Metformin Safety Signals & FDA Actions: What Every Woman Needs to Know

What Metformin Actually Does in the Body

Metformin is a biguanide that lowers blood glucose primarily by suppressing hepatic gluconeogenesis. It does not stimulate insulin secretion, which is why it does not cause hypoglycemia when used alone. Secondary mechanisms include improving peripheral insulin sensitivity in skeletal muscle and slowing intestinal glucose absorption. At the molecular level, metformin activates AMP-activated protein kinase (AMPK), a cellular energy sensor, and inhibits mitochondrial complex I, which reduces the NAD+/NADH ratio and limits the substrate available for gluconeogenesis.

Why the Mechanism Matters for Women Specifically

Women with polycystic ovary syndrome (PCOS) carry a disproportionate burden of hyperinsulinemia. Excess insulin drives ovarian androgen production, which worsens the hormonal picture. By reducing circulating insulin, metformin can lower androgen levels, partially restore ovulatory cycles, and improve the metabolic profile without the weight gain associated with insulin secretagogues. This is mechanistically distinct from simply "lowering blood sugar," and it explains why metformin appears in PCOS management even when a woman does not have overt type 2 diabetes.

Insulin sensitivity in women is not static. It changes across the menstrual cycle, drops sharply in the second half of pregnancy, and declines again in perimenopause as estrogen falls. Because metformin's effect depends on the degree of insulin resistance it is acting against, the same dose may produce different glycemic effects depending on where you are in your hormonal life.

The FDA's Boxed Warning: Lactic Acidosis

The single most serious safety signal on metformin's label is lactic acidosis. The FDA placed a boxed warning on all metformin products because the drug can, under specific circumstances, impair mitochondrial lactate clearance and cause a life-threatening accumulation of lactic acid.

How Common Is It Really?

Lactic acidosis attributed to metformin occurs at roughly 3 cases per 100,000 patient-years, based on pharmacovigilance data. That figure is low, but fatality rates when it does occur are approximately 50 percent, which is why the warning exists. The risk concentrates in people who use metformin despite contraindicated renal function, significant hepatic impairment, or acute illness causing tissue hypoxia.

The Renal Function Threshold

The FDA revised its renal guidance in 2016, moving away from a serum creatinine cutoff (which systematically disadvantages women because they produce less creatinine per unit of muscle mass) toward estimated glomerular filtration rate (eGFR). Current FDA guidance:

• eGFR <30 mL/min/1.73 m²: metformin is contraindicated
• eGFR 30-45 mL/min/1.73 m²: use requires caution, dose reduction, and more frequent monitoring
• eGFR >45 mL/min/1.73 m²: generally safe to continue

The shift to eGFR is clinically meaningful for women. An older serum creatinine threshold of 1.4 mg/dL (the prior female cutoff in labeling) could exclude women with genuinely adequate renal function and include women who were actually at risk because their lean muscle mass kept creatinine artificially low.

Illness, Surgery, and Contrast Media

The FDA requires metformin be held before iodinated contrast procedures in patients with eGFR <60, and it should be withheld during acute illness causing dehydration, sepsis, or hemodynamic instability. These situations create the acute renal hypoperfusion that raises lactic acid risk. After contrast, recheck renal function at 48 hours before restarting.

The NDMA Contamination Recall: What Actually Happened

In May 2020, the FDA announced that some extended-release metformin products had tested above acceptable daily intake limits for N-nitrosodimethylamine (NDMA), a probable human carcinogen. The FDA's NDMA investigation led to voluntary recalls of specific lots from manufacturers including Apotex, Amneal, and Extended Pharm.

What NDMA Is and Is Not

NDMA is a nitrosamine impurity that can form during synthesis or degrade into from certain drug precursors. It was the same compound that triggered the massive ranitidine (Zantac) recall. Immediate-release metformin tested at levels well below the FDA's acceptable daily intake limit of 96 nanograms per day. Extended-release tablets, which spend longer at high temperature in formulation, showed higher levels in some lots.

The FDA did not recall all extended-release metformin. Patients were advised to check the FDA recall list and switch to an unaffected product or the immediate-release formulation rather than stopping metformin abruptly. Abrupt discontinuation in women with PCOS or type 2 diabetes carries real glycemic risk.

What This Means for You Now

As of 2024, the FDA continues periodic testing of metformin products. The agency's current metformin testing data shows that most lots on the market test below the acceptable limit. If you are on extended-release metformin, ask your pharmacist to confirm your current manufacturer's lot status.

