PCOS Risks to Baby: Pregnancy Outcomes and the Heritability Question

This page is about what the evidence says for the baby specifically. The maternal side of PCOS in pregnancy — gestational diabetes, blood pressure and miscarriage — already has its own numbers and monitoring pathway covered elsewhere on this site, and the specific cesarean-rate question sits in its own page too. What follows covers preterm birth, growth and NICU admission for the baby, how much of that risk is explained by metabolic factors rather than the PCOS diagnosis itself, whether conceiving through IVF changes any of it, and the question a lot of pregnant readers are actually asking: whether a daughter will inherit this too.

Does PCOS Raise the Odds of Preterm Birth and NICU Admission?

PCOS raises the odds of preterm birth by about half again over baseline, and roughly doubles the odds a newborn needs neonatal intensive care. A 2024 meta-analysis of 73 studies and 92,881 offspring found preterm birth (defined as delivery before 37 weeks in most included studies) at OR 1.53 (95% CI 1.33–1.75), fetal growth restriction at OR 1.84 (95% CI 1.09–3.10), and low birth weight at OR 1.28 (95% CI 1.04–1.59) in babies born to mothers with PCOS, alongside a mean birthweight roughly 58 grams lower. Those are outcomes measured in the baby directly, not the mother.

A separate meta-analysis focused specifically on neonatal intensive care found the same direction with its own number: pooling 27 studies covering 4,982 women with PCOS against 119,692 without, NICU admission ran at OR 2.32 (95% CI 1.40–3.85). Most babies born to mothers with PCOS are still born at term and go home without a NICU stay — the odds are elevated, not the expected outcome.

Is a Baby More Likely to Be Large When the Mother Has PCOS?

PCOS does not raise the odds of an unusually large baby, despite PCOS roughly doubling the odds of gestational diabetes — a condition well known for producing bigger babies in other contexts. The same 2024 meta-analysis found macrosomia at OR 1.13 (95% CI 0.95–1.35) across 62,489 offspring, and large for gestational age at OR 1.14 (95% CI 0.98–1.33) across 11,081, neither reaching statistical significance. A second, independent 2025 meta-analysis restricted to pregnancies conceived through IVF or ICSI found the same null result for macrosomia and large-for-gestational-age babies, across 16,365 women with PCOS and 111,503 without. Two separate analyses, two different populations, the same absence of a signal — that consistency is itself worth noting.

How Much of This Risk Traces to Gestational Diabetes and Body Weight, Not PCOS Itself?

A large share of the excess preterm-birth risk tracks with body mass index rather than the PCOS label on its own. In the same 2024 meta-analysis, PCOS was linked to higher preterm-birth odds in studies that matched women by age alone — but that link disappeared in the smaller set of studies that matched by age and BMI together. Fetal growth restriction told an even starker story: the overall pooled estimate showed higher odds in PCOS, but neither the age-matched nor the age-and-BMI-matched subgroup analyses found a significant difference at all. Lower average birthweight and higher odds of being small for gestational age were the two outcomes that held up in matched comparisons — a pattern that looks more independent of body weight than the preterm-birth and growth-restriction findings above.

The clearest single piece of evidence on mediation comes from the same paper’s sensitivity analysis, restricted to pregnancies where gestational diabetes was already present in both groups: the odds of adverse birth outcomes were similar between PCOS and non-PCOS pregnancies once gestational diabetes had occurred in both. Once gestational diabetes is on the chart, having a PCOS diagnosis alongside it did not add further measurable risk to the baby in this analysis. The authors are upfront that this comparison drew on a small number of studies and carries limited statistical power — a signal, not a settled finding — but it points the same direction as the BMI-matching results: much of what looks like a PCOS-specific risk to the baby is a metabolic risk that gestational diabetes screening already exists to catch.

A 2026 study of women with PCOS undergoing IVF or ICSI ran a formal mediation analysis and reached a similar conclusion from a different angle: preconception BMI and insulin resistance, rather than the PCOS diagnosis alone, were identified as the mediating pathways behind the higher odds of hypertensive disorders and pre-eclampsia in PCOS pregnancies — both conditions that can themselves affect the baby through reduced placental blood flow. This is the practical reasoning behind the 2023 international guideline’s recommendation that PCOS status be flagged at the start of pregnancy: the modifiable part of this risk runs through insulin resistance and weight, both already screened for, rather than through PCOS as a fixed label.

Does It Matter Whether Conception Was Spontaneous or Through IVF?

