Recurrent Miscarriage With PCOS: The Workup and What Treatment Evidence Actually Shows

Does PCOS Change the Recurrent-Loss Workup?

No — and that single fact is the most useful thing on this page. The standard evaluation for recurrent pregnancy loss is the same test panel whether or not the person has PCOS, and having a PCOS diagnosis already on the chart is, in practice, one of the more common reasons that evaluation gets skipped. A clinician sees an existing explanation for irregular cycles and reduced fertility, attributes the loss to it, and the four-part workup that would have run for anyone else with the same history of loss never gets ordered. Nothing in the actual PCOS miscarriage data supports treating a PCOS diagnosis as a reason to stop looking — a full diagnostic workup found PCOS itself present in only 8–10% of recurrent-loss cases once precise diagnostic criteria were applied, far below the ultrasound-appearance figure most people have heard.

This page covers what that workup actually contains, and — separately — what the evidence says about treatments marketed for preventing another loss. It is a companion to, not a replacement for, the rate question above — the two are answered on separate pages on purpose.

You may see PCOS written as polyendocrine metabolic ovarian syndrome (PMOS), after a 2026 global consensus of more than 50 organisations renamed it. Nothing about the workup or the evidence below changes under either name.

How Many Losses Count as “Recurrent,” and Does It Include a Biochemical Pregnancy?

Two losses is the threshold the current American Society for Reproductive Medicine committee opinion uses to define recurrent pregnancy loss, and its position on biochemical losses is explicit: a pregnancy confirmed only by a positive urine or blood hCG test counts toward that number, ultrasound or tissue confirmation is not required, and the losses do not need to be consecutive. The 2022 update to the European Society of Human Reproduction and Embryology’s guideline lands on the same number — two or more pregnancies, excluding ectopic and molar pregnancy — so the two bodies most often cited in this space actually agree on the count. The guideline that departs from both is the UK’s, and it says so directly: the Royal College of Obstetricians and Gynaecologists’ Green-top Guideline No. 17 defines recurrent miscarriage as three or more first-trimester losses, though it gives a clinician discretion to start investigating after two if something about the pattern looks pathological rather than sporadic.

The practical upshot: if you have had two losses and your clinician cites “three” as the reason to wait, that is the UK guideline’s threshold, not the only one in use internationally — and even RCOG allows earlier investigation at a clinician’s discretion when the pattern warrants it.

What Does a Full Recurrent-Loss Workup Actually Test?

Four categories make up the standard evaluation, and none of them is specific to PCOS: clotting antibodies, thyroid function, chromosome analysis, and the shape of the uterine cavity. Each exists to find one specific, sometimes treatable explanation — and each is worth naming so you can check whether it was actually done, rather than assumed unnecessary because a PCOS diagnosis already existed.

Antiphospholipid Syndrome Testing

Antiphospholipid syndrome (APS) is tested with three specific blood markers — lupus anticoagulant, anticardiolipin antibodies, and anti-β2-glycoprotein I antibodies — measured on two occasions at least 12 weeks apart, per the international consensus classification criteria that define the condition. APS is one of the few causes of recurrent loss with both a confirmatory blood test and a treatment with real trial support behind it, which makes it worth confirming or ruling out early rather than assuming a pregnancy loss is unexplained without having checked.

Thyroid Function

Thyroid testing matters here for a reason distinct from the general PCOS diagnostic work-up: a 2023 meta-analysis of 10 observational studies found that recurrent-loss patients with positive thyroid antibodies had a 46% higher risk of a subsequent pregnancy loss than those without (RR 1.46, 95% CI 1.20–1.85), and that association held even in women whose thyroid hormone levels were otherwise normal. The 2023 international PCOS guideline already recommends TSH testing to rule out thyroid disease as a mimic before a PCOS diagnosis is even confirmed — this is a separate, additional reason to check it once loss has actually occurred, not a test the PCOS work-up already covered for this purpose.

Parental Karyotype and Products-of-Conception Testing

Chromosomal analysis of the pregnancy tissue itself — not just the parents’ chromosomes — is the test that tells you what actually happened in a specific loss, and a large retrospective study of 1,309 women with 2 to 20 consecutive first-trimester losses found something counterintuitive about it: the proportion of losses with an abnormal embryonic karyotype fell, and the proportion with a normal karyotype rose, as the number of prior losses increased. A normal karyotype in a tested loss was actually a stronger predictor of a further miscarriage than an abnormal one — the opposite of what “it was just chance” would predict. In plain terms: the more losses someone has already had, the less likely a random chromosomal event explains the next one, and the more reason there is to look for a specific, findable cause rather than write off another loss as bad luck.

