TSH Blood Test: 3rd Generation, Normal Range, High & Low

At a Glance

TSH (thyroid-stimulating hormone, thyrotropin) is made by the pituitary gland, not the thyroid. It is the signal the pituitary sends telling the thyroid how much hormone to produce. Because the pituitary raises TSH when thyroid hormone runs low and suppresses it when thyroid hormone runs high, TSH usually moves in the opposite direction to thyroid function. A high TSH generally points toward an underactive thyroid; a low TSH generally points toward an overactive one.

That inverse relationship is the most common source of confusion on this test.

Your report may label the result TSH, TSH, 3rd Generation, TSH-ICMA, or TSH, High Sensitivity. Test code 899 is Quest-specific, while TSH with reflex to FT4 names an ordering workflow whose trigger is laboratory-specific. These labels belong to the same broad analyte, but they are not interchangeable report identities.

If you remember only one thing

TSH is the signal; free T4 tests the feedback loop. In primary thyroid disease they usually move in opposite directions. When they do not, think about timing, treatment, illness, pituitary disease, pregnancy, or assay interference before forcing the result into “high means low thyroid” or “low means high thyroid.”

What the label on your report means

Laboratories use several related names. Read the exact label and source laboratory before assigning its range or reflex behavior.

TSH, 3rd Generation / TSH, High Sensitivity / TSH Ultrasensitive. “Generation” describes low-end assay sensitivity, not a newer hormone. LOINC provides method-specific terms at detection limits of ≤0.05 and ≤0.005 mIU/L, while Quest calls test 899 third-generation and publishes 0.01 mIU/L sensitivity. That mismatch is why the marketing label alone cannot determine the method-specific LOINC. Use the laboratory's actual result code; 3016-3 is the generic serum-or-plasma TSH code when method is unspecified.

Quest Diagnostics, which reports TSH under test code 899, states that it measures TSH using an ultrasensitive third-generation chemiluminescent assay with a sensitivity of 0.01 mIU/L.

The lower detection limit matters most when TSH is strongly suppressed. Do not infer assay generation solely from whether the report prints a small number; use the method or test identity supplied by the laboratory.

TSH-ICMA. ICMA stands for immunochemiluminometric assay — antibodies capture the TSH molecule and the amount captured is read as emitted light. It describes the instrument, not a variant of the test.

TSH w/Reflex to FT4 (also written TSH rfx, TSH reflex, w/rflx to ft4). This is an ordering rule rather than a different test. The laboratory measures TSH first. If the TSH falls outside the laboratory's limits, the lab automatically runs free T4 on the same tube of blood. If TSH is normal, no second test is run. The practical benefit is that an abnormal screening result gets its most important follow-up test without you returning for a second blood draw.

If you ordered a reflex test and see only TSH, the reflex criteria may not have been met—but the exact trigger must come from that panel's source specification, not from the words “reflex to FT4” alone.

Your number, and what it means

Before looking anything up, check one thing. Several numbers people search alongside "TSH" are not results at all — they are the boundaries of the reference range printed on the report. 0.26, 0.40, 0.43, 0.45, 0.55, 2.66, 2.73, 2.91, 4.30 and 4.50 are all commonly printed range limits. If the number you are trying to interpret is one of these, and it appears next to words like "reference range" or "trimester," you may be reading the range rather than your own value. Your result is usually the number set apart in its own column, sometimes flagged with an H or an L.

With that checked, the working adult range on this page is 0.45 – 4.5 mIU/L. Your own report may print something marginally different: Quest publishes 0.40–4.50 mIU/L for adults 20 years and older, and Labcorp publishes 0.450–4.500 μIU/mL above age 10. Note where those actually differ — the upper limits are the same number, and the disagreement is confined to a 0.05 window at the bottom. A result between 0.40 and 0.45 is the only case where the two labs would classify you differently. Always read against the interval printed on your own report.

Units matter more than that gap does. mIU/L and μIU/mL are numerically equivalent, so 2.1 of one is 2.1 of the other. IU/L is not equivalent — one IU/L equals 1,000 mIU/L. A range written as 0.45–4.5 IU/L is a thousandfold error, not an alternative unit. Confirm the printed unit before comparing values.

