Skip to main content

Longevity Library · Lab reference

Written by Daniel Tagge, MD
Updated October 10, 2026
Reviewed October 12, 2026
29 sources · Measured in nmol/L

System: Metabolism & Metabolomics

SHBGSex hormone-binding globulin · Free testosterone (calculated)

SHBG is the liver protein that decides how much testosterone and estradiol is free, and its level reads out the liver's metabolic state: in nested case-control studies of 359 women and 170 men, the highest SHBG quartile had about one tenth the odds of type 2 diabetes of the lowest, genetic variants that raise SHBG lower diabetes risk, and SHBG falls with liver fat and rises with age, thyroid excess and oral estrogen. In 149,436 UK Biobank men and nine pooled cohorts, higher SHBG went with higher, not lower, mortality, so no guideline names an optimal value; what SHBG changes is the reading of a total testosterone, which is why guidelines ask for a free testosterone whenever SHBG is abnormal or the total is borderline.

Sex hormone-binding globulin is made in the liver and carries testosterone and estradiol in the blood; only the unbound fraction is free to act. Laboratories print about 10 or 17 to 50 or 56 nmol/L in men and a wider, higher range in women. Low SHBG tracks liver fat and insulin resistance and predicts diabetes; high SHBG comes with age, thyroid excess and oral estrogen, and can hide a low free testosterone behind a normal total. The number matters most for what it does to the other numbers.

01 · The scale

Where the laboratory range and the evidence sit.

Risk sat at the low end for diabetes and at the high end for death. The highest SHBG quartile had about one tenth the odds of type 2 diabetes of the lowest in nested case-control studies of 359 women and 170 men, and alleles that raise SHBG lowered diabetes risk in 27,657 cases and 58,481 controls; in 149,436 UK Biobank men and in nine pooled cohorts of older men, lower SHBG went with lower all-cause mortality (hazard ratios 0.68 and 0.85 for the lowest versus the highest quintile); in 3,334 European men, a low calculated free testosterone carried symptoms whether or not the total was low; no trial has targeted an SHBG number.

020406080100

  1. Common laboratory reference range for adult men (laboratories print about 10 or 17 to 50 or 56); women's ranges run wider and higher

    17 to 56 nmol/L · Typical US laboratories

  2. Lowest quartile in the diabetes cohorts (medians 15.2 in men, 17.1 in women), where type 2 diabetes risk was highest; usually a liver-fat and insulin pattern

    0 to 20 nmol/L · Ding 2009; Peter 2010

  3. Below the 28.3 cutoff above which men's diabetes risk was about half in pooled prospective studies; in older men, the lowest-mortality quintile had a median of 20.6

    20 to 28 nmol/L · Ding 2006; Yeap 2024

  4. Upper quartiles of middle-aged men in the diabetes cohorts (top quartile median 38.0); about half the diabetes risk of the bottom

    28 to 50 nmol/L · Ding 2006; Ding 2009

  5. Common in older men, in women, with thyroid excess and oral estrogen; the highest quintile in nine cohorts of older men (median 68.3) had the highest mortality; a normal total testosterone can hide a low free testosterone here

    50 and above nmol/L · Yeap 2024; Yeap 2021; Bhasin 2018

  • Laboratory range
  • Evidence supports
  • Guideline goal or unstudied
  • Higher risk in studies
Scale in nmol/L. SHBG is reported in nmol/L everywhere. Free testosterone is reported in pg/mL in the United States and pmol/L elsewhere; multiply pg/mL by about 3.47 for pmol/L. Bands show where studies found something, not a recommendation for any one person.

02 · What it is

What SHBG measures.

SHBG is a glycoprotein made by the liver that binds androgens and estrogens with high (nanomolar) affinity and specificity, while albumin binds them loosely in bulk; between them they set the small unbound fraction of each hormone that is considered biologically active, which is why a review calls these proteins the primary gatekeepers of steroid action. The Endocrine Society's 2018 guideline treats SHBG as the reason a total testosterone can mislead, and asks for a free testosterone by equilibrium dialysis or an accurate formula whenever the total is near the lower limit of normal or a condition alters SHBG.1,2Grade B

What sets the level is mostly the liver, not the scale. In 225 adults at risk of diabetes measured by magnetic resonance before and during a 9-month lifestyle program, liver fat was the strongest determinant of circulating SHBG (ahead of sex), total body fat and visceral fat were not independent determinants once liver fat was in the model, and the change in liver fat tracked the change in SHBG (r = -0.37) while changes in total and visceral fat no longer did. In liver cells and transgenic mice, glucose and fructose reduced SHBG production by switching off the transcription factor HNF-4alpha through fat synthesis, which a 2015 review takes as evidence that liver fat content rather than body mass index, and perhaps not insulin itself, regulates SHBG. A low SHBG is, in short, a liver finding.3,4,5Grade C

The range depends on sex and age. In the same 225 adults, mean SHBG was 24.8 nmol/L in men and 56.2 in women. In 2,395 European men aged 40 to 79 followed for 4.4 years, SHBG rose by about 0.56 nmol/L a year while testosterone fell by about 0.1 nmol/L a year, so free testosterone fell faster than the total; in the UK Biobank authors' words, testosterone declines with age while SHBG increases.3,6,7Grade B

03 · Origins

Where it came from, and where it sits in the network.

