SHBG Unit Converter
SHBG Unit Converter
Sex hormone-binding globulin is reported in nmol/L essentially everywhere, and nmol/L is what every free testosterone and free androgen index formula expects — so the ladder here is short on purpose. The µg/dL figure you may have been given is a calibration against an assumed 90 to 95 kDa dimer, not a unit conversion, and this page keeps it off the dropdown and explains why.
SHBG converter
Molar prefixes onlySHBG 32 nmol/L in a man aged 34
The ladder, and the conversion that is deliberately not on it
µg/dL → nmol/L = × 10 000 ÷ molecular weight of the dimer
= × 0.1111 at 90 kDa, or × 0.1053 at 95 kDa — which is the problem
- nmol/L
- the reporting unit. Every published SHBG reference interval, every Vermeulen free testosterone calculation and every free androgen index expects this unit, and this page opens in it for that reason
- pmol/L
- a thousand times the nmol/L figure, offered because free testosterone is reported in pmol/L and the two get confused. An SHBG of 32 nmol/L is 32,000 pmol/L; a free testosterone of 325 pmol/L is 0.325 nmol/L. They are a thousandfold apart and neither is the other
- no mass unit on the ladder
- SHBG is a heavily glycosylated homodimer. Published masses for the dimer run from about 90 to 95 kDa, and the glycosylation varies between individuals and is increased by a common genetic variant, so grams per mole is not a constant. A mass-to-molar factor for SHBG is a calibration, and a unit dropdown would present it as an identity
- why 90 to 95 kDa
- the mature subunit is 373 residues, near 40.5 kDa before its carbohydrate, so the unglycosylated dimer is about 81 kDa. The remaining 9 to 14 kDa is sugar, at two N-linked sites and one O-linked site per subunit, with a third N-linked site in carriers of the D356N variant
- what SHBG is for
- not usually a diagnosis in itself. It is the number that decides how much of a total testosterone is available, which is why it is measured alongside testosterone and albumin rather than alone
Worked example
SHBG 32 nmol/L in a man aged 34
32 nmol/L is 32.0 nmol/L — this page opens in the unit the result is already in, because that is the unit everything downstream wants
= 32,000 pmol/L, which is a relabelling and not a conversion
Within Labcorp's 16.5–55.9 nmol/L interval for men aged 20 to 49 on a Roche electrochemiluminescence assay
In mass units it is about 2.9 mg/L, or 288 µg/dL, if the dimer is taken as 90 kDa — and about 3.0 mg/L, or 304 µg/dL, at 95 kDa. That 5% spread is why those units are not in the dropdown above
Feed the 32 nmol/L, not a converted mass figure, into the calculated free testosterone or the free androgen index
The µg/dL relation, and the two assumed masses behind it
| Assumed dimer mass | µg/dL → nmol/L | nmol/L → µg/dL | nmol/L → mg/L (= µg/mL) | 32 nmol/L becomes |
|---|---|---|---|---|
| 90 kDa — the commonly quoted figure | × 0.1111 | × 9.0 | × 0.090 | 288 µg/dL, or 2.88 mg/L |
| 95 kDa — from a 2022 review of SHBG structure | × 0.1053 | × 9.5 | × 0.095 | 304 µg/dL, or 3.04 mg/L |
| 81 kDa — the unglycosylated dimer, for reference only | × 0.1235 | × 8.1 | × 0.081 | 259 µg/dL, or 2.59 mg/L |
Two reference laboratories, two platforms, two sets of intervals
| Group | Labcorp 082016 (Roche ECLIA) | Mayo SHBG1 (immunoenzymatic) |
|---|---|---|
| Male, adult | 16.5–55.9 nmol/L (20–49 y) 19.3–76.4 nmol/L (over 49 y) | 13.3–89.5 nmol/L (18 y and over) |
| Female, reproductive age | 24.6–122.0 nmol/L (20–49 y) | 18.2–135.5 nmol/L (18–46 y) |
| Female, older | 17.3–125.0 nmol/L (over 49 y) | 16.8–125.2 nmol/L (47–91 y) |
| Boys, Tanner I → V | Not published | 17–135 → 3.9–40 nmol/L |
| Girls, Tanner I → V | Not published | 16–182 → 10–79 nmol/L |
| Pregnancy | Not published | Reference values not established |
What moves SHBG, and by how much where anyone has measured it
| Direction | Cause | Quantified where available |
|---|---|---|
| Raises SHBG | Ageing | In men over 70, SHBG rose with age (r = 0.2, P < 0.0001) while total testosterone increased marginally and free testosterone fell — Health in Men Study |
| Raises SHBG | Oestrogen, and the route matters more than the dose | At 12 months of feminising therapy: +242% on oral estradiol with a GnRH agonist, +81% on oral estradiol with cyproterone acetate, +63% on transdermal estradiol with a GnRH agonist, and no significant change on transdermal estradiol with cyproterone acetate |
| Raises SHBG | Thyroid hormone excess — hyperthyroidism | T3 specifically stimulated SHBG secretion by human hepatoma cells, dose responsively, half-maximal at about 3 × 10⁻¹¹ mol/L free T3 |