Assisted reproductive technology (ART) itself raises the odds of gestational diabetes, regardless of PCOS. A 2020 meta-analysis of 1.89 million singleton pregnancies found gestational diabetes at RR 1.53 (95% CI 1.39–1.69) after ART compared with spontaneous conception, and that elevated risk showed up whether or not the underlying studies included women with PCOS — meaning ART itself, not only the diagnosis that often leads someone to it, is part of the picture. For a reader who conceived through IVF or ICSI because of PCOS, ART-related risk and PCOS-related risk are stacking on the same pregnancy rather than one replacing the other.

Within PCOS pregnancies achieved through ART specifically, the complication pattern the mother sees largely mirrors PCOS pregnancies generally — higher odds of gestational diabetes, hypertensive disorders, pre-eclampsia and preterm birth, in a 2024 meta-analysis of 33 studies and 92,810 ART pregnancies — while the earlier finding on macrosomia and large-for-gestational-age babies held in the ART-only population too. The 2024 meta-analysis found one comparison worth naming here: restricted to post-ART pregnancies only, the odds of small for gestational age were actually lower in the PCOS group than in the non-PCOS group conceiving the same way — the reverse of the pattern in the full, mostly-spontaneous-conception dataset, where the two groups did not differ significantly. ART does not simply add PCOS-shaped risk on top of itself in every direction.

One factor that changes the picture more than ART status alone is whether the pregnancy is a singleton or a twin pregnancy. A 2026 study following 696 women with PCOS and 1,392 matched controls with singleton births, and 201 with PCOS against 402 matched controls with twin births, all conceived through IVF or ICSI, found gestational diabetes at OR 2.23 and very preterm birth at OR 2.08 specifically in the twin-birth group — risks the singleton group did not show at the same magnitude. A multiple-embryo transfer decision, not PCOS alone, is part of what sets that risk.

Will My Daughter Inherit PCOS? What the Evidence Actually Shows

A mother’s PCOS diagnosis raises a daughter’s odds of the condition; it does not fix the outcome. The clearest family-clustering data comes from a study of 78 mothers and 50 sisters of 93 women already diagnosed with PCOS, which found PCOS in 24% of mothers and 32% of sisters, against roughly 4% in the general-population comparison the researchers used — rates that rose to 35% and 40% respectively once the count was narrowed to relatives who were premenopausal and not on hormonal treatment that could mask symptoms. Read the other direction, the same numbers say most relatives of a woman with PCOS — 65 to 76% of them — are not affected at all, even sharing a first-degree relative who is. The twin studies and gene-mapping evidence behind this pattern, including the specific gene regions researchers have identified, covers the mechanism in depth; this page’s job is what it means for a baby, not the underlying genetics.

There is a biological signal detectable even before a daughter reaches puberty. A study of 28 daughters of mothers with PCOS and 33 daughters of unaffected mothers, aged 8 to 16, found average AMH levels of 24.0 pmol/L in the PCOS-mother group against 14.4 pmol/L in the comparison group, alongside higher androgen markers and larger average ovarian volume. That is a real, measurable difference between groups — not a diagnosis, and not a predictive test for an individual girl. Guidelines explicitly recommend against using AMH to diagnose PCOS in adolescence, precisely because puberty itself produces a wide, overlapping range of values in girls who will never develop PCOS at all.

None of this is a reason to run hormone panels on an infant or a young daughter with no symptoms. If a daughter later develops persistent irregular cycles well past the first year after her first period, or clear signs of excess androgen, that symptom pattern — not the family history on its own — is what would warrant a conversation with her pediatrician.

Who This Risk Data Does Not Apply To

Twin pregnancies carry a meaningfully different risk profile than singleton ones, as the gestational-diabetes and very-preterm-birth numbers above show — the singleton-pregnancy figures elsewhere on this page are not the right reference point for a twin pregnancy. Pre-existing type 1 or type 2 diabetes is a different situation from PCOS-related insulin resistance and gestational diabetes; anyone in that position is almost certainly on a monitored plan already, and the mediation findings above, built around gestational diabetes as the marker, do not describe pre-existing diabetes. None of the meta-analyses above broke results down by PCOS phenotype — androgen-dominant, lean, or otherwise — so no figure here can be read as more or less applicable to one phenotype than another; that segmentation simply has not been studied at this scale yet. And a family history of PCOS is not a substitute for a formal diagnosis in a daughter — the criteria and testing pathway stay the same regardless of what a mother’s history predicts.

You may see PCOS referred to as polyendocrine metabolic ovarian syndrome (PMOS), after a 2026 global consensus of more than 50 organisations renamed it. Every figure above still applies under either name — only the label changed, and this article uses PCOS because that is still what most readers search. For what changes once a pregnancy is confirmed and underway, the first-trimester diet shift this diagnosis brings and what a second pregnancy’s numbers show both sit alongside this page in this site’s full fertility section.