Thyroid Panel Explained: TSH, T4, T3, and What AI Looks For

How the thyroid axis actually works

You cannot read a thyroid panel well without a mental model of the feedback loop that produces the numbers. The thyroid is a butterfly-shaped gland in the front of the neck, but it does not decide on its own how much hormone to release. It sits at the bottom of a three-tier control system called the hypothalamic-pituitary-thyroid (HPT) axis. The hypothalamus releases thyrotropin-releasing hormone (TRH); TRH tells the anterior pituitary to release thyroid-stimulating hormone (TSH); and TSH tells the thyroid gland to make and release thyroxine (T4) and a smaller amount of triiodothyronine (T3). When circulating thyroid hormone rises, it feeds back on the pituitary and hypothalamus to switch the signal off. When it falls, the brakes come off and TSH climbs. This is a classic negative-feedback loop, and almost every pattern you will ever see on a panel is that loop caught mid-adjustment.

The single most important consequence of this design is leverage. The relationship between TSH and free thyroid hormone is not linear — it is roughly log-linear. A small drop in Free T4 that stays inside the "normal" range can drive TSH to double or triple. That amplification is why TSH, a pituitary hormone, is the most sensitive early warning for a thyroid gland that is starting to fail. It also explains a pattern that frustrates patients constantly: your TSH can be clearly abnormal while your T4 still reads normal. Nothing is contradictory there — the pituitary is simply shouting louder to keep a struggling gland producing enough hormone.

The gland needs raw material to do its job: iodine. The World Health Organization identifies iodine deficiency as the world's leading preventable cause of thyroid dysfunction and cognitive impairment, which is why salt iodization programs exist globally. In iodine-sufficient regions such as most of North America and much of Europe, the leading cause of an underactive thyroid is not iodine at all — it is autoimmune destruction of the gland (Hashimoto's thyroiditis). Knowing which world you live in changes how a high TSH should be read, and it is one of the contextual clues a good analyzer weighs.

TSH: the master signal

TSH (thyroid-stimulating hormone, also called thyrotropin) is made by the pituitary and tells the thyroid how much hormone to produce. It is the most sensitive single test for thyroid dysfunction because small changes in thyroid output produce large changes in TSH. The typical adult reference range is roughly 0.4 – 4.0 mIU/L, with tighter ranges in pregnancy and a well-documented upward drift with age — people over 80 commonly sit at the high end or slightly above it without any disease. You can read the marker-level detail on the TSH biomarker page, but the interpretation always depends on what the free hormones are doing alongside it.

Because of the log-linear leverage described above, TSH is best thought of as a hypersensitive smoke alarm rather than a thermometer. It tells you loudly that something is off before the fire is visible, but it does not tell you how big the fire is. That is why the American Thyroid Association and the Endocrine Society both position TSH as the recommended first-line screen, then use the free hormones to grade severity and localize the problem. A single TSH also has real biological noise: it follows a circadian rhythm, peaking overnight and in the early morning and dipping in the afternoon, and the pituitary secretes it in pulses. Within one healthy person, repeat measurements can vary by up to roughly 50%. A borderline TSH should almost always be repeated before anyone acts on it.

Direction matters more than most patients realize. A high TSH points to primary hypothyroidism — the pituitary is compensating for a gland that is not keeping up. A low or suppressed TSH points to hyperthyroidism — there is so much thyroid hormone that the pituitary has stopped signaling. The important exceptions are central (pituitary) disease, where TSH can be inappropriately normal or low despite a genuinely low T4, and acute illness, where TSH can transiently drop and then rebound. Those exceptions are exactly where reflexive "high equals hypo, low equals hyper" thinking goes wrong, and where reading TSH in isolation is dangerous.