| Raises SHBG | Cirrhosis and hepatitis, HIV disease, anticonvulsants, SHBG gene polymorphisms | Listed by the Endocrine Society testosterone guideline |
| Lowers SHBG | Androgens, progestins and glucocorticoids | Masculinising testosterone therapy: −43% at 12 months (95% CI −48 to −36) |
| Lowers SHBG | Obesity, insulin resistance and type 2 diabetes | A low SHBG is described as a marker of insulin resistance; hepatic SHBG production falls when de novo lipogenesis and inflammatory cytokines suppress HNF-4α |
| Lowers SHBG | Hypothyroidism, nephrotic syndrome, acromegaly, SHBG gene polymorphisms | Listed by the Endocrine Society testosterone guideline |
One unit, and a number that is only useful as an input
Sex hormone-binding globulin is the high-affinity carrier for testosterone and oestradiol in plasma. It is a homodimer, each subunit 373 amino acids with three oligosaccharide side chains, and each dimer binds one steroid molecule. Almost all circulating testosterone is bound — tightly to SHBG, loosely to albumin — and only the small unbound fraction, with some of the albumin-bound portion, is available to tissue. So SHBG determines how much of a total testosterone is actually doing anything, which is the whole of its clinical interest and the reason it is measured alongside testosterone and albumin rather than on its own.
That makes the unit question unusually simple and unusually important. SHBG is reported in nmol/L by essentially every laboratory, and nmol/L is what the calculations downstream require: the Vermeulen equation for calculated free testosterone takes total testosterone in nmol/L, SHBG in nmol/L and albumin in g/dL, and the free androgen index is total testosterone divided by SHBG with both in nmol/L. An SHBG entered in the wrong unit does not produce an obviously wrong answer; it produces a free testosterone that is out by a clean factor of nine or of a thousand, which is exactly the kind of error that gets acted on. This page therefore opens in nmol/L and offers only pmol/L beside it, and pmol/L is there mainly as a warning: free testosterone is reported in pmol/L and SHBG in nmol/L, a thousandfold apart, and confusing the two is a real and repeated mistake.
The conversion that is missing is the one to mass units, and its absence is deliberate. SHBG is heavily glycosylated, and the sugar is a large share of the molecule: the unglycosylated dimer is about 81 kDa while published figures for the glycosylated dimer run from about 90 to 95 kDa. Glycosylation varies between individuals, and a common genetic variant at residue 356 creates an additional N-linked site, so grams per mole is not a constant for this measurand. At 90 kDa a µg/dL result converts to nmol/L by multiplying by 0.1111; at 95 kDa by 0.1053. That is a 5% systematic spread before any assay’s own calibration is taken into account, which makes the figure a calibration rather than a unit conversion. Putting it in a dropdown would present an assumption as arithmetic, so it sits in a table above with both derivations shown, and a converted µg/dL value should never be fed into a free testosterone calculation.
What the number means depends almost entirely on what is moving it, which is why the last table is longer than the rest of the page. The Endocrine Society’s testosterone guideline lists decreased SHBG with obesity, diabetes mellitus, glucocorticoid, progestin and androgenic steroid use, nephrotic syndrome, hypothyroidism, acromegaly and SHBG gene polymorphisms, and increased SHBG with ageing, HIV disease, cirrhosis and hepatitis, hyperthyroidism, anticonvulsant use, oestrogen use and the same gene polymorphisms. Each of those shifts the free fraction, which is why the same guideline recommends determining free testosterone either by equilibrium dialysis or by calculation from total testosterone, SHBG and albumin — and warns against direct analogue-based free testosterone immunoassays as inaccurate.
The clearest demonstration of why this matters comes from ageing. In men over 70 followed in the Health in Men Study, total testosterone increased marginally with age while free testosterone declined, and SHBG rose. A rising SHBG binds more of the same total, so the total looks stable while the available hormone falls — and a clinician reading only the total concludes nothing is happening. The mirror image is the obese or insulin-resistant patient, whose SHBG is low: the total testosterone reads low and the free testosterone may be normal, so treating the total would be treating the carrier protein. In both directions the SHBG is what turns a total testosterone into an interpretable result, and the unit it arrives in has to be nmol/L for any of that arithmetic to work.
