Key takeaways
- T4 is a stable prohormone reservoir; T3 is the hormone that acts at the receptor.
- Free measurements matter because bound hormone is inert and binding proteins move for reasons unrelated to the thyroid.
- Conversion is controlled locally by deiodinases, so serum free T3 is an average across a system that does not behave like an average.
- TSH reports on the pituitary's own supply, which leaves it blind to conversion problems and to central hypothyroidism.
- A low free T3 with a normal TSH is a question about energy intake, illness, inflammation and cofactors — not a diagnosis.
The thyroid gland barely makes the hormone that does the work. It produces mostly T4, which is close to inactive, and the body then converts that T4 into T3 in the tissues that need it — at rates each tissue sets for itself. That single fact is why a thyroid panel is not one measurement but two, why a normal TSH does not settle the question, and why the most common pattern in symptomatic people is one that a standard screening panel cannot see.
Free versus total, and why it matters first
Almost all circulating thyroid hormone is bound to carrier proteins — thyroxine-binding globulin mainly, plus transthyretin and albumin. Bound hormone cannot enter cells or act on receptors. It is a reservoir.
Total T4 and total T3 measure bound plus unbound together, so they move whenever the carrier proteins move — and those move for reasons unrelated to the thyroid. Oestrogen raises binding globulin, which is why pregnancy and oral oestrogen raise total T4 in women with entirely normal thyroid function. Androgens, significant illness and some medications lower it. Free T4 and free T3 measure the unbound, biologically available fraction, which is why "free" is the measurement worth ordering — and why a total T4 in isolation misleads in exactly the populations most likely to be tested.
T4 is a prohormone; T3 is the hormone
T3 binds the nuclear thyroid hormone receptor with far higher affinity than T4 and is conventionally described as around four times more potent. T4's contribution at the receptor is small; its real function is to circulate as a stable, long-lived reservoir — half-life in days, against less than one for T3 — from which tissues generate active hormone on demand. The thyroid does not deliver a finished product. It delivers raw material, and the conversion step is where much of the regulation happens.
The conversion machinery, and why it is local
Conversion is done by three selenium-containing deiodinases, doing different jobs in different places (Bianco et al., Endocr Rev 2002).
- D1, mainly in liver and kidney, generates a large share of the T3 that appears in the circulation.
- D2, in brain, pituitary, skeletal muscle and brown fat, generates T3 inside the cell for that cell's own use. Its output is not necessarily reflected in a blood test at all.
- D3 is the off switch. It converts T4 into reverse T3 — an inactive isomer — and T3 into an inactive metabolite, reducing thyroid signalling locally.
The design implication matters: thyroid status is not one number for the whole body. Each tissue sets its own D2 and D3 activity, so the brain can be adequately supplied while muscle is not. Serum free T3 is an average across a system that does not behave like an average.
Several things reliably shift the balance toward D3: significant illness or surgery, sustained caloric restriction, inflammation, high cortisol, impaired liver or kidney function, and deficiency in the cofactors these enzymes depend on — selenium in particular, since the deiodinases are selenoproteins (Köhrle, Int J Mol Sci 2023). Iron and zinc matter too, which is worth knowing before chasing an exotic explanation.
Why TSH does not close the question
TSH tests one thing well: whether the pituitary considers itself adequately supplied. It is treated as the master test because in classic primary hypothyroidism it is exquisitely sensitive, rising well before free T4 leaves the reference range.
The limitation follows from the deiodinase biology. The pituitary is unusually rich in D2 and makes its own T3 locally from T4, so it reports on its own intracellular status — which tracks free T4 more closely than the T3 available to muscle, liver or brain. A comfortable pituitary does not guarantee comfortable peripheral tissue.
This is not fringe. It explains a documented observation in treated hypothyroidism: patients on T4-only replacement with a normal TSH frequently have lower free T3 and higher free T4 than before treatment, because one circulating source cannot reproduce a gland secreting both (Gullo et al., PLoS One 2011). It does not mean TSH is useless, or that every symptom is hidden thyroid disease — see the TSH range debate.