Free T4 and Free T3

T4 (thyroxine) is the storage and transport hormone; T3 (triiodothyronine) is the active form that does the work at the tissue level. The thyroid secretes mostly T4, and peripheral tissues then convert T4 to T3 using deiodinase enzymes according to local demand. This means T3 is partly under the control of your organs, not just your gland — an elegant system, but one that complicates interpretation. The word "free" matters: more than 99% of thyroid hormone in blood is bound to carrier proteins (thyroxine-binding globulin, transthyretin, albumin) and is metabolically inert. Only the tiny unbound fraction is biologically active, so Free T4 and Free T3 are far more reliable than the older total T4/T3 tests, which move whenever binding proteins change — in pregnancy, on estrogen, in liver disease.

Most clinicians check Free T4 routinely alongside TSH and reflex to Free T3 when TSH is suppressed or the picture is unclear. Typical adult ranges are approximately Free T4 0.8 – 1.8 ng/dL and Free T3 2.3 – 4.2 pg/mL, though the exact interval is assay-specific and should always be read against the range printed on your own report. Free T4 is the workhorse for grading severity and for spotting central disease. Free T3 earns its keep in one specific scenario: T3-toxicosis, an early form of hyperthyroidism (often Graves' disease) in which Free T3 rises before Free T4 does. If you order only Free T4, you can miss it.

A crucial caution about Free T3: it is the least useful number in a sick or hospitalized patient. Any significant systemic illness suppresses the T4-to-T3 conversion as an adaptive, energy-sparing response, dropping Free T3 first. A low Free T3 in someone with the flu, a recent surgery, or a serious infection usually reflects the illness, not the thyroid. This is why endocrinology guidelines discourage routine thyroid testing during acute illness unless thyroid disease is genuinely suspected.

Reverse T3 and non-thyroidal illness

When the body converts T4, it can go down one of two paths: to active T3, or to an inactive mirror-image molecule called reverse T3 (rT3). Reverse T3 occupies the receptor without switching it on — it is a way to dial thyroid activity down without touching the gland. In serious illness, starvation, trauma, or after major surgery, the body deliberately shunts T4 toward reverse T3 and away from active T3. The result is a distinctive laboratory picture called non-thyroidal illness syndrome (NTIS), or euthyroid sick syndrome: low Free T3, normal or low Free T4, a TSH that is usually normal or low, and a raised reverse T3.

The clinically important point, endorsed by the American Thyroid Association and the Endocrine Society, is that NTIS is generally a physiological adaptation, not a thyroid disease, and treating it with thyroid hormone has not been shown to help and may harm. Reverse T3 testing is therefore of limited value in routine screening and is not recommended by mainstream guidelines for diagnosing hypothyroidism, despite its popularity in some wellness circles. Where a reverse T3 result is genuinely useful is in helping distinguish the temporary NTIS pattern of a sick patient from true central hypothyroidism — a nuance a careful analyzer will flag rather than over-interpret.

Reference ranges at a glance

The table below gives typical adult reference ranges. Two caveats before you use them. First, reference intervals are assay-dependent: a Free T4 of 1.0 ng/dL is normal on one analyzer and borderline on another, which is why international bodies such as the IFCC (International Federation of Clinical Chemistry) run programs to harmonize thyroid assays, and why age- and sex-specific pediatric intervals from projects like CALIPER, and adult Nordic intervals from NORIP, exist. Second, a range describes the middle 95% of a reference population — it is a statistical fence, not a personal target. Always compare against the range on your own lab report.

The patterns to know

Almost all thyroid interpretation collapses into a handful of recognizable patterns. Learn these and you can read most panels at a glance. The core idea is to always look at TSH and Free T4 together, add Free T3 when TSH is suppressed, and layer antibodies on top when autoimmunity is in question.

Two subtleties separate a novice reading from an expert one. First, a normal Free T4 with a high TSH is not the same clinical entity as a low Free T4 with a high TSH — the first is subclinical, the second is overt, and they carry different treatment thresholds. Second, an inappropriately low or normal TSH paired with a genuinely low Free T4 is a red flag for central (pituitary) disease and must never be filed under "subclinical hypothyroidism." An AI lab test analyzer flags this discordant combination rather than mislabeling it, because the workup and urgency are entirely different.