Uterine Cavity Assessment

A structural uterine anomaly is found in roughly one in six women with recurrent loss — 16.7%, against 6.7% in the general population — per a critical review of the diagnostic literature (Saravelos et al. 2008), which also found that two-dimensional ultrasound and standard hysterosalpingography are not accurate enough on their own to rule an anomaly out. Sonohysterography, three-dimensional ultrasound, or combined hysteroscopy and laparoscopy are what that review found necessary for a definitive answer — worth knowing before accepting a normal-looking standard scan as the final word on cavity shape.

Does Metformin Prevent Miscarriage? Report the Evidence, Don’t Round It Up

The honest answer is that the strongest study design available found no clear benefit, despite an older, much larger-looking effect in weaker data. PregMet2, a randomised, placebo-controlled trial, tested metformin taken throughout pregnancy against a combined outcome of late miscarriage and preterm birth and found 5% on metformin versus 10% on placebo — a gap that did not reach statistical significance in the trial on its own. An earlier, much smaller retrospective study at a single US clinic had reported a far larger difference, but a retrospective design cannot rule out that women who tolerated and stayed on the medication differed systematically from those who did not — exactly the kind of bias a randomised trial exists to remove. Metformin’s full fertility evidence picture, beyond this specific pregnancy-loss question, is covered separately. Based on the trial evidence available now, metformin through pregnancy is not an established way to prevent recurrent loss in PCOS specifically.

Does Progesterone Prevent Recurrent Miscarriage? A Well-Powered Trial Was Negative

The PROMISE trial is the clearest single answer available, and it does not support routine progesterone for this purpose. In 836 women with a history of unexplained recurrent miscarriage, first-trimester vaginal progesterone (400 mg twice daily from a positive pregnancy test through 12 weeks) produced a live birth rate of 65.8%, against 63.3% on placebo — a 2.5 percentage-point difference that did not reach statistical significance (relative rate 1.04, 95% CI 0.94–1.15). This was a well-powered, double-blind, placebo-controlled trial specifically in the population progesterone is most often marketed to, and it found no benefit. Progesterone is not shown to prevent recurrent loss in women whose losses are unexplained.

Do Aspirin and Heparin Prevent Recurrent Miscarriage? It Depends Entirely on Why the Loss Happened

This is the clearest case on this page where the same treatment has two different evidence verdicts depending on what the workup actually found — which is exactly why the workup above comes first.

In women with recurrent loss and no confirmed clotting disorder, low-dose aspirin combined with heparin does not improve the live-birth rate over placebo. A randomised trial of 364 women with unexplained recurrent miscarriage found live-birth rates of 54.5% on aspirin plus heparin, 50.8% on aspirin alone, and 57.0% on placebo — no significant difference, and placebo numerically highest. The 2023 ALIFE2 trial tested the same combination in 326 women who specifically had recurrent loss and a confirmed inherited thrombophilia, and still found no benefit — 72% live births on heparin versus 71% on standard care — leading the trial’s own authors to advise against both routine heparin use and routine inherited-thrombophilia screening in recurrent loss.

In women with confirmed antiphospholipid syndrome specifically, the picture reverses. A network meta-analysis of 54 randomised trials and 4,957 women with confirmed APS found that aspirin combined with heparin substantially increased the live-birth rate over aspirin alone or no treatment, with odds ratios for various regimens ranging from 2.88 to 11.24 — and the review’s authors concluded that low-dose aspirin plus heparin is the appropriate first-line treatment specifically for confirmed APS.

What If the Workup Finds Nothing? Unexplained Is Not the Same as Untreatable

A large share of recurrent pregnancy loss remains unexplained even after a complete four-part workup, and that is a statement about the limits of current testing, not a verdict on the pregnancy or on you. Both the ASRM and ESHRE guidance behind the numbers on this page were written specifically because clinicians manage this population routinely — an unexplained result is a common, well-documented outcome of a proper work-up, not a dead end. What “unexplained” does mean in practice: none of the four tests above found antiphospholipid syndrome, a thyroid problem, a chromosomal cause, or a structural uterine finding. It does not mean nothing can be offered, and it does not mean the next pregnancy is doomed to repeat the pattern — most women who conceive again after an unexplained loss, including in the trial populations cited above, go on to a live birth either way, with or without added medication.

Who This Page Does Not Answer For

This page is one piece of the site’s wider PCOS fertility coverage, and it describes what a standard workup contains and what trial evidence shows about specific treatments — it is not a substitute for the work-up itself, and it does not tell you which result applies to your case. If you have not yet had two losses formally evaluated, the referral timelines and what a fertility specialist adds beyond a GP are covered separately. If a loss has already happened and you are still processing it, that is a conversation for a GP, a midwife, or a bereavement-focused support service, not something this page or any statistic can resolve. And none of the treatment evidence above changes the guideline-based monitoring that applies once a pregnancy is ongoing — the four PCOS pregnancy risks worth tracking beyond miscarriage are a separate, related picture.