My antibodies were positive but 'don't count' — what now?

This article exists because of a specific, common and genuinely painful situation: you've had losses, an antibody test came back positive, and you've been told it doesn't count.

Sometimes that's right. Often it's more complicated than the person saying it has time to explain. Either way you're entitled to understand the actual argument, because this is one of the places where "the guidelines don't support that" is doing more work than the evidence behind it can bear.

Classification criteria were not built to decide who gets treated

This is the piece almost nobody explains, and once you understand it the rest follows.

The APS criteria — the Sydney criteria, and the newer 2023 ACR/EULAR version — are classification criteria. They were designed to define clean, homogeneous groups of patients for research studies, so that trials enrol people who unambiguously have the disease. To do that job they're deliberately tuned for high specificity, which is a technical way of saying they are built to exclude borderline cases on purpose.

They are not diagnostic criteria, and the literature says so directly: classification criteria "are not equivalent to diagnostic criteria." A person can fail to meet them and still have the condition. That isn't a loophole or wishful thinking — it's the acknowledged, designed-in limitation of the tool.

So when a clinician says "you don't meet criteria," the accurate translation is "you wouldn't have been enrolled in the trials" — not "you don't have this."

Non-criteria obstetric APS is a real, named category

Emerging evidence There's an established framework for exactly this group, and published outcome data. Women with obstetric APS features who don't meet full criteria are usually sorted into three subgroups:

• Inconclusive serology, with pregnancy complications that aren't on the criteria list.
• The full clinical picture — the losses — but serology that doesn't quite satisfy the lab criteria. This is the largest source of the argument, and it's where most people reading this will find themselves.
• Antibodies that clearly meet the lab criteria, without the clinical history yet.

In a single-centre study of 140 women stratified this way, those three groups made up 22%, 35% and 43% of the non-criteria cohort. Treated with standard care — low-dose aspirin, with or without low-molecular-weight heparin — the maternal and fetal outcomes were broadly similar to women who did meet full criteria, with a significant improvement in live births. A separate cohort of 91 non-criteria patients reached comparable conclusions.

Safety consideration Read that carefully, because it cuts both ways. It's meaningful evidence that this group can be identified and treated, from observational cohorts rather than randomized trials — which is exactly why guidelines stop short of a general recommendation. Cohort data can't fully separate the effect of treatment from the fact that most women with recurrent loss eventually have a successful pregnancy anyway. It is real evidence. It is not the same grade of evidence as a large randomized trial, and anyone who tells you otherwise is overselling.

Antibodies move — and they move up in pregnancy

Emerging evidence If you've been positive at one point and negative at another, you have probably been told the positive was a fluke. Here's the part that's rarely mentioned: antiphospholipid antibody levels are not stable, and pregnancy itself changes them.

In a 2026 longitudinal study that measured antibodies in 100 women with recurrent loss at three time points, 61% were positive for at least one antibody subtype at one or more of those points — while only one woman met formal APS criteria. Levels rose significantly after conception compared with before it, most strongly for anticardiolipin IgG. Pre-conception positivity was associated with a live birth rate of 44.8%, versus 66.2% in women who tested negative before conceiving.

So "positive during pregnancy, negative outside it" is a documented pattern, not a contradiction that discredits the positive result. It also means when the blood was drawn is part of the result, and a single negative draw at the wrong moment doesn't close the question.

None of which repeals the confirmation rule — positives are meant to be repeated at least twelve weeks apart precisely because infections and other transient causes can produce them. The point is narrower and more useful: one negative doesn't undo a previous positive any more than one positive is a diagnosis.

What the anticoagulation evidence actually covers

If you take one thing from this page, make it this, because it is the most frequently mis-stated fact in recurrent loss.

The major trials showing that blood thinners don't improve live birth rates studied women without APS. ALIFE enrolled women with unexplained recurrent miscarriage. ALIFE2 enrolled women with inherited thrombophilia — Factor V Leiden, prothrombin mutation, protein C and S deficiencies — and explicitly excluded antiphospholipid syndrome. Both were well-conducted, both were negative, and both are the correct basis for not putting antibody-negative women on heparin.

Meanwhile, for women with confirmed obstetric APS, low-dose aspirin plus heparin is the established treatment, associated in observational series with live birth rates around 79%.

Which means: if your antibodies are positive, a clinician citing "the studies show heparin doesn't help" is citing studies you were not in. That is a fair and specific thing to raise, and it's a much stronger question than an argument about whether you feel strongly.

Questions that move this conversation forward

• "Exactly which antibodies were positive, at what titre, and on what date relative to that pregnancy?" Get the numbers, not the summary. "Borderline" is not a result.
• "Was that ever repeated at least twelve weeks later? If not, can we repeat it now, while I'm not pregnant?"
• "Could the timing of the draw explain a negative — and would you interpret a result during pregnancy differently?"
• "Do I fall into what's described as non-criteria obstetric APS? How do you approach that group?" This question tells you a lot, fast. A specialist who works in this area will recognize the term immediately.
• "What are the risks of treating me as though I have it, if I don't?" You want them named out loud: bleeding, injection-site bruising, cost, delivery and epidural planning.
• "What's the risk of not treating, if I do have it?" Ask both. The asymmetry is usually the whole decision.
• "If you don't think this is indicated, would you refer me to a maternal-fetal medicine specialist or a rheumatologist who focuses on obstetric APS?"

The honest summary

There are two failure modes here and they're both real. One is a woman with meaningful antibodies who is told she doesn't qualify, does nothing differently, and loses another pregnancy. The other is a woman with no real antibody problem who spends a pregnancy injecting an anticoagulant she never needed, carrying its risks for nothing. Guidelines are written to prevent the second. They are less good at preventing the first.

If you want to find someone who works in this specific area, how to research a doctor covers reading their published research on PubMed — which, for a question this specialized, tells you more than any review site will.