Frequently asked questions
What unit is SHBG reported in?
nmol/L, almost universally, and it is the unit you want: every published reference interval and every calculated free testosterone or free androgen index formula expects nmol/L. This page opens in nmol/L for that reason and offers pmol/L beside it, which is simply a thousand times the same number.
How do I convert SHBG from µg/dL to nmol/L?
You can approximate it by multiplying by 0.111, but be clear that this is a calibration rather than a unit conversion. It assumes the dimer weighs 90 kDa; a 2022 review of SHBG structure gives approximately 95 kDa, which would make the factor 0.1053. The glycosylation that accounts for the difference varies between individuals, so the answer carries at least a 5% systematic uncertainty and should not be used in a free testosterone calculation.
Why does this converter not offer µg/dL or mg/L?
Because SHBG is a heavily glycosylated homodimer with no fixed mass per mole. The unglycosylated dimer is about 81 kDa and published figures for the glycosylated form run from 90 to 95 kDa, with a common genetic variant adding a further glycosylation site. Offering the factor in a unit dropdown would present that assumption as exact arithmetic, so it is published in a table with its derivation instead.
What is a normal SHBG?
It depends on sex, age and the platform, and by more than most people expect. Labcorp gives 16.5–55.9 nmol/L for men aged 20 to 49 and 19.3–76.4 above 49, with 24.6–122.0 and 17.3–125.0 for the corresponding women, on a Roche electrochemiluminescence assay. Mayo, on an immunoenzymatic assay, gives 13.3–89.5 for adult men. Use the interval on your own report.
What raises and lowers SHBG?
It rises with ageing, oestrogen use, hyperthyroidism, cirrhosis and hepatitis, HIV disease and anticonvulsants; it falls with obesity, diabetes, androgens, progestins and glucocorticoids, nephrotic syndrome, hypothyroidism and acromegaly. Route of administration matters: oral oestradiol with a GnRH agonist raised SHBG by 242% at a year, while transdermal oestradiol with cyproterone acetate did not change it significantly.
Why does my total testosterone look normal when I have symptoms?
A high SHBG is one of the commonest explanations. It binds more of the same total testosterone, so the total reads reassuringly while the free fraction is genuinely low — the pattern seen with ageing, oestrogen use and hyperthyroidism. Calculate the free testosterone from the total, the SHBG in nmol/L and albumin, or measure it by equilibrium dialysis.
Related calculators
References
- Bhasin S, Brito JP, Cunningham GR, et al. Testosterone therapy in men with hypogonadism: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2018;103(5):1715–1744. Source of the lists of conditions raising and lowering SHBG and of the recommendation to use equilibrium dialysis or calculation rather than a direct analogue immunoassay.
- Labcorp. Sex Hormone-binding Globulin, test 082016. Electrochemiluminescence immunoassay on Roche cobas; reference intervals 16.5–55.9 and 19.3–76.4 nmol/L in men, 24.6–122.0 and 17.3–125.0 nmol/L in women.
- Mayo Clinic Laboratories. Sex Hormone-Binding Globulin, Serum, test SHBG1. Immunoenzymatic assay; adult males 13.3–89.5 nmol/L, females 18–46 years 18.2–135.5, females 47–91 years 16.8–125.2, with Tanner-staged paediatric intervals.
- Yeap BB, Almeida OP, Hyde Z, et al. In men older than 70 years, total testosterone remains stable while free testosterone declines with age. The Health in Men Study. Eur J Endocrinol. 2007;156(5):585–594. SHBG increased with age, r = 0.2, P < 0.0001.
- van Velzen DM, et al. The influence of gender-affirming hormone therapy on serum concentrations of hormone-binding proteins. Eur J Endocrinol. 2025;192(5):614. SHBG at 12 months: +242% on oral estradiol with a GnRH agonist, +81% on oral estradiol with cyproterone acetate, +63% transdermal with a GnRH agonist, and −43% with masculinising testosterone therapy.
- Rosner W, et al. Effects of thyroid hormone on sex hormone-binding globulin gene expression in human cells. J Clin Endocrinol Metab. 1992;75(1):116–120 — SHBG was specifically and dose-responsively stimulated by T3, half-maximal at about 3 × 10⁻¹¹ mol/L free T3.
- Bourebaba Y, et al. Sex hormone binding globulin as a potential drug candidate for liver-related metabolic disorders treatment. Biomed Pharmacother. 2022. Describes SHBG as a homodimeric plasma glycoprotein of approximately 95 kDa whose weight depends partly on its glycosylation status, each subunit of 373 amino acids with three oligosaccharide side chains.
- UniProt Consortium. UniProtKB P04278 (SHBG): 402-residue precursor, 43,779 Da, homodimer, N-glycosylation at Asn380 and Asn396, O-glycosylation at Thr36, and an additional N-glycosylation site created by the D356N variant.
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.