PCOS in Pregnancy: The Four Risks Worth Monitoring, With Real Numbers

Does PCOS affect pregnancy? What a 106,690-pregnancy analysis found

Yes, measurably: a 2024 systematic review and meta-analysis pooled 104 studies covering 106,690 pregnancies — 17,384 in women with PCOS and 89,306 without — and found higher odds of miscarriage, gestational diabetes, gestational hypertension, pre-eclampsia and cesarean section in the PCOS group. This page sits inside the wider fertility picture: conceiving is one part of it, and what happens once you are pregnant is the other.

This is the largest dataset assembled on the question, and it confirms the pattern was not a fluke of smaller, older studies: those higher odds remained statistically significant even after matching for age and BMI, and even when the analysis was restricted to the highest-quality studies alone. This is not simply a weight story.

A companion 2024 analysis of 73 studies and 92,881 offspring found the same pattern extends to the baby: higher odds of preterm birth, fetal growth restriction and low birth weight, and a lower average birthweight, in babies born to mothers with PCOS.

The same 2024 analysis also found higher odds of cesarean delivery in PCOS, at OR 1.23 (95% CI 1.06–1.43) across 37 studies — the smallest effect of the outcomes it measured, and a plausible downstream consequence of the other four rather than an independent risk in its own right, since gestational diabetes, hypertensive disorders and growth concerns are all reasons a clinician might recommend a planned delivery.

Neither finding means pregnancy with PCOS is unsafe by default. It means four specific risks are worth naming, quantifying honestly, and monitoring for — which is the rest of this page.

Gestational diabetes and PCOS: the numbers

PCOS raises the odds of gestational diabetes 2.4-fold, at OR 2.41 (95% CI 1.95–2.99) across 55 studies in the 2024 meta-analysis — consistent with the earlier 2006 meta-analysis of 15 smaller studies, which found an OR of 2.94 (95% CI 1.70–5.08) for the same comparison. Two analyses, fifteen years apart, different study pools, the same direction and a similar magnitude — that consistency is itself evidence.

The mechanism is the one PCOS is built on: insulin resistance is already present in a large share of PCOS before pregnancy, and pregnancy itself is a naturally insulin-resistant state as placental hormones rise through the second and third trimesters. Stack a pregnancy-driven insulin-resistance increase on top of a pre-existing one and glucose tolerance is more likely to tip over the diagnostic threshold. Insulin resistance and PCOS covers what that mechanism is doing outside of pregnancy.

Hypertensive disorders: gestational hypertension and pre-eclampsia

PCOS roughly doubles the odds of both conditions: OR 2.20 (95% CI 1.82–2.67) for gestational hypertension across 39 studies, and OR 2.30 (95% CI 1.88–2.82) for pre-eclampsia specifically across 34 studies, in the same 2024 meta-analysis. The 2006 meta-analysis found even wider odds ratios on a smaller dataset — 3.67 (95% CI 1.98–6.81) for pregnancy-induced hypertension and 3.47 (95% CI 1.95–6.17) for pre-eclampsia — which is a reminder that odds ratios narrow as evidence accumulates; the 2024 figures, built on a much larger and more recent dataset, are the more reliable estimate today.

Preterm birth and what happens to the baby

PCOS raises the odds of preterm birth by about half again over baseline: OR 1.53 (95% CI 1.33–1.75) across 54 studies and nearly 82,000 offspring in the 2024 birth-outcomes analysis. The same analysis found higher odds of fetal growth restriction (OR 1.84, 95% CI 1.09–3.10) and low birth weight (OR 1.28, 95% CI 1.04–1.59), and a lower average birthweight — about 58 grams lower — in babies born to mothers with PCOS. Small-for-gestational-age odds were not meaningfully different between groups, which is a useful distinction: PCOS pregnancies trend toward earlier delivery and a modestly lower average birthweight, not toward babies who are undersized for the gestational age they actually reach.

Some of this preterm-birth signal is plausibly explained by the other three risks rather than by an independent mechanism of its own. Gestational diabetes and hypertensive disorders are two of the most common reasons an obstetric team recommends a planned early delivery, and both are more common in PCOS to begin with — so a portion of the excess preterm birth in PCOS pregnancies is likely medically initiated rather than spontaneous labour arriving early. Neither the 2024 nor the 2006 meta-analysis separated spontaneous from medically indicated preterm birth, so this is a plausible reading of the data rather than a settled finding.

Miscarriage: the risk, and why relative and absolute are not the same number

PCOS raises the odds of miscarriage by about half: OR 1.49 (95% CI 1.20–1.85) across 44 studies in the 2024 meta-analysis. Reported on its own, “49% higher odds of miscarriage” sounds alarming. Read against an absolute baseline, it is a smaller shift than the relative number implies — and telling the two apart properly is the actual point of this section.

Here is the arithmetic, using the widely-cited general-population estimate that roughly 10% of confirmed pregnancies end in early loss. A 49% relative increase on a 10% baseline works out to approximately 14–15%, not 59%. The odds ratio describes how much more likely an event is relative to the comparison group; it says nothing on its own about how common the event is in absolute terms. Multiplying a small base rate by a real relative increase still produces a modest absolute change — most pregnancies in PCOS, including this one, do not end in miscarriage.

What actually gets monitored for each risk

None of the four risks above are being discovered for the first time when they appear — each has a monitoring pathway already built into standard antenatal care, made more relevant, not invented, by a PCOS diagnosis.

There is one medication-specific piece of evidence worth naming here: the PregMet2 trial tested whether metformin taken through pregnancy reduced a combined outcome of late miscarriage and preterm birth in PCOS. The trial’s own result — 5% versus 10% — did not reach statistical significance on its own, though a pooled analysis with two earlier trials did. That is not strong enough evidence for a standing recommendation, and it is a question for your own care team, not a decision this page can make for you. Metformin and fertility covers what the trial found in full.

Who is at higher risk within PCOS — and who is not

The 2024 meta-analysis is deliberately blunt on one point: these risks held up after matching for BMI, which means lean PCOS is not a protective category here. Insulin resistance, androgen levels and chronic low-grade inflammation can all be part of PCOS at a normal body weight, and the mechanisms behind gestational diabetes and hypertensive disorders do not require a high BMI to operate.

That said, BMI and PCOS are not independent of each other in most pregnancies that occur — a higher average BMI at conception is itself part of what the 2024 analysis measured in the PCOS group, and higher BMI carries its own additive risk for gestational diabetes and hypertensive disorders regardless of PCOS status. The honest summary is that PCOS adds risk on its own, weight can add further risk on top of it, and the two are worth discussing as separate, stacking factors with your antenatal team rather than assuming either one alone explains your personal risk profile. If you are still working out your own odds of conceiving in the first place, can you get pregnant with PCOS is the earlier question this page assumes has already been answered.

None of the risks above end at delivery, either — PCOS after pregnancy covers what continues afterward, including the specific long-term metabolic risk for anyone who develops gestational diabetes alongside PCOS.

You may see PCOS written 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.