Your own trend is often more informative than a population interval. Quest's FAQ cites a biological-variation study reporting 0.7 mIU/L as a significant serial difference. This is not a universal decision threshold: baseline TSH, assay, timing, pregnancy, illness, and treatment state all still matter. The usable comparison is between results drawn under similar conditions and, where possible, on the same assay.

How TSH and free T4 read together

TSH alone is a screening signal. The pairing of TSH with free T4 is what identifies the pattern. This table is the reasoning most clinicians apply to a thyroid panel.

Two rows deserve emphasis. High TSH with normal free T4 supports a subclinical hypothyroidism pattern; T3 is not what defines that category. Low or inappropriately normal TSH with low free T4 can point toward central hypothyroidism, but the pattern requires clinical evaluation rather than diagnosis from the pair alone.

Reference ranges

Adults: this page uses 0.45 – 4.5 mIU/L as its default band, and your own report's interval takes precedence over it. Quest uses 0.40–4.50 mIU/L at age 20 and older; Labcorp uses 0.450–4.500 μIU/mL above age 10. The upper limits agree; the two differ only between 0.40 and 0.45.

The upper limit is method- and population-dependent. Quest's clinical FAQ recommends using the interval established by the laboratory performing the assay. This is why a value near a boundary should be compared with the range printed on that report rather than a universal internet cutoff.

Children: ranges are higher and narrow with age. Labcorp publishes 0.700–5.970 mIU/L for ages 1–5, 0.600–4.840 for ages 6–10, and the adult 0.450–4.500 above age 10. Newborn values are far higher still and fall rapidly in the first days of life.

Pregnancy: TSH shifts downward, most sharply in the first trimester, because hCG stimulates the TSH receptor. Quest publishes these assay-specific pregnancy reference ranges:

These are Quest ranges, not universal pregnancy cutoffs. Use the pregnancy interval supplied by the laboratory that performed the assay. The 2017 American Thyroid Association guideline prioritizes population- and assay-specific ranges; when those are unavailable, it describes adjusting the nonpregnant upper limit rather than automatically applying the older fixed 2.5/3.0 thresholds.

Why your TSH changes between tests

Time of day. TSH follows a daily rhythm, peaking around midnight and reaching its low point around midday at roughly half the peak value. This has a consequence often stated backwards: population reference intervals are largely derived from daytime samples, closer to the trough than the peak. An early-morning draw therefore sits nearer the peak and can read somewhat higher than an afternoon draw from the same person on the same day. The defensible advice is not a particular hour but consistency — same approximate time, same laboratory, when results are compared over time.

Recent illness or hospitalization. Acute illness disturbs the thyroid axis in ways that resolve on recovery. Thyroid testing during or shortly after significant illness is difficult to interpret.

Assay platform. TSH is not fully standardized across manufacturers, so different platforms can return meaningfully different numbers from the same sample. Comparing results within one laboratory is more reliable than across two.

Medication timing. After a change in thyroid hormone dose, TSH takes time to settle — commonly 6 to 8 weeks. Testing sooner measures a system still in transit.

What can make the number wrong

Biotin. On susceptible platforms, high-dose biotin can produce falsely low TSH together with falsely high free T4 or T3, mimicking hyperthyroidism. Not every assay uses the affected chemistry. The American Thyroid Association advises avoiding biotin for at least two days before thyroid testing; the required interval can be longer depending on dose, kidney function, and platform, so follow the laboratory or clinician's instructions.

Heterophile antibodies, including human anti-mouse antibodies (HAMA). These can bridge the assay's antibodies and produce an artificially high or artificially low result. Where interference is suspected, the sample can be re-run using a different method or with specific blocking agents; Quest offers this as a dedicated TSH with HAMA Treatment test.

Anti-streptavidin antibodies. Less common than biotin but increasingly recognized, and producing the same falsely-low-TSH pattern. Unlike biotin, this interference is endogenous, so a washout period does not help.

Macro-TSH. TSH bound into large immune complexes is cleared slowly and accumulates, producing a persistently elevated TSH in someone with normal free T4 and no symptoms.

Medications. Amiodarone, lithium, corticosteroids and dopamine all affect TSH through genuine physiological mechanisms rather than assay interference.

Two people with the same result

Hypothetical example. Two people both have a TSH of 6.2 mIU/L and a free T4 within the laboratory interval. On the panel alone, the pair is identical.