Hypothyroidism: high TSH in depth

Hypothyroidism means the thyroid is not producing enough hormone, and the pituitary responds by driving TSH up. Overt disease shows a high TSH with a low Free T4; symptoms are the familiar slowdown — fatigue, cold intolerance, weight gain, constipation, dry skin, hair thinning, low mood, heavy or irregular periods, and slowed thinking. Because these symptoms are common and nonspecific, laboratory confirmation matters: you cannot diagnose an underactive thyroid on symptoms alone, and equally, a normal panel is strong evidence that vague fatigue is coming from somewhere else.

The mechanism behind a high TSH varies, and the cause changes both prognosis and monitoring. In iodine-sufficient countries the dominant cause is Hashimoto's thyroiditis, an autoimmune attack on the gland; worldwide, iodine deficiency remains the leading cause. Other important causes include prior thyroid surgery, radioactive iodine treatment, external-beam radiation to the neck, and a long list of drugs. The table below groups the common culprits.

Treatment of overt hypothyroidism is levothyroxine (synthetic T4), typically dosed near 1.6 micrograms per kilogram of body weight per day in otherwise healthy adults, with lower, cautiously titrated starting doses in older people and those with heart disease. The dose is then adjusted to bring TSH back into range, rechecked no sooner than about six weeks after any change because the axis takes that long to re-equilibrate. This is educational context, not a prescription — dosing decisions belong to your clinician, who will individualize them to your age, heart, pregnancy status, and symptoms.

Hyperthyroidism: low TSH in depth

Hyperthyroidism (thyrotoxicosis) is the mirror image: too much thyroid hormone, so the pituitary shuts TSH down toward or below the lower limit. Overt disease shows a suppressed TSH with a high Free T4 and/or Free T3. Symptoms are the body running hot — palpitations, a fast or irregular heartbeat, weight loss despite a good appetite, tremor, heat intolerance, sweating, anxiety, insomnia, frequent bowel movements, and in Graves' disease specifically, eye changes (a stare, bulging, grittiness) and sometimes a goiter. In older adults the presentation can be deceptively quiet — sometimes the only sign is new atrial fibrillation, which is why the American Heart Association and ACC flag thyroid testing in the workup of new-onset AF.

The most common cause of hyperthyroidism is Graves' disease, an autoimmune condition in which stimulating antibodies (TRAb/TSI) mimic TSH and drive the gland relentlessly. Other causes include toxic multinodular goiter, a single hyperfunctioning ("hot") nodule, the thyrotoxic phase of thyroiditis, and excess thyroid hormone intake. The distinction matters enormously because the treatments diverge: Graves' and toxic nodules are ongoing overproduction that may need antithyroid drugs, radioactive iodine, or surgery, whereas the thyrotoxic phase of thyroiditis is a self-limited leak of stored hormone that usually needs only symptom control and time.

Because the underlying cause dictates management, a suppressed TSH is a starting point, not an endpoint. Clinicians typically confirm with free hormones, then use TRAb/TSI antibodies and, where needed, a radioactive-iodine uptake scan to separate overproduction from a leak. Untreated hyperthyroidism is not benign: it accelerates bone loss and raises the risk of atrial fibrillation and, in severe cases, a life-threatening thyroid storm. Persistent palpitations with a suppressed TSH deserve prompt medical attention, not a wait-and-see approach.

The subclinical gray zone

"Subclinical" is the word that generates the most confusion and the most unnecessary anxiety. It means the free hormones are still normal while TSH sits just outside its range — mildly high in subclinical hypothyroidism, mildly low in subclinical hyperthyroidism. The gland is compensating; the body is, for now, still getting roughly the right amount of hormone. Many people in this zone feel completely well, and a meaningful fraction of borderline TSH results simply normalize on a repeat test weeks later, which is why guidelines from the American Thyroid Association and the Endocrine Society advise confirming a borderline TSH before labeling anyone.

Whether subclinical hypothyroidism should be treated is one of the genuinely nuanced questions in endocrinology, and it depends on how high the TSH is, the person's age, their symptoms, their antibody status, and whether they are pregnant or trying to conceive. Broad strokes below — but note that this is exactly the kind of multi-factor judgment that belongs to a clinician, not to a single number or an app.

