Thyroxine Binding Globulin (TBG)

What Thyroxine Binding Globulin (TBG) actually measures — the analyte, and the assay

The analyte is thyroxine-binding globulin: a 54 kDa hepatic glycoprotein of the serpin family that carries the large majority of circulating T4 — conventionally around three-quarters of it, with transthyretin and albumin carrying most of the rest. It is not a thyroid hormone and it has no receptor. It exists on this site for one reason: it is the commonest explanation for a thyroid panel where the total hormones look wrong and the person is well.

TBG itself is measured by immunoassay in mg/L or µg/mL, and that measurement is the easy part. The hard part is the assumption sitting underneath it — that binding-protein changes move total hormone and leave free hormone alone. That assumption is a good approximation and it is not exactly true, because automated ‘free T4’ assays do not measure free hormone directly. They estimate it, and the estimate is sensitive to the binding proteins in the sample.

The size of that error has been measured. Against LC-MS/MS, automated fT4 immunoassays deviated by +7% to +29% more in sera from pregnant women than in healthy controls — falsely high, in the condition where TBG is most raised Jansen 2022. So in exactly the situation TBG is invoked to explain, the free hormone number is drifting too.

And TBG is not the only binding protein that can do this. In familial dysalbuminemic hyperthyroxinemia an albumin variant with increased T4 affinity produces a discordant, apparently high thyroxine in a completely euthyroid person Ting 2021. A normal TBG does not close the binding-protein hypothesis.

Thyroxine Binding Globulin (TBG): what changes the blood, and what only changes the reading

What changes the protein in your blood:

1. Estrogen — by a wide margin the biggest. Pregnancy, combined oral contraceptives and oral (not transdermal) HRT raise TBG substantially, by increasing its sialylation and so slowing its hepatic clearance. This is the mechanism behind the well-documented shifts in thyroid results and thyroxine requirement seen with estrogen-containing contraceptives Alkahlout 2025. It is also the exposure people most often fail to declare, because a contraceptive does not feel like a drug.
2. Androgens and glucocorticoids lower it — the mirror image, and directly relevant to anyone running exogenous androgens who finds their total T4 and T3 sitting low with a normal TSH.
3. Liver and kidney. Cirrhosis lowers synthesis; nephrotic syndrome loses TBG in the urine; acute hepatitis can raise it.
4. Inherited TBG excess or deficiency, X-linked, and commoner than people expect — partial TBG deficiency is a routine cause of an isolated low total T4 in a well man.

What changes only the reading — on this test and on the thyroid tests beside it:

1. The fT4 assay's own binding-protein sensitivity, quantified above at +7% to +29% in pregnancy Jansen 2022.
2. Biotin. On streptavidin-biotin platforms high-dose biotin distorts thyroid results in a characteristic pattern — typically high free hormones with a suppressed TSH, mimicking Graves’ disease in someone who has none Balzer 2023.
3. Heterophile antibodies, anti-streptavidin antibodies, macro-TSH and anti-T4 autoantibodies, each producing its own signature of discordance. The current review catalogs them Al-Bahadili 2024.
4. An albumin variant with altered T4 affinity Ting 2021.

Reference interval or decision threshold — which kind of number Thyroxine Binding Globulin (TBG) is

A reference interval, method-specific, and there is no other kind of number available here. No guideline body publishes an optimal TBG, and none ever will, because TBG is a carrier rather than a signal: nobody has shown that moving it changes an outcome, and there is no plausible route by which it would.

That makes TBG unusual and worth stating plainly: this is a test whose value is not the value. Its job is to explain a discordance. A TBG sitting mid-interval has done its job if it rules the binding-protein hypothesis out, and a TBG at the top of the interval in a woman on the pill has done its job if it rules it in. Reading it as a level to optimize is a category error.

The decision thresholds in thyroid medicine belong to TSH and to free hormone — and even those move with the assay and the physiological state, which is what the interference literature is about Al-Bahadili 2024.

How you would know your Thyroxine Binding Globulin (TBG) was wrong — and when to redraw

Roughly five days — that is TBG's own plasma half-life — so changes take weeks, not days, to complete. An estrogen-driven rise builds over several weeks after starting the pill or conceiving, and unwinds over a similar period after stopping. A TBG drawn inside six weeks of an estrogen change is measuring a transition, and a total T4 drawn alongside it is measuring the same transition twice.

The hormone's own kinetics stretch that window further. Thyroxine has a plasma half-life of roughly 7 days — unusually long, precisely because so much of it is protein-bound — against about 1 day for T3. Five half-lives is the conventional point at which a new steady state has been reached, so a total T4 needs on the order of 5 weeks to finish responding to a change in TBG. Redrawing at 2 weeks and calling the result a plateau is the commonest timing error on this panel.

What would have to change for the second value to mean anything. TBG is never retested for its own sake. Redraw it on the same sample as TSH, free T4 and free T3, and ask whether the discordance that prompted it has resolved. If the frees and the TSH are now concordant and the person is well, the story is finished and no further TBG is needed — that is a successful test, not an inconclusive one.

If a discordance persists with a normal TBG, the next test is not another TBG. The remaining hypotheses are assay interference — biotin, heterophile or anti-streptavidin antibodies, macro-TSH Al-Bahadili 2024 Balzer 2023 — or a binding variant such as familial dysalbuminemic hyperthyroxinemia Ting 2021, and each of those is answered by running the sample on a different platform or by equilibrium dialysis LC-MS/MS, not by repeating the same one.

What Thyroxine Binding Globulin (TBG) cannot tell you

It cannot tell you whether you are hypothyroid or hyperthyroid. It contains no information about thyroid function at all.

It cannot be treated, and there is no reason to want to. A raised TBG on the pill is a normal physiological response, not a finding.

It adds nothing when your panel already reports free hormones and TSH agrees with them. Ordering it routinely is spending money to confirm something the panel has already established.

And it does not explain every discordant panel. Non-thyroidal illness, biotin, heterophile antibodies, macro-TSH and an albumin variant all produce the same shape on the report Al-Bahadili 2024 Ting 2021. The wrong inference — and this one causes real harm — is a woman on a combined contraceptive whose total T4 is read as thyroid disease and who is started on levothyroxine for a binding protein.

Sources read for these sections

• Jansen HI, et al. Pregnancy disrupts the accuracy of automated fT4 immunoassays. European Thyroid Journal 2022 · PMID 36219545
• Ting MJM, et al. Familial Dysalbuminemic Hyperthyroxinemia as a Cause for Discordant Thyroid Function Tests. Journal of the Endocrine Society 2021 · PMID 33728390
• Al-Bahadili H, et al. The Complex Web of Interferences with Thyroid Function Tests. Endocrine Practice 2024 · PMID 39477092
• Alkahlout L, et al. The Association Between Estrogen-Containing Oral Contraceptive Pills and Hypothyroidism. International Journal of Endocrinology 2025 · PMID 40135042
• Balzer AHA, et al. An Analysis of the Biotin-(Strept)avidin System in Immunoassays: Interference and Mitigation Strategies. Current Issues in Molecular Biology 2023 · PMID 37998726