Thyroid Disorders Diagnosis
Thyroid Disorders Diagnosis
Thyroid testing asks two different questions, and each is answered by a different investigation. Function — how much hormone the gland is delivering — is answered by a blood test in which TSH and free T4 are read together and move in opposite directions. Structure — what the gland looks like — is answered by an ultrasound. This page sets out what each test establishes, what it cannot show, and why one abnormal result is usually the beginning of the assessment rather than the end of it.
The blood test first: TSH, and why it moves before the hormone does
TSH is released by the pituitary and tells the thyroid to produce hormone. The pituitary responds to the hormone level rather than the other way round, so it raises TSH when the gland is underdelivering and suppresses it when the gland is overdelivering. That loop makes TSH the most sensitive single measure of thyroid function: it moves before the hormone level does, and it moves further.
In practice this means a gland that is beginning to fail shows a rising TSH while free T4 is still inside the reference range, and measuring the hormone alone at that point shows nothing at all. It also means a borderline result is read against the range printed on that report, since ranges differ slightly between laboratories.
Free T4 is then measured alongside it, and its job is confirmation. TSH says which way the control signal has moved; free T4 says whether the hormone level itself has moved. Where both have moved, the picture is established. Where only one has, the result is subclinical, which is a different question and is covered below.
Free T3 has a narrower role. It is added when overactivity is suspected and the free T4 is normal, or when the severity of an established overactive state needs describing. It is not a routine addition to every thyroid panel, and a free T3 measured on its own, without TSH, is close to uninterpretable.
Reading the pair, in which the two move in opposite directions
The single most useful thing to know about thyroid results is that TSH and free T4 move in opposite directions. When the gland is failing, the hormone level falls and the pituitary pushes harder, so TSH rises. When the gland is overproducing, the hormone level rises and the pituitary withdraws, so TSH falls. Every pattern in thyroid testing follows from that one fact.
| TSH | Free T4 | Reading |
|---|---|---|
| High | Low | An underactive gland, with the hormone level confirming it. The combination that usually leads to replacement |
| High | Normal | Subclinical underactivity: the hormone level is holding and the decision is not automatic |
| Low | High | An overactive gland, with the hormone level confirming it |
| Low | Normal | Subclinical overactivity: the degree, the age and the heart rhythm decide what follows |
| Normal | Normal | Function is normal. This says nothing at all about structure, which is a separate question |
The misreading that follows from the name is common enough to be worth stating directly: a high TSH does not mean a strong or overactive thyroid. It is the pituitary shouting at a gland that is not responding. People who read it the other way arrive expecting the opposite of what they are about to be told, and the appointment is spent correcting the assumption before anything useful can be discussed.
Subclinical: an abnormal TSH with a normal hormone level
Subclinical means the control signal has moved while the hormone level has not. It is a description of a result, not a diagnosis of a person, and it is the situation in which thyroid medicine is most often over-treated and under-explained.
Whether it calls for treatment rests on several things rather than on the direction of an arrow: how far the TSH has moved, whether antibodies are present, what symptoms there are, your age, whether a pregnancy is planned, and whether anything makes the person more sensitive to thyroid hormone.
The other half of the answer is that a single abnormal measurement is often transient. TSH moves with intercurrent illness, with a large iodine load from contrast agents or supplements, with pregnancy and the months after it, with several medicines, with the time of day the sample was drawn, and with nothing identifiable at all. Repeat testing on a separate occasion is therefore part of the assessment rather than a delay imposed on it.
What a reasonable sequence looks like: a repeat measurement with both tests after an interval, an antibody measurement, and an ultrasound if the gland is enlarged or a nodule has been felt. If the abnormality persists and the rest of the picture supports it, treatment becomes a decision made with the person rather than a reflex. Whether it applies in any individual case is a clinical judgement made with the full picture.
