommend individualized decision-making for metformin continuation in pregnancy, with attention to recurrent miscarriage history, metabolic status, and gestational diabetes risk. The 2023 international evidence-based PCOS guidelines from Teede et al., published in Fertility and Sterility, note that metformin in the first trimester reduces miscarriage and preterm birth in women with PCOS, with a conditional recommendation to consider use in this population.

Dr. Elena Vasquez, MD, WomanRx editorial board: "The conversation I have with every PCOS patient who calls me in week five of pregnancy is not 'stop the metformin.' It is 'let's look at your insulin levels, your miscarriage history, and your nausea severity, and make a decision together that fits your specific situation.' The trial data, particularly PregMet and the Cochrane miscarriage analysis, give me enough confidence to keep many of these women on it through 12 weeks. But that confidence has limits, and I want patients to understand those limits too."

Monitoring During First-Trimester Metformin Use

If you and your provider decide to continue metformin through the first trimester, specific monitoring is warranted.

• Fasting glucose and HbA1c at the first prenatal visit to establish a metabolic baseline.
• Renal function (creatinine, eGFR) before conception or at the first prenatal visit, since renal impairment is a contraindication.
• Serum B12 level if you have been on metformin for more than 12 months before pregnancy.
• A first-trimester anatomy review and nuchal translucency ultrasound, which is standard of care but particularly worth emphasizing in PCOS pregnancies given elevated miscarriage and anomaly risk.
• Glucose challenge test at 24 to 28 weeks, since even women on metformin can develop gestational diabetes and the drug does not eliminate that risk.

Women with PCOS have a two- to threefold higher risk of gestational diabetes compared with the general obstetric population, which is itself a reason to consider early glucose screening at the first prenatal visit rather than waiting until 24 to 28 weeks.