01Purpose and principlesWhat the assessment is for and the core concepts behind it.
The hypothalamic-pituitary-thyroid axis is a feedback system: small changes in circulating free thyroid hormone can generate a large inverse change in TSH. This makes TSH a sensitive first test when the pituitary is intact, but an unreliable stand-alone marker when pituitary function is impaired, during some acute illnesses or after certain medicines.
Interpretation has three layers. First classify the biochemical pattern; next decide whether it fits symptoms and physiology; finally test for cause only when the abnormality is persistent and credible. Reference intervals are method- and population-dependent, so the reporting laboratory's range is part of the result rather than decorative text.
NICE advises against testing solely because a patient is acutely unwell unless thyroid dysfunction is suspected as the cause, because non-thyroidal illness can produce transient abnormalities. When repeat testing is needed after a stable outpatient result, it is generally unhelpful before six weeks unless clinical deterioration demands earlier assessment.
Key points
- Start with the patient, indication, pregnancy status, current illness, medicines and timing of the sample; numbers without context invite diagnostic error.
- For most adults without suspected pituitary disease, NICE recommends TSH first, with laboratory reflex free T4 when TSH is high and free T4 plus free T3 when TSH is low.
- When hypothalamic or pituitary disease is plausible, request TSH and free T4 together because a reassuring-looking TSH can coexist with central hypothyroidism.
- Raised TSH with low free T4 supports primary hypothyroidism; raised TSH with normal free T4 is the biochemical pattern of subclinical hypothyroidism.
- Suppressed TSH with raised free T4 and/or free T3 indicates overt thyrotoxicosis, whereas normal peripheral hormones describe subclinical hyperthyroidism after confounders are excluded.
- Low free T4 with low, normal or only modestly raised TSH prompts central disease, non-thyroidal illness, drug effect or assay interference rather than automatic levothyroxine prescribing.
- Ask specifically about high-dose biotin in supplements, amiodarone, lithium, glucocorticoids and sample timing after levothyroxine; each can distort the apparent pattern.
- Use pregnancy- and assay-specific reference intervals; ordinary adult ranges and total hormone concentrations are not interchangeable with pregnancy interpretation.
- If results conflict with the clinical picture, contact the biochemistry laboratory, repeat when appropriate and consider an alternative assay before investigating rare endocrine syndromes.
02Indications, selection and cautionsWhen it is useful, when urgency changes and important limitations.
TSH is above the assay range and free T4 is low. Autoimmune thyroiditis, previous radioiodine or thyroid surgery, iodine exposure and medicine effects are common causes; severe symptoms or pregnancy change urgency.
Free T4 is low while TSH is low, normal or inappropriately unremarkable for the degree of hormone deficiency. Seek headache, visual change, amenorrhoea, reduced libido, other pituitary deficits and glucocorticoid deficiency before treatment.
TSH is suppressed with free T4 and/or free T3 above range. Pulse, temperature, rhythm, heart failure, agitation and gastrointestinal or hepatic features determine whether this is uncomplicated disease or a threatened thyroid emergency.
A low TSH with normal free T4 but raised free T3 may be early Graves disease or nodular autonomy. Confirm persistence and investigate cause rather than labelling the patient subclinical from free T4 alone.
Acute systemic illness can lower T3, alter TSH transiently and, when severe, lower free T4. The profile must be interpreted against illness trajectory and usually repeated after recovery rather than treated empirically.
Raised free hormones without suppressed TSH, abrupt fluctuations or a result inconsistent with examination suggests biotin or antibody interference, dosing immediately before sampling, non-adherence with catch-up dosing, medicine effects, or much less commonly thyroid hormone resistance or a TSH-secreting tumour.
03Method and interpretationA systematic approach to the test and its findings.
Read from the initial assessment onwards. Tests may run in parallel in urgent care; first-line, preferred, confirmatory, definitive and gold-standard labels appear only when the chapter explicitly states them.
- 01
TSH with reflex free T4 and free T3First step - Why
- Classify thyroid status efficiently when pituitary disease is not suspected.
- Interpretation and limitations
- A high TSH triggers free T4 measurement; a low TSH needs both free T4 and free T3 to detect T3-predominant thyrotoxicosis. Apply the laboratory's own intervals.
- 02
Paired TSH and free T4 - Why
- Avoid missing central hypothyroidism in a relevant clinical setting.
- Interpretation and limitations
- Low free T4 with a TSH that is not appropriately raised is abnormal even if TSH lies inside its printed range; discuss pituitary evaluation with endocrinology.
- 03
Repeat thyroid profile - Why
- Confirm persistence after a borderline, transient or discordant outpatient result.
- Interpretation and limitations
- Repeat using comparable conditions and usually no sooner than six weeks; defer until recovery from intercurrent illness when safe, and record levothyroxine or supplement timing.
- 04
Thyroid peroxidase antibodies - Why
- Support autoimmune aetiology when TSH is persistently raised.
- Interpretation and limitations
- A positive result increases the likelihood of autoimmune thyroiditis and progression, but antibody titre does not measure replacement adequacy and should not be repeatedly trended.
- 05
TSH-receptor antibodies - Why
- Confirm Graves disease in biochemically established thyrotoxicosis.
- Interpretation and limitations
- A positive TRAb result supports Graves; when negative and the cause remains uncertain, NICE advises considering technetium scintigraphy rather than indiscriminate ultrasound.