Vitamin B12 Depletion: The Underappreciated Signal

Long-term metformin use reduces vitamin B12 absorption by approximately 30 percent over a decade, according to data from the Diabetes Prevention Program Outcomes Study (DPPOS). Metformin appears to interfere with calcium-dependent uptake of the vitamin B12-intrinsic factor complex in the terminal ileum.

Why Women Carry More of This Risk

Women are already at higher baseline risk for B12 deficiency, particularly those who follow vegetarian or vegan diets (common in women with PCOS seeking dietary interventions), women with autoimmune gastritis, and women over 50 whose gastric acid production declines. Peripheral neuropathy from B12 deficiency can mimic diabetic neuropathy, meaning the deficiency may go undetected and be attributed to the wrong cause.

Current American Diabetes Association (ADA) guidance suggests periodic monitoring of B12 levels in patients on long-term metformin, particularly those with symptoms of neuropathy or anemia. Annual testing is reasonable for women on metformin for more than three years.

The Landmark Efficacy Trial and What It Does (and Does Not) Tell Women

The United Kingdom Prospective Diabetes Study 34 (UKPDS 34), published in The Lancet in 1998, remains the foundational efficacy trial for metformin. In overweight patients with newly diagnosed type 2 diabetes, metformin produced a 32 percent reduction in any diabetes-related endpoint compared to conventional dietary therapy, with a 42 percent reduction in diabetes-related death and a 36 percent reduction in all-cause mortality.

The Evidence Gap for Women

UKPDS 34 enrolled 1,704 participants, of whom fewer than half were women, and the trial did not power or pre-specify sex-stratified analyses. The mortality and cardiovascular benefit data are therefore extrapolated to women rather than directly studied in them. This is a genuine limitation. Women have historically been under-represented in diabetes trials, and sex-specific pharmacokinetic data for metformin remain sparse. What we do know from pharmacokinetic studies is that women tend to have slightly higher peak plasma concentrations of metformin at equivalent doses, which may translate to both greater efficacy and marginally higher GI side-effect rates.

How Metformin's Safety Profile Shifts Across Women's Life Stages

Reproductive Years and PCOS

For women in their reproductive years with PCOS, metformin is often started at 500 mg once daily with the evening meal and titrated over four to eight weeks to 1,500-2,000 mg daily to minimize gastrointestinal side effects. The GI side effects (nausea, diarrhea, loose stools) are the primary reason women discontinue. Slow titration and consistent food co-administration reduce but do not eliminate them. Extended-release formulations produce fewer GI symptoms in head-to-head comparisons, though the NDMA contamination history makes lot verification important.

Trying to Conceive

For women with PCOS who are trying to conceive, the American Society for Reproductive Medicine (ASRM) notes that metformin may improve ovulation rates and reduce the risk of ovarian hyperstimulation syndrome (OHSS) when used alongside gonadotropin stimulation. The evidence for metformin as a standalone ovulation inducer in PCOS is weaker than for letrozole, which ASRM currently considers first-line for ovulation induction in this population.

Perimenopause and Menopause

Perimenopause deserves specific attention. Estrogen decline reduces insulin sensitivity independent of weight change, and many women notice rising fasting glucose or HbA1c during the menopause transition even without significant dietary changes. The Menopause Society (formerly NAMS) recognizes metabolic deterioration as a core concern of the transition. For a woman who was previously well-controlled on a stable dose of metformin, her glycemic targets may shift, and her dose may need upward adjustment during perimenopause. Conversely, menopausal hormone therapy (MHT) with estrogen can improve insulin sensitivity and may reduce the glucose-lowering dose required.

Pregnancy and Lactation: A Required Clinical Discussion

Metformin's use in pregnancy is one of the most clinically contested areas in women's health, and it deserves a clear framework rather than a vague statement.

Pregnancy Safety Data

Metformin carries an older FDA Pregnancy Category B designation, meaning animal studies showed no harm and adequate human data are reassuring but not conclusive. Metformin crosses the placenta freely. Fetal plasma concentrations approximate maternal concentrations. This is not automatically harmful, but it means the fetus is exposed.

