Your thyroid, your weight, and a warning that stops at borders

It does something. It does less than it is given credit for, and the part it does is not mostly what people assume.

Thyroid hormone sets the pace of resting metabolism, so when there is too little of it, resting energy expenditure falls. Weight goes up. But the increase is typically modest — a matter of a few kilograms rather than a transformation — and a substantial share of it is retained fluid and salt rather than fat tissue. Hypothyroidism causes the body to hold water in a way that is visible in the face and the ankles, and that is weight on a scale without being fat.

Which is why the weight that comes off after treatment starts also tends to be modest, and also tends to be largely fluid. People who begin levothyroxine expecting the diagnosis to explain everything and the treatment to reverse everything are usually disappointed twice.

None of that means the diagnosis does not matter. An untreated underactive thyroid is worth finding and worth treating for a long list of reasons that have nothing to do with weight — fatigue, cold intolerance, mood, cholesterol, heart rate, fertility. It means only that it is rarely the whole explanation, and that being told it is sets someone up to stop looking.

Subclinical hypothyroidism — a raised TSH with normal thyroid hormone levels — is a separate question and a genuinely unsettled one. Treating it does not reliably produce weight loss, and the decision to treat rests on symptoms, antibody status, the degree of TSH elevation and pregnancy plans rather than on the scale.

This section exists because thyroid hormone is sold on the grey market as a weight-loss product, frequently alongside stimulants, and because the warning against it is rarely quoted in the places where it would do the most good.

The UK label for levothyroxine is unusually direct. Thyroid hormones should not be given for weight reduction. In people whose thyroid function is normal, levothyroxine does not cause weight reduction. And substantial doses may cause serious or even life-threatening undesirable effects, particularly in combination with certain substances for weight reduction, and especially with sympathomimetic amines.

Read that last clause again, because it is not generic caution. Sympathomimetic amines are stimulants. The label is describing a specific combination — thyroid hormone plus a stimulant, for weight — and calling it potentially life-threatening. That combination is exactly what is sold in the parts of the internet that sell this sort of thing, and it has been for decades.

The mechanism of harm is not mysterious. Thyroid hormone in excess drives heart rate and cardiac workload, provokes atrial fibrillation, and accelerates bone loss. Add a stimulant and the cardiac effects compound. The weight that comes off under those conditions includes muscle, and the risk does not scale with how much weight anybody wanted to lose.

There is also a quieter interaction worth knowing if you take both kinds of medicine. Orlistat can reduce levothyroxine absorption, which may produce hypothyroidism in someone who was adequately treated before, so thyroid function should be monitored if the two are taken together. That is a monitoring point rather than a prohibition, and it is the sort of thing that gets missed when the two prescriptions come from different places.

This is the finding this post was written for, and it explains a great deal of confusion that patients are blamed for.

In the United States, the labels for GLP-1 receptor agonists licensed for weight management carry a boxed warning about thyroid C-cell tumours, and they contraindicate the medicines in anyone with a personal or family history of medullary thyroid carcinoma, or with Multiple Endocrine Neoplasia syndrome type 2. A boxed warning is the strongest warning that label format has.

In the United Kingdom and the European Union, they do not. The Summary of Product Characteristics for Wegovy lists exactly one contraindication: hypersensitivity to the active substance or the excipients. The Saxenda SmPC lists exactly one: hypersensitivity to liraglutide or the excipients. There is no boxed warning, because the format does not exist, and there is no contraindication for medullary thyroid carcinoma or MEN 2.

The underlying evidence is the same evidence in both places. Two-year carcinogenicity studies in rats and mice found non-lethal thyroid C-cell tumours, described in the UK labels as a class effect for GLP-1 receptor agonists. The labels explain the mechanism: non-genotoxic, specifically mediated through the GLP-1 receptor, in a way to which rodents are particularly sensitive. And they state the conclusion in careful words worth quoting precisely — the relevance for humans is considered to be low, but cannot be completely excluded.

So two regulators looked at one set of rodent data and translated it into two different instructions. That is not a scandal and neither is wrong; it is what happens when a finding of uncertain human relevance meets two different regulatory traditions. But it has a practical consequence that nobody mentions.

It means that most of what you will find online about thyroid history and these medicines was written for a different licence than the one that governs your prescription.

The volume of English-language health content is overwhelmingly American. So a reader in Lisbon or Lyon searching whether a family history of thyroid cancer rules them out will find, repeatedly and confidently, that it does — because in the jurisdiction those pages were written for, it does. Under their own regulator it is not a contraindication.

That cuts both ways and the symmetry matters. Nobody should read this and conclude that a thyroid history is irrelevant. A prescriber assessing you in Europe will still take a thyroid history, will still weigh a personal or family history of medullary thyroid carcinoma in the balance of a clinical decision, and may well decide against treatment on exactly that basis. The difference between a contraindication and a clinical judgement is not that one matters and the other does not. It is that one removes the decision and the other leaves it with a clinician who knows your history.

There is one more label difference that almost nobody notices, and it is inside the European labels rather than between continents. The liraglutide label advises caution in patients with pre-existing thyroid disease, because thyroid adverse events such as goitre were reported in diabetes trials particularly in that group. The semaglutide label does not carry that wording. Two medicines in the same class, two different cautions about the same organ.

The useful instruction from all of this is narrow and worth following: bring your thyroid history, and your most recent thyroid blood results, to the assessment. Not because the answer is predetermined in either direction, but because the person deciding should be the one who can see them.

If thyroid disease and weight are both in the picture, sequence matters more than most things in this post.

An untreated or inadequately treated underactive thyroid should be addressed first, and settled, before conclusions are drawn about anything else. Not because it will solve the weight — it usually will not, for the reasons in the first section — but because it changes the baseline against which everything is measured. Starting a weight treatment while thyroid replacement is still being titrated means two moving variables and no way to attribute what happens to either.

And if nobody has ever checked, that is worth asking for. Thyroid function is a routine blood test. The point is not that it is likely to be the answer. The point is that it is cheap to rule in or out, and it is the sort of thing that goes unchecked for years while someone is told to try harder.