- 06
Assay interference assessment - Why
- Resolve biochemistry that is internally inconsistent or clinically implausible.
- Interpretation and limitations
- Ask about biotin, discuss dilution or blocking studies and repeat on another analytical platform through the laboratory. Exclude interference before ordering pituitary imaging for rare diagnoses.
- 07
Pituitary and adrenal assessment - Why
- Investigate a low free T4 with inappropriately low or normal TSH.
- Interpretation and limitations
- Morning cortisol and other pituitary axes require endocrine interpretation; acute headache, visual deficit, hypotension or hypoglycaemia warrants emergency assessment rather than routine outpatient testing.
04Clinical next stepsHow the result changes management or prompts escalation.
01First requestChoose the correct initial profileFirst stepSymptoms or signs make thyroid dysfunction a reasonable clinical possibility.+
- 1Check for acute illness, pregnancy, pituitary features, previous thyroid treatment, relevant medicines and high-dose biotin before selecting tests.
- 2Request TSH alone with the local reflex cascade when pituitary function is likely intact; request paired TSH and free T4 if central disease is suspected.
- 3Interpret the reported pattern with the examination and assay range, then decide whether urgent physiological assessment, cause-specific testing or planned confirmation is required.
02DiscordanceResolve a result that does not fitThyroid hormones, TSH and clinical findings do not form a coherent pattern.+
- 1Recheck identity, units, reference intervals, pregnancy status, acute illness and exact timing of prescribed hormone, antithyroid medicine and supplements.
- 2AlternativeDiscuss suspected interference with the duty biochemist and arrange repeat or alternative-platform testing rather than simply requesting the same assay repeatedly.
- 3Only after artefact and common physiological explanations are addressed should endocrine specialists pursue macro-TSH, resistance to thyroid hormone or a TSH-secreting pituitary lesion.
03Low free T4Separate primary from central diseaseFree T4 is below range on a credible sample.+
- 1Use TSH direction: a clearly raised concentration supports primary failure, whereas an inappropriately low, normal or modest result raises central disease or illness effects.
- 2If central disease is plausible, assess haemodynamic and glycaemic stability, obtain cortisol if it will not delay care and involve endocrinology before thyroid replacement.
- 3Treat possible adrenal insufficiency before levothyroxine when clinically indicated, then investigate other pituitary axes and imaging under specialist guidance.
04Suppressed TSHConfirm severity and identify causeTSH is below the reporting laboratory's lower reference limit.+
- 1EscalationMeasure free T4 and free T3, examine cardiovascular and neurological status, and urgently escalate fever, marked tachyarrhythmia, heart failure or delirium.
- 2Review exogenous hormone, amiodarone, recent iodine and pregnancy; request TRAb when overt endogenous thyrotoxicosis is confirmed.
- 3If TRAb is negative or nodular autonomy remains likely, use endocrinology-directed scintigraphy; reserve ultrasound for a palpable or otherwise structural indication.
05Risks, monitoring and follow-upComplications, safety checks and further assessment.
- Record the assay, units and reference interval when following a result across laboratories, because numerical values are not automatically interchangeable.
- For a stable borderline outpatient abnormality, repeat timing should reflect the likely physiology and is usually at least six weeks rather than a few days.
- Track pulse, rhythm, temperature, blood pressure, mental state and heart-failure features when biochemistry suggests thyrotoxicosis; clinical deterioration overrides routine follow-up.
- In treated primary hypothyroidism use TSH as the main biochemical target, adding free T4 when symptoms persist; central disease instead requires specialist free-T4-based assessment.
- Reassess using pregnancy-specific guidance promptly after conception or when pregnancy is planned, as treatment targets and monitoring frequency differ.
06Special situationsVariants, exceptions and circumstances that change the usual approach.
Normal TSH has conditions
A TSH inside range is reassuring only when the pituitary axis is intact, sampling is interpretable and no medicine or illness has uncoupled the feedback relationship.
Magnitude needs chronology
TSH may remain suppressed after free hormones improve and may take months to fall after profound longstanding hypothyroidism; one asynchronous value should not trigger reflex dose escalation.
Biotin can mimic disease
High supplement doses can produce falsely high or low immunoassay results depending on platform design. Ask explicitly, then follow laboratory advice on withholding and retesting.
Total hormone is binding-sensitive
Pregnancy and oestrogen increase thyroxine-binding globulin, altering total T4 and T3. Use validated pregnancy and assay pathways rather than ordinary total-hormone ranges.
Ultrasound answers structure
An abnormal thyroid profile alone is not an ultrasound indication. Imaging cannot reliably determine the biochemical cause of uncomplicated thyrotoxicosis without a structural finding.
07Common pitfallsFrequent interpretation and management errors.
- 01
Ordering TSH alone despite a credible history of pituitary surgery, irradiation or mass symptoms.
- 02
Diagnosing thyroid disease from tests taken during severe intercurrent illness without judging whether thyroid dysfunction caused that illness.
- 03
Calling low TSH with normal free T4 subclinical hyperthyroidism without checking free T3.
- 04
Ignoring supplements because the medication list records only prescribed drugs.
- 05
Requesting thyroid ultrasound as the default response to any biochemical abnormality.
- 06
Pursuing a TSHoma before the laboratory has assessed assay interference and dosing behaviour.
- 07
Applying non-pregnant reference ranges and thresholds to a pregnant patient.