For gestational diabetes mellitus (GDM), several randomized trials including the MiG Trial (Rowan et al., NEJM 2008) found that metformin was noninferior to insulin for glycemic control, with fewer maternal hypoglycemic episodes and less gestational weight gain. However, the MiG trial also reported that 46 percent of participants required supplemental insulin. ACOG Practice Bulletin 190 acknowledges metformin and glyburide as acceptable pharmacologic options in GDM when insulin is not feasible, while noting that long-term offspring data (particularly regarding childhood adiposity and neurodevelopment) are still accumulating.

For women with PCOS, many reproductive endocrinologists continue metformin through the first trimester to reduce miscarriage risk, though the evidence for this practice remains mixed. The decision should be individualized.

What to Do If You Become Pregnant on Metformin

Do not stop metformin abruptly without speaking to your provider. In women with type 2 diabetes, abrupt discontinuation may cause glycemic deterioration that itself carries fetal risk. The conversation with your OB or maternal-fetal medicine specialist should happen before conception if possible.

Lactation

Metformin passes into breast milk at low concentrations. A pharmacokinetic study by Briggs et al. found infant exposure of approximately 0.28 percent of the weight-adjusted maternal dose, well below the 10 percent threshold typically used to flag concern. No adverse effects in nursing infants have been reported in the published literature. Both the Academy of Breastfeeding Medicine and most endocrine guidelines consider metformin compatible with breastfeeding, though infant glucose monitoring has been suggested by some experts as a precaution.

Contraception Considerations

Metformin is not a teratogen in the classical sense, and it does not require mandatory contraception the way drugs like valproate or isotretinoin do. However, because metformin is used in PCOS and can partially restore ovulation in previously anovulatory women, it may inadvertently increase fertility. Women who do not wish to conceive and who are starting metformin for PCOS should be counseled that pregnancy is now more possible, and reliable contraception should be discussed.

Drug Interactions Relevant to Women

Several interactions deserve attention for common prescribing scenarios in women.

Carbonic anhydrase inhibitors (topiramate, zonisamide, acetazolamide) increase the risk of lactic acidosis when combined with metformin by reducing renal tubular secretion of the drug. Topiramate is sometimes prescribed off-label for migraine prevention in women of reproductive age, and this combination warrants monitoring.

Hormonal contraceptives (particularly high-dose estrogen formulations) can worsen insulin resistance and blunt metformin's glycemic effect slightly, though this is rarely clinically significant with modern low-dose pills. No dose adjustment is routinely required.

Alcohol potentiates the risk of lactic acidosis by impairing hepatic lactate clearance. Heavy alcohol use is a relative contraindication to metformin use, not a minor footnote.

Who This Drug Is Right For (and Who Should Be Cautious)

Generally Appropriate

• Women with type 2 diabetes and eGFR >45
• Women with PCOS and insulin resistance or anovulation
• Women with prediabetes, particularly those with BMI >25 and high-risk features (per the Diabetes Prevention Program)
• Women with GDM when insulin is declined or inaccessible, in consultation with OB

Requires Extra Caution or Avoidance

• Women with eGFR <45 (use cautiously; avoid if <30)
• Women with hepatic cirrhosis or significant alcohol use
• Women hospitalized with acute illness, sepsis, or heart failure causing hypoperfusion
• Women undergoing contrast imaging in the setting of renal impairment
• Women over 70 with declining renal reserve, particularly if on NSAIDs for osteoarthritis

The ADA's 2024 Standards of Care note that metformin remains the preferred initial pharmacologic agent for type 2 diabetes in most patients, provided renal function permits.

As Dr. Anne Peters, Director of USC's Clinical Diabetes Programs, has stated in published commentary: "Metformin is one of the safest drugs we have, but 'safe' does not mean 'appropriate for everyone.' The renal function check before prescribing is non-negotiable."

GI Side Effects: Managing the Most Common Reason Women Stop

Gastrointestinal side effects affect up to 25 percent of women starting metformin. Nausea, diarrhea, and abdominal cramping are dose-related and usually transient. Strategies with evidence behind them:

• Start at 500 mg once daily with the largest meal. Increase by 500 mg no faster than every one to two weeks.
• Take with food every single time. Not occasionally. Every time.
• Switch to extended-release if GI symptoms persist on immediate-release; clinical trials show meaningfully lower rates of diarrhea.
• If GI symptoms persist beyond 12 weeks despite these steps, re-evaluate the dose rather than abandoning the drug.

Women tend to report higher rates of GI intolerance than men in observational data, which may reflect the pharmacokinetic differences mentioned above (higher peak concentrations per dose). Slower titration is worth the extra time.