Subclinical hyperthyroidism (a low TSH with normal free hormones) deserves its own respect, especially in adults over 65, because even mild, persistent TSH suppression is associated with a higher risk of atrial fibrillation and reduced bone density. Persistent — not one-off — suppression is what matters, and confirming it on a repeat test before acting is again the rule.

Antibodies and autoimmunity

Antibodies are what turn a functional pattern into a diagnosis with a name and a trajectory. TPO antibodies (against thyroid peroxidase) and thyroglobulin (Tg) antibodies are the markers of autoimmune thyroid disease. Strongly positive TPO antibodies in someone with subclinical hypothyroidism predict progression to overt disease at a rate of several percent per year, which is why their presence lowers the monitoring threshold and, for some patients, the treatment threshold. TRAb and TSI are a different beast: these are stimulating antibodies that cause Graves' disease, and they are the confirmatory test for hyperthyroidism of autoimmune origin, useful both for diagnosis and for predicting relapse or fetal risk in pregnancy.

Two honest caveats keep antibody results in perspective. First, a meaningful minority of healthy people — more women than men, and more with age — carry low-level TPO antibodies and never develop disease, so a positive antibody is a risk marker, not a verdict. Second, autoimmune thyroid disease clusters with other autoimmune conditions (type 1 diabetes, celiac disease, pernicious anemia, vitiligo), so a positive thyroid antibody is a nudge to keep an eye on the wider picture. Hashimoto's and Graves' can even coexist or convert over time, which is why antibodies are read alongside the functional pattern, never on their own.

Central and rare patterns

Most thyroid disease is primary — the problem is in the gland. But a small and important subset is central, meaning the problem is upstream in the pituitary or hypothalamus. In central hypothyroidism, the failing pituitary cannot mount a proper TSH response, so you see a genuinely low Free T4 with a TSH that is low or, misleadingly, still within the normal range. Anyone who anchored on TSH alone would call this normal and miss a serious diagnosis. Central hypothyroidism can be the first clue to a pituitary tumor, prior head injury or radiation, or postpartum pituitary damage, and it usually comes bundled with deficiencies in other pituitary hormones — so it changes the whole workup.

Rarer still are patterns that break the usual rules entirely: a TSH-secreting pituitary adenoma (high TSH and high free hormones), resistance to thyroid hormone (high free hormones with a non-suppressed TSH), and assay artifacts from interfering antibodies that produce numbers no physiology can explain. The common thread is discordance — TSH and free hormones pointing in directions that should not coexist. When a result is internally contradictory, the right move is not to force it into a named pattern but to repeat it, question the assay, and refer. This is precisely where a rules-based analyzer earns its place: it can hold up its hands and say "this combination doesn't fit a standard pattern; here is why it needs a specialist," rather than manufacturing false confidence.

Pregnancy and trimester ranges

Pregnancy rewrites the thyroid rulebook. In the first trimester, the pregnancy hormone hCG cross-reacts with the TSH receptor and mildly stimulates the thyroid, which physiologically lowers TSH; rising estrogen also increases binding proteins, shifting total hormone levels. As a result, non-pregnant reference ranges do not apply. The American Thyroid Association's 2011 guidance popularized a first-trimester upper TSH limit around 2.5 mIU/L, but the 2017 update walked that back: it recommends using population- and trimester-specific reference ranges wherever they exist, and where they do not, using an upper reference limit of roughly 4.0 mIU/L rather than the stricter 2.5. The net effect is that TSH targets are lowest early and drift back toward non-pregnant values by the third trimester.

The stakes are high, which is why the caution is warranted. Untreated overt hypothyroidism in pregnancy is associated with miscarriage, pre-eclampsia, preterm birth, low birth weight, and impaired neurodevelopment in the child — the fetus depends on maternal thyroid hormone, especially before its own gland is working. Levothyroxine requirements typically rise during pregnancy, and thyroid status is monitored closely throughout. TPO-antibody-positive women are watched even more carefully because they are prone to postpartum thyroiditis in the months after delivery. blood-test.life applies trimester-specific reference ranges automatically when pregnancy is indicated at upload, so a TSH that would read as "normal" on a standard adult range is correctly reframed against the pregnancy-appropriate one.