The first also has positive TPO antibodies and a similar elevation on a previous test.

The second has negative antibodies and a previously normal TSH, recorded before a documented recent illness.

The first pattern supports persistent autoimmune thyroid dysfunction more strongly; the second may be transient. The distinguishing information is the antibody result and the earlier value, not the TSH. Neither result alone determines treatment.

What this result cannot tell you

It cannot identify why the thyroid is underperforming — autoimmune disease, iodine imbalance, prior surgery or radioiodine, and medication effects all raise TSH. Antibody testing and history make that distinction.

It cannot separate subclinical from overt disease; that requires free T4.

It cannot establish that every borderline result represents persistent thyroid dysfunction; repeat testing may be appropriate depending on context.

It cannot rule out hypothyroidism in someone with pituitary disease, where the compensating rise in TSH never happens.

Common interpretation mistakes

Reading TSH in the same direction as thyroid function. High TSH does not mean an overactive thyroid. It usually means the opposite.

Treating a single borderline value as a diagnosis. A TSH of 4.8 on one draw does not by itself establish persistent thyroid dysfunction or determine treatment. Repeat testing may be appropriate depending on free T4 and clinical context.

Comparing results across laboratories. Platform differences can move the number without anything having changed in you.

Assuming "3rd generation" means a more serious or more advanced test was ordered. It is the routine assay in most laboratories today.

Believing TSH can be lowered through diet. Where a raised TSH reflects genuine thyroid underactivity, the treatment is thyroid hormone replacement. No food, supplement or exercise regimen reliably lowers TSH, and iodine supplementation taken on the assumption that it will can make matters worse.

Questions your doctor may ask

• Are you taking biotin, or a hair, skin and nail supplement? At what dose, and when did you last take it?
• Have you been significantly ill or hospitalized in the past two months?
• Are you taking levothyroxine or any thyroid medication, and how consistently, and how long since the last dose change?
• Are you taking amiodarone, lithium, corticosteroids, or a medication containing iodine?
• Is there thyroid disease in your family?
• Are you pregnant, or might you be?
• What time of day was the blood drawn, and was it the same laboratory as last time?

Read together with

TSH is a screening test and is rarely interpreted alone. Free T4 is the essential companion and separates subclinical from overt primary disease patterns. Free T3 can add information when hyperthyroidism is suspected. Thyroid peroxidase antibodies and thyroglobulin antibodies support evaluation for autoimmune thyroid disease. Thyrotropin receptor antibodies support evaluation for Graves' disease.

Clinical pearls

Trends are most comparable under similar conditions — same laboratory or assay, similar draw time, and enough time after a medication change for TSH to re-equilibrate. Quest's FAQ cites a study reporting 0.7 mIU/L as a significant serial difference, though that figure does not define a threshold for every patient or assay.

Low TSH with a normal or low free T4 in someone who feels unwell is not reassuring. It is the pattern that should prompt thought about the pituitary rather than the thyroid.

Suppressed TSH can persist for months after successful treatment of Graves' disease, even once thyroid hormone levels have normalized. During that window free T4 is the more reliable guide.

A high TSH while taking levothyroxine has several explanations, including inconsistent dosing, absorption problems, interacting medicines, or a dose that needs review. Calcium and iron are timing-sensitive interactions; proton pump inhibitors act through longer-lasting changes in gastric acidity.

In adults over 70, a mildly raised TSH may be age-appropriate. The reference range drifts upward with age, and treating to a young adult's target risks overtreatment.

Clinical Takeaway

A single TSH value is a screening signal, not a diagnosis. Its meaning depends on the free T4 measured alongside it, on your own previous results, and on whether anything — biotin, recent illness, a dose change, pregnancy — was disturbing the system on the day of the draw. Borderline results may need confirmation before they drive decisions.

In one sentence

TSH is the pituitary's instruction to the thyroid, it moves opposite to thyroid function, and it means something specific only when read together with free T4 and your own previous values.

Bottom line

If TSH is high or low, free T4 tests whether the usual inverse-feedback pattern holds. T3 can matter when TSH is low and free T4 is normal. Use the interval printed by the source laboratory; Quest and Labcorp publish slightly different adult limits. A normal TSH is reassuring in the usual screening context, but it does not override pregnancy-specific ranges, pituitary disease, treatment targets, or a clinically discordant free T4.