Thyroid and Fertility: Which Finding Actually Changes Treatment

It depends which finding it is, and the four that get grouped together behave very differently. Where the thyroid is genuinely underactive, that is a diagnosis in its own right and is managed on its own account. Where thyroid function is normal and only antibodies were found, the two largest randomised trials found that treating made no difference to live birth.

That second result is the one that tends not to reach the consulting room. TABLET randomised 952 women who had thyroid peroxidase antibodies, normal thyroid function, and a history of miscarriage or infertility, to levothyroxine or placebo from before conception through pregnancy. Live birth after at least 34 weeks occurred in 37.4% of 470 women given levothyroxine and 37.9% of 470 given placebo — a relative risk of 0.97, and no difference in pregnancy loss, preterm delivery or the newborn outcomes either.

So the useful question is not whether your thyroid is "optimal". It is which of these four your report actually describes.

These are not degrees of the same problem. They are different findings with different evidence behind them:

• Overt hypothyroidism — the thyroid is not keeping up: TSH is raised and free T4 is low. This is unambiguous, and it is treated as a medical condition rather than as a fertility intervention. None of the trials below apply to it. The Cochrane review of thyroxine in subfertile women specifically excluded women with previously known clinical hypothyroidism or already taking thyroxine.
• Subclinical hypothyroidism — TSH above the reference range with a normal free T4. This is a laboratory description, and whether treating it helps fertility is genuinely unsettled rather than merely debated. Which reference range your clinic applies decides who gets this label at all.
• Antibodies with normal thyroid function — thyroid peroxidase antibodies present, TSH and free T4 normal. This is the group TABLET and POSTAL studied, and the group ESHRE recommends against treating.
• Hyperthyroidism — an overactive thyroid. A different condition with its own management, addressed separately in the American Thyroid Association's 2026 guidelines, which cover hyperthyroidism and Graves' disease in preconception and pregnancy alongside hypothyroidism.

Reading one number in isolation is how these four get confused. Our separate page on what a TSH result actually means goes through the threshold question in detail, including where the widely quoted 2.5 mIU/L figure came from and why it moved.

Under NICE's fertility guideline NG257, not necessarily. Recommendation 1.18.12 is that thyroid function testing should be offered to women with possible fertility problems only if they have symptoms of thyroid disease. A routine TSH on every woman who walks into a fertility clinic is not what that recommendation describes.

This matters because a test you did not need can still produce a borderline number, and a borderline number is hard to un-know. Once it exists it tends to get treated, and the evidence that treating it helps is the weakest part of this whole area.

None of which means a thyroid result should be ignored. If you have symptoms, a family history of thyroid disease, or a previous abnormal result, testing is reasonable and repeating a borderline value before acting on it is more reasonable still.

On the best available evidence, no. Two large randomised trials asked this in different populations and agreed.

TABLET, above, found no difference in live birth. POSTAL asked the same question inside IVF: 600 women at a single Beijing centre with normal thyroid function and positive antithyroperoxidase antibodies were randomised to levothyroxine or no levothyroxine alongside identical IVF and embryo transfer protocols. Miscarriage was 10.3% against 10.6%, clinical intrauterine pregnancy 35.7% against 37.7%, and live birth 31.7% against 32.3%.

The guidelines have followed the trials. ESHRE's 2022 update to its recurrent pregnancy loss guideline states that euthyroid women with thyroid antibodies and recurrent pregnancy loss should not be treated with levothyroxine, and grades that as a strong recommendation.

A positive antibody result is still worth having on record, because it raises the chance of thyroid function changing later and is a reason to keep an eye on it. It is not, on this evidence, a reason to start treatment while function is normal. Thyroid autoimmunity sits alongside other immune findings in fertility care, which we cover in what is actually proven about autoimmune disease and fertility.

This is the honest gap, and it is worth naming rather than papering over. The Cochrane review found a signal in favour of treatment here — live birth relative risk 2.13 (95% CI 1.07-4.21) — but that figure comes from a single randomised trial of 64 women, was graded low quality, and its confidence interval is wide enough to be compatible with a small benefit or a very large one. The reviewers' overall verdict was that no clear conclusions could be drawn.

NICE reached a similar place from the other direction. For subclinical hypothyroidism in people having fertility treatment, the NG257 committee reviewed the evidence and issued a recommendation for research — that is, it asked for a trial rather than recommending the treatment.

What follows practically is modest but real. If you have been given this label, the two questions worth asking are which reference range produced it and whether the result has been repeated, because both can change the answer before any treatment decision is reached.

No. Nothing on this page is a reason to stop a prescribed medicine, and stopping thyroid hormone on your own is not a neutral act — if you are one of the people who genuinely needs it, coming off it causes harm. This page is about which question your result answered, not about overriding the person who examined you.

What it does give you is a better conversation. Reasonable things to ask: which of the four findings above was mine, what were the TSH and free T4 values and the range they were read against, was the result repeated, and was the decision based on thyroid function or on antibodies alone. If the answer is antibodies with normal function, the trial evidence above is a fair thing to raise.

Your prescriber may still have a reason particular to you — a previous thyroid condition, a borderline free T4, a symptom pattern. That is exactly why the question belongs with them rather than with an article.

Not on the basis of living in India. The India Iodine Survey 2018-19 sampled 21,406 households across every state and union territory and measured salt iodine by titration. Median urinary iodine concentration was 173.4 ug/L in pregnant women, 172.8 ug/L in lactating women and 178.0 ug/L in non-pregnant, non-lactating women — all within the adequate range by WHO criteria.

Coverage is not uniform. Of 21,406 households sampled, those holding salt with adequate iodine, meaning at least 15 parts per million, came to 76.3% nationally, with 10 states and 3 union territories reaching universal salt iodisation and 11 states and 2 union territories below the national average. So this is a question with a local answer rather than a national one, and it belongs with a clinician who knows where you live.

The general point stands against the common advice to load up on iodine for fertility: at a population level, iodine nutrition in Indian women of reproductive age measured as adequate, and iodine is a nutrient where more is not automatically better.

Being specific about the gaps is more useful than a summary:

• That treating antibodies improves fertility. TABLET and POSTAL both looked and found no effect on live birth. What the trialists suggest instead is that the risk in these women may sit in the immune environment rather than in a shortfall of thyroid hormone.
• That treating subclinical hypothyroidism improves live birth. The only randomised signal comes from one trial of 64 women, graded low quality, and NICE asked for research rather than recommending treatment.
• That there is an agreed "fertility TSH target". This page deliberately quotes no threshold; where the commonly cited figures come from is set out on our TSH page, and the current edition of the American Thyroid Association guidance is the 2026 one.
• That selenium, iodine loading or dietary changes improve fertility outcomes in thyroid autoimmunity. No trial retrieved for this page tested that, so it is unsupported here rather than disproved.
• That any of this was measured in an Indian fertility population. TABLET ran in UK hospitals and POSTAL in a single Beijing centre. The biology is not expected to differ; the direct evidence is simply not Indian.

One more limit, acknowledged by the guideline writers themselves: the ATA's 2026 guidelines state that much of the underlying evidence in this field remains of low-to-moderate quality. Confident advice about thyroid and fertility is running ahead of what has been measured.