Who is at risk and who should test

Thyroid disease is common — hypothyroidism affects a few percent of the general population and rises with age — and it is markedly more common in women, in whom autoimmune thyroid disease predominates. Risk is higher with a family history of thyroid or autoimmune disease, after any radiation or surgery to the neck, in people with other autoimmune conditions, during the postpartum year, and with certain medications (lithium, amiodarone, immune checkpoint inhibitors). Age itself shifts the picture: TSH reference ranges legitimately rise in older adults, so a mildly high TSH in an 80-year-old is often normal aging, not disease.

On population-wide screening of people with no symptoms and no risk factors, the guidance is genuinely split, and honesty about that is part of good evidence. The U.S. Preventive Services Task Force (USPSTF) concluded there is insufficient evidence to recommend for or against routine screening of asymptomatic, non-pregnant adults — an "I" statement. Other bodies, including some endocrine groups, favor case-finding in higher-risk groups such as older women. What is not in dispute is that anyone with suggestive symptoms, a relevant medication, pregnancy, or a strong family history has a clear reason to test. If you are building a broader picture of your health, TSH sits naturally alongside the rest of a standard panel — you can see how the pieces connect on our how-it-works page and browse individual markers in the biomarker library.

How thyroid tests are measured

TSH and free hormones are measured on automated immunoassay analyzers, most commonly using chemiluminescent or electrochemiluminescent detection. Modern TSH assays are described by "generation": today's third-generation assays can reliably detect very low concentrations, which is what makes them good enough to distinguish subclinical from overt hyperthyroidism. Free T4 and Free T3 are trickier to measure because the free fraction is minuscule and can be disturbed during the assay itself; the reference method is equilibrium dialysis, but most labs use faster automated estimates that are excellent for routine care while carrying more uncertainty at the extremes and in altered-binding states like pregnancy or critical illness.

This is why reference intervals are assay-specific and why standardization efforts exist. The IFCC coordinates programs to harmonize thyroid measurements across manufacturers so that a result means the same thing wherever it is drawn. Reference-interval projects such as CALIPER (which established robust pediatric and adolescent intervals) and NORIP (Nordic adult intervals) supply the age- and sex-adjusted ranges that make interpretation accurate at the edges of life. The practical takeaway for you as a patient: never compare a result from one lab against a range printed by a different lab, and be cautious about tracking a trend across labs that use different platforms.

What skews your results

A surprising number of everyday factors can move thyroid numbers without any change in the thyroid itself. The most notorious is biotin (vitamin B7), sold in high doses for hair and nails. Because many immunoassays use biotin chemistry, high supplemental intake can produce a falsely low TSH together with falsely high Free T4 and Free T3 — a picture that mimics Graves' disease and has led to misdiagnosis. The U.S. FDA issued a formal safety communication about exactly this. The fix is simple: stop biotin for a few days before testing, and tell your lab and clinician if you take it.

• Time of day: TSH peaks overnight and early morning and dips in the afternoon; standardize your draw time when tracking a trend.
• Acute illness and recovery: can transiently lower TSH and free hormones (non-thyroidal illness), then rebound — avoid testing during acute illness unless necessary.
• Medications: amiodarone, lithium, corticosteroids, dopamine, estrogen, and immune checkpoint inhibitors each shift results by distinct mechanisms.
• Levothyroxine timing: taking your dose right before a draw can transiently raise Free T4; consistency matters more than any single value.
• Heterophile antibodies: rare interfering antibodies can produce impossible, discordant results that no physiology explains.
• Supplements and diet: very high or very low iodine intake, and biotin, are the common offenders.

The unifying principle is that a single surprising value should be questioned before it is believed. If your panel does not fit your clinical picture — you feel completely well but the numbers scream Graves' — the most likely explanation is interference or timing, not a dramatic hidden disease. Repeating the test, ideally under standardized conditions, resolves the great majority of these puzzles.