Reverse T3, honestly
When D3 activity rises, more T4 is diverted to reverse T3, which does nothing. Low free T3 with high reverse T3 and a normal or low-normal TSH is well described in serious illness and termed non-thyroidal illness, or low-T3, syndrome (Savvidis et al., World J Crit Care Med 2025). In critical illness it looks like a protective adaptation, and treating it with thyroid hormone has not improved outcomes. In ambulatory people the assay is poorly standardised with no treatment threshold. It is a clue, not a diagnosis and not a target (reverse T3 and chronic stress).
Reading the panel
| Pattern | Interpretation | What it points to |
|---|---|---|
| TSH high, free T4 low, free T3 low | Primary hypothyroidism | Check antibodies; this is the straightforward case |
| TSH high, free T4 normal | Subclinical hypothyroidism | Repeat before acting; antibodies and symptoms guide the decision |
| TSH normal, free T4 normal, free T3 low | Reduced conversion | Look at illness, dieting, inflammation, cortisol, selenium, iron |
| TSH normal, free T4 high-normal, free T3 low, on T4 therapy | Conversion is the limiting step on replacement | A prescribing conversation, not a supplement one |
| TSH low, free T4 high, free T3 high | Hyperthyroidism | Requires prompt evaluation |
| TSH low, free T4 low, free T3 low | Central (pituitary) hypothyroidism | Rare, and easily missed by TSH-first screening |
That last row is the genuine blind spot in TSH-only screening: in central hypothyroidism the pituitary is the problem, so its output cannot report on itself. Free T4 catches it; TSH alone does not (the HPT axis in depth).
Ranges, and what "optimal" can and cannot mean
Commonly used targets are free T4 around 1.2-1.6 ng/dL against a reference of roughly 0.8-1.8, and free T3 around 3.2-4.2 pg/mL against a reference of roughly 2.3-4.2. Reference intervals differ between assays, so these are only interpretable next to the range printed on your own report.
One caution. Reference ranges are population-derived and an individual's set point is narrower — the legitimate basis for "in range but wrong for you". That argument is often stretched into treating any low-normal free T3 as a disease. Most low free T3 in otherwise well people is a downstream marker of dieting, over-training, poor sleep or inflammation, and corrects when those resolve (optimal versus normal ranges).
The clinical pearl: a low free T3 with a normal TSH is a finding, not a diagnosis. It says conversion is being restrained, and the useful next question is always why. In an ambulatory adult the answer is far more often a prolonged calorie deficit, chronic under-sleeping, inflammation or an iron or selenium shortfall than a primary thyroid problem — and treating the number without answering the question is how people end up on thyroid hormone they did not need.
What to actually do with the result
Order the full picture at once rather than in stages: TSH, free T4, free T3 and thyroid antibodies. Antibodies change the meaning of a borderline result entirely — a raised TSH with positive TPO antibodies is a trajectory, the same TSH without them may be nothing (thyroid antibodies). Draw in the morning, and before the day's dose if already on replacement. Repeat anything abnormal before acting on it; single values have started a great many unnecessary treatments.
If free T3 is low with a normal TSH and free T4, work the upstream list first: energy intake, training load, sleep, alcohol, inflammatory markers, ferritin and selenium (selenium and the thyroid). Correct what is correctable and re-test after a couple of months.
And leave the treatment decision where it belongs. Whether replacement is indicated, and whether combination therapy has a role, is a prescribing judgement — guidance still positions T4 monotherapy as standard and treats the combination question as unsettled (Jonklaas et al., Thyroid 2014).
Bottom line
The thyroid secretes mostly T4, a stable prohormone, and tissues generate active T3 themselves using deiodinases they control locally. Free measurements matter because bound hormone is inert and binding proteins move for unrelated reasons. TSH reports on the pituitary's own supply, which is why it is excellent for classic primary hypothyroidism and blind to conversion problems and central disease. Order TSH, free T4, free T3 and antibodies together, repeat before acting, and treat a low free T3 with a normal TSH as a question rather than a diagnosis. Interpretation belongs with a physician — the 60-second assessment is where that starts.
Educational content, not medical advice. Laboratory interpretation and any treatment decision are made by a licensed physician after individual evaluation. Individual results vary.