Antibodies — what they establish and what they do not
Two antibody measurements are used in routine practice. Antibodies against thyroid peroxidase indicate that the immune system is reacting against the gland; they are the marker of autoimmune thyroiditis, the commonest background for an underactive gland. Antibodies against the TSH receptor indicate a different process, in which the antibody stimulates the gland rather than damaging it.
What they establish is the mechanism. That has two practical uses: it explains why the gland has failed or is overworking, and it contributes to the judgement about subclinical results, because an abnormal TSH with antibodies present is more likely to progress than one without.
What they do not do is measure function. A person can carry antibodies at a high level for years with entirely normal thyroid function, and the presence of antibodies alone is not an indication for treatment. Nor is the level a measure of severity: antibody titres do not track with how unwell a person feels or with how much hormone needs replacing, and repeating them to monitor progress adds nothing.
Ultrasound, and where a nodule leads
Ultrasound is the structure axis, and nothing in the blood tests substitutes for it. It describes the size of the gland, whether it is diffusely enlarged, whether the texture is uniform, and whether any nodule is present. Where a nodule is found, the scan describes its size, whether it is solid or cystic, its echogenicity, the shape of its margins, and whether there are calcifications — features that a radiologist combines into a category.
That category, together with the size, is what decides whether a nodule is sampled. Sampling is done by fine-needle aspiration, a thin needle guided by the ultrasound, and it is decided by appearance rather than by the hormone results: a nodule in a gland with entirely normal function can still warrant sampling, and a nodule in an overactive gland usually does not need it if the scan pattern is characteristic.
What ultrasound is not is a screening test for everyone. It is indicated when something has been felt, when the gland is enlarged, when the neck has been irradiated, where there is a family history of thyroid cancer, and in some people with antibodies — because a scan performed without an indication frequently finds small nodules that would never have mattered.
The characterisation of a nodule, and the decision about sampling, belongs to the thyroid nodules guide and, where sampling establishes a diagnosis, to the thyroid cancer guide. This page hands over at that point rather than repeating it. Where sampling is indicated, our endocrinology department performs thyroid nodule aspiration to establish what a nodule is early rather than watching it indefinitely.
Pregnancy, iodine, medicines and illness — what changes the reading
Several situations change what a thyroid result means, and each of them is a reason for the result to be read by someone who knows the context rather than against a range printed on a page.
| Situation | What changes, and what follows |
|---|---|
| Pregnancy and the year after delivery | A pregnancy hormone stimulates the TSH receptor, so TSH is often lower in the first trimester and is read against trimester-specific ranges. Requirements for thyroid hormone often rise early in pregnancy, so the dose is reviewed as soon as pregnancy is confirmed and monitored more often through it. The months after delivery carry their own thyroiditis, which is easy to mistake for something else |
| Iodine, in either direction | Too little iodine and too much both disturb the gland. A large load — from contrast agents, from amiodarone, or from concentrated seaweed or kelp supplements — can push an otherwise normal gland in either direction, which is why the history includes supplements rather than only medicines |
| Biotin supplements | Biotin interferes with the assay many laboratories use and can produce a pattern that looks like overactivity. It is stopped for the period the laboratory specifies before the sample is drawn |
| Acute illness of any kind | Thyroid results taken during a significant illness are frequently misleading and usually recover with it. Testing is deferred where it can be, and a result drawn during an admission is interpreted with that in mind |
| Medicines that act on the thyroid | Amiodarone, lithium, interferon and several cancer treatments change the gland or the test. A medicine started before the symptoms appeared is the clue |
| Oestrogen and the pill | Oestrogen changes the proteins that carry thyroid hormone, which moves total hormone measurements while leaving free hormone — and therefore the person — unaffected. It is one of several reasons free rather than total hormone is measured |
What all six have in common is that the number is the same and the meaning is not: whether a sample was drawn during an illness, during a pregnancy, or while a supplement was being taken is part of the result rather than a footnote to it.