Metformin in First Trimester: Pre-gestational Diabetes vs PCOS
For women with pre-gestational diabetes, metformin can be safely discontinued at the end of the first trimester, while for women with PCOS, continuing metformin throughout the first trimester may reduce miscarriage risk and improve pregnancy outcomes.
Pre-gestational Diabetes (Type 2 Diabetes)
The most recent joint Endocrine Society/European Society of Endocrinology guideline 1 and the 2025 ADA Standards of Care 2 provide clear direction: metformin can be safely stopped at the end of the first trimester in women with pre-existing diabetes due to poor placental crossing from lack of cation transporters during this period 1.
Key Evidence for Pre-gestational Diabetes:
No composite neonatal benefit: Recent RCTs (MiTY and MOMPOD) showed no differences in composite neonatal health outcomes between insulin alone versus insulin plus metformin 2
Concerning findings: One trial showed a doubling of small-for-gestational-age (SGA) neonates with metformin 2, and the MiTY trial specifically identified increased SGA risk in women with chronic hypertension or nephropathy 1
Long-term offspring concerns: The MiG TOFU study found 9-year-old children exposed to metformin had higher BMI, waist-to-height ratio, and waist circumference in one cohort 2, though the 2-year MiTy Kids follow-up showed no differences 2
Critical Contraindication:
Avoid metformin entirely in women with chronic hypertension or nephropathy - this subgroup showed the highest risk of SGA without reduction in preeclampsia 1.
Polycystic Ovary Syndrome (PCOS)
The evidence strongly favors continuing metformin throughout the first trimester in PCOS pregnancies, contrary to pre-gestational diabetes.
Key Evidence for PCOS:
Reduced miscarriage risk: A 2025 systematic review and meta-analysis 3 demonstrated that continuing metformin throughout the first trimester showed possible reduction in miscarriage (OR 0.64) and possible increase in live birth (OR 1.24), while stopping metformin after positive pregnancy test showed an increase in miscarriage rate (OR 1.46)
Indirect comparison advantage: The same meta-analysis found that continuing metformin through first trimester versus stopping it showed consistent trends favoring continuation: miscarriage OR 0.44 and live birth OR 1.14 3
Reduced preterm delivery: Meta-analysis of RCTs showed metformin associated with reduced preterm delivery risk (RR 0.45) 4
No teratogenic effects: Multiple studies confirm no demonstrable teratogenic effects, intrauterine deaths, or developmental delays 5, and no major congenital malformations in metformin-exposed PCOS pregnancies 6
FDA Labeling Context:
The FDA label 7 states that published data have not reported a clear association with metformin and major birth defects or miscarriage, though it notes insufficient data to definitively establish absence of risk. The label emphasizes that poorly controlled diabetes poses greater risks than metformin exposure.
Clinical Algorithm
Step 1: Identify the underlying condition
- Pre-gestational Type 2 diabetes → Consider discontinuation at end of first trimester
- PCOS (without pre-existing diabetes) → Continue through first trimester
Step 2: Screen for high-risk features (if pre-gestational diabetes)
- Chronic hypertension present? → Do not use metformin
- Nephropathy present? → Do not use metformin
- Extreme insulin resistance or barriers to insulin use? → May continue metformin with informed consent
Step 3: Patient counseling
- Discuss placental crossing and fetal exposure 2
- For PCOS: Emphasize miscarriage reduction benefit with continuation 3
- For diabetes: Discuss lack of neonatal benefit and potential SGA risk 2
- Address long-term offspring data showing possible increased childhood adiposity 2, 8
Important Caveats
Placental transfer: Metformin readily crosses the placenta, resulting in umbilical cord blood levels equal to or higher than maternal levels 2. This is relevant for all first-trimester exposures.
No benefit for GDM prevention: Meta-analysis showed metformin does not reduce GDM risk in high-risk individuals with obesity or PCOS 2, and one large RCT found no reduction in GDM when given from first trimester to delivery in PCOS 9.
Gastrointestinal tolerance: Drug intolerance occurred more frequently in some metformin groups 2, which may affect adherence.
The critical distinction: The divergent recommendations stem from different mechanisms - in pre-gestational diabetes, insulin remains the gold standard with metformin offering no clear neonatal advantage and potential SGA risk, while in PCOS, metformin appears to address underlying pathophysiology affecting early pregnancy maintenance 3, 10.