Related biomarkers and downstream effects

The thyroid does not act in isolation, and its state ripples across the rest of your panel — which is one reason interpreting thyroid results in a vacuum misses so much. Hypothyroidism slows metabolism broadly: it commonly raises LDL cholesterol and total cholesterol, so an unexplained rise in lipids is a recognized prompt to check TSH before escalating cholesterol treatment. It can also mildly raise creatine kinase, cause a low-grade anemia, and lift prolactin. Hyperthyroidism does the opposite to lipids (often lowering cholesterol) and speeds heart rate and bone turnover. If your lipids look off, our lipid panel guide explains how thyroid status factors into the interpretation.

The overlap runs the other way too, because thyroid symptoms are notoriously nonspecific and share territory with several correctable deficiencies. Fatigue, hair changes, and brain fog are just as easily driven by iron deficiency or low vitamin D as by an underactive thyroid, and the conditions frequently coexist — autoimmune thyroid disease clusters with pernicious anemia (low B12) and celiac disease (which impairs iron absorption). A thoughtful workup of "thyroid symptoms" therefore often includes ferritin, B12, and vitamin D. If your tiredness is being blamed on your thyroid but the panel is normal, it is worth reading our guides on iron deficiency anemia and vitamin D deficiency to make sure the real culprit is not being missed.

How AI reads a thyroid panel in context

The value of an AI lab test analyzer on a thyroid panel is not that it knows a single reference range — any table can do that — but that it reads the numbers together, the way a clinician does. The Kantesti AI engine behind blood-test.life starts by classifying the functional pattern from TSH and Free T4, adds Free T3 when TSH is suppressed, and then asks the questions that separate look-alike results: is this discordant (a red flag for central disease or assay interference), is it internally consistent, and does it fit the person's context — age, pregnancy status, medications, and prior results?

• Pattern first, not marker first: it labels euthyroid, subclinical, overt, T3-toxicosis, central, or non-thyroidal-illness patterns rather than reacting to one out-of-range value.
• Context-aware ranges: it applies trimester-specific ranges in pregnancy and accounts for the legitimate upward drift of TSH with age.
• Discordance detection: a low/normal TSH with a low Free T4 is flagged as possible central disease, not filed as "subclinical."
• Trend over snapshot: a single borderline TSH is treated with appropriate humility given up to ~50% within-person variation; your trajectory is weighted over any one point.
• Cross-panel reasoning: it connects a high TSH to a rising LDL, or thyroid-like symptoms to ferritin, B12, and vitamin D, rather than tunnel-visioning on the thyroid.
• Honest uncertainty: when a result does not fit any standard pattern, it says so and points toward a clinician instead of forcing a false diagnosis.

None of this makes the analyzer a diagnosis. blood-test.life is an educational tool, not a medical device, and it does not replace a clinician's judgment, a physical examination, or the antibody and imaging workup that real thyroid diagnosis often requires. What it does well is give you a clear, honest, context-aware read of what your numbers are most likely saying and which of them deserve a conversation with your doctor. You can see the reasoning approach in more depth on our how-it-works page, and if you want to understand the machine-learning side, our blog on reading blood test results with AI goes further. Access is described on the pricing page.

Lifestyle, monitoring, and next steps

If your thyroid is genuinely abnormal, the effective treatments are medical — levothyroxine for hypothyroidism, and antithyroid drugs, radioactive iodine, or surgery for hyperthyroidism — and lifestyle changes do not substitute for them. That said, a few evidence-based habits support the system and prevent avoidable problems. Get enough, but not excessive, iodine: in iodine-sufficient countries most people meet their needs from iodized salt, dairy, and seafood, and megadoses of iodine or kelp supplements can actually trigger dysfunction. Selenium is being studied in autoimmune thyroiditis but is not an established treatment, and mainstream guidelines do not recommend routine high-dose supplementation.

1. Confirm before you act: repeat a borderline TSH under standardized conditions (same lab, morning, off biotin) before drawing conclusions.
2. Track the trend, not the snapshot: keep your results in one place so the trajectory is visible over months and years.
3. Take levothyroxine consistently: same time daily, on an empty stomach, separated from calcium, iron, and coffee, which all reduce absorption.
4. Recheck at the right interval: allow roughly six weeks after any dose change before rechecking TSH, because the axis re-equilibrates slowly.
5. Address the co-travelers: correct iron, B12, and vitamin D deficiencies that mimic or compound thyroid symptoms.
6. Flag pregnancy and new medications: both change targets and interpretation, sometimes urgently.

For the many people whose thyroid panel is entirely normal, the useful next step is to stop blaming the thyroid and look elsewhere for the symptom — sleep, iron status, vitamin D, blood sugar, mood, and stress are far more common drivers of fatigue and brain fog than a normal thyroid ever is. A normal, clinician-reviewed thyroid panel is genuinely reassuring information, not a dead end.

When to see a clinician

This guide and any analyzer are for understanding, not diagnosis. Some findings deserve prompt, real-world medical attention rather than watchful waiting, and a few are genuinely urgent. Use the list below as a prompt to pick up the phone, and remember that context — your symptoms, your history, your medications — always outranks a single number.

• A suppressed TSH with palpitations, an irregular or racing heartbeat, tremor, or unexplained weight loss — get seen; new atrial fibrillation is a medical emergency.
• A clearly high TSH with a low Free T4, or a TSH above 10 mIU/L — established hypothyroidism that usually warrants treatment.
• A discordant result — a low or normal TSH with a genuinely low Free T4 — which can signal pituitary disease and needs specialist assessment.
• Any abnormal or borderline result in pregnancy or if you are trying to conceive — time-sensitive.
• A neck lump, difficulty swallowing, hoarseness, or eye changes (bulging, double vision, pain).
• New thyroid abnormalities after starting lithium, amiodarone, or an immune checkpoint inhibitor.
• Severe symptoms of either extreme — confusion and profound cold in hypothyroidism, or high fever and agitation in hyperthyroidism — which can indicate the life-threatening states of myxedema coma or thyroid storm and require emergency care.

Myths and misconceptions

"A normal TSH rules out all thyroid disease."

Mostly true, but not absolutely. TSH is the most sensitive single screen for the common primary thyroid disorders, and a normal TSH makes them unlikely. But it can miss central hypothyroidism (where a failing pituitary produces an inappropriately normal TSH despite a low Free T4) and early T3-toxicosis if free hormones are not also checked. This is why discordant symptoms sometimes justify going beyond TSH alone.

"If I feel tired, my thyroid must be the problem."

Fatigue is real, but it is one of the least specific symptoms in medicine. Iron deficiency, low vitamin D, poor sleep, depression, and blood-sugar problems all cause it, and all are more common than undiagnosed thyroid disease in someone with a normal panel. A normal, reviewed thyroid result is a reason to look elsewhere, not to keep re-testing the thyroid.

"Reverse T3 tells me my thyroid is secretly failing."

Reverse T3 is popular online but of limited value in routine diagnosis. A high reverse T3 usually reflects the body's normal adaptation to illness or stress (non-thyroidal illness), and mainstream bodies including the American Thyroid Association do not recommend it for diagnosing hypothyroidism or for guiding treatment.

"Everyone with a mildly high TSH needs a pill."

No. Subclinical hypothyroidism is common, often transient, and frequently does not require treatment, especially in older adults with a TSH under 10 and no antibodies. Overtreating a borderline number can push someone into a mild hyperthyroid state with its own risks to heart and bone. The decision is individualized — the opposite of a reflex.

"Supplements and diets can cure an autoimmune thyroid."

There is no supplement, elimination diet, or detox that reliably reverses Hashimoto's or Graves'. Adequate iodine matters, and correcting genuine deficiencies helps you feel better, but established disease is treated medically. Beware of megadose iodine or kelp, which can actually trigger dysfunction rather than prevent it.

Frequently asked questions