SHBG is a liver protein with a long vertebrate history. Fish make it in the liver as we do, and in zebrafish it has a second, striking job: the gills carry large amounts of it, and it works as a portal that pulls sex steroids out of the surrounding water and controls their flux between blood and environment, a function that synthetic steroids in waste water, ethinylestradiol and progestins among them, can hijack. Salmon keep a second, poorly conserved SHBG gene expressed only in ovary, muscle and gill. In mammals below the primates the same gene is also switched on in the Sertoli cells of the testis under follicle-stimulating hormone, where its product, androgen-binding protein, shapes androgen action in the seminiferous tubules and epididymis; in humans that testicular expression is gone, and a germ-cell isoform accumulates in the sperm's acrosome instead. The thread through all of these animals is the same: a protein that binds sex steroids tightly and decides where they go.8,9Grade C

What SHBG does for the hormone system is buffer it. By binding androgens and estrogens with nanomolar affinity it sets their distribution in plasma, their clearance and their availability to tissues, holding most of the circulating testosterone and estradiol in reserve and leaving a small free fraction to act; it also accumulates outside the vessels in some tissues and inside certain epithelial cells, where it has effects of its own on androgen and estrogen action. In the network it therefore sits between the gonads and every steroid-responsive tissue, and between the liver and the sex hormones: testosterone, estradiol and thyroxine all stimulate its production in human liver cells, and it in turn sets the free hormone the tissues see, which is why Total testosterone (men) cannot be read without it.8,1,10Grade C

The reason SHBG reads as a metabolic marker is the liver's wiring. In 1988, human liver cells in culture were shown to cut SHBG production when exposed to insulin (from 65.0 to 46.8 nmol per million cells) and to prolactin, and insulin also blunted the rise that estradiol and thyroxine produce; the authors proposed insulin as a regulator of SHBG in people. In 2007 the mechanism was traced through sugar: in liver cells and transgenic mice, glucose and fructose lowered SHBG by driving fat synthesis, which lowered the transcription factor HNF-4alpha that switches the gene on. The current reading is that liver fat, more than insulin as such, is the proximate signal, with insulin and sugar upstream of the fat. Either way, the number on the report is the liver's summary of its own metabolic state, which is why a low SHBG sits beside Fasting insulin, HOMA-IR and ALT on this shelf, and why it predicts type 2 diabetes years ahead with genetic support.10,4,5,11Grade B

04 · The laboratory range

What a laboratory calls normal.

About 10 or 17 to 50 or 56 nmol/L in adult men, depending on the laboratory and assay; wider and higher in women; most laboratories print age- and sex-specific intervals.

The reference interval is a distribution of the people a laboratory tested, not a map of risk, and the cohorts drew their lines inside it. Across 43 prospective and cross-sectional studies (6,974 women and 6,427 men), men with SHBG above 28.3 nmol/L had about half the risk of type 2 diabetes of men at or below it (relative risk 0.48), and women above 60 nmol/L had about one fifth the risk of women at or below it (0.20). In the Women's Health Study and Physicians' Health Study II, the quartile medians ran from 17.1 to 55.8 nmol/L in postmenopausal women and from 15.2 to 38.0 in middle-aged men, and risk fell across every step.12,11Grade B

The same quartile labels sit at different numbers in different people. The top quartile of middle-aged physicians had a median SHBG of 38.0 nmol/L, while the top quintile of older men in nine pooled cohorts had a median of 68.3 and the bottom quintile 20.6, because SHBG rises with age. A value of 45 is high for a 35-year-old man with a fatty liver and ordinary for a 75-year-old; the number needs the age and the sex beside it before it needs a label.11,13,6Grade B

The Endocrine Society's guideline lists what moves SHBG outside the usual range. Lower: obesity, diabetes mellitus, glucocorticoids, some progestins and androgenic steroids, nephrotic syndrome, hypothyroidism, acromegaly, and variants in the SHBG gene. Higher: aging, HIV disease, cirrhosis and hepatitis, hyperthyroidism, some anticonvulsants, estrogens, and variants in the SHBG gene. Any of these is a reason to read a total testosterone through a free testosterone rather than on its own.2Grade A

05 · The evidence

What the studies support.

The clearest prospective evidence is a nested case-control study inside the Women's Health Study (359 postmenopausal women not using hormone therapy who developed type 2 diabetes and 359 controls), replicated in men of the Physicians' Health Study II (170 cases and 170 controls). Compared with the lowest quartile of SHBG, the multivariable odds ratios for diabetes in women were 0.16 in the second quartile, 0.04 in the third and 0.09 in the fourth; in men the highest quartile carried an odds ratio of 0.10. Two variants in the SHBG gene that shift the protein level by about 10 percent shifted diabetes risk in the matching directions, and the Mendelian randomization estimate was an odds ratio of 0.28 per standard deviation of SHBG in women and 0.29 in men. The authors concluded that low SHBG is a strong predictor of type 2 diabetes in both sexes and may have a causal role.11Grade B

Larger genetic studies confirm the direction and shrink the size. In 27,657 people with type 2 diabetes and 58,481 controls from 15 studies, each copy of the SHBG-raising allele at rs1799941 lowered the odds of diabetes (odds ratio 0.94), almost precisely what the allele's effect on SHBG (0.2 standard deviations per copy) and the SHBG difference between patients and controls (0.23 standard deviations lower) would predict; results were similar in men and women. In three Finnish cohorts and the DIAGRAM consortium (41,439 cases, 103,870 controls), the causal estimate was an odds ratio of 0.83 per standard deviation of SHBG, considerably weaker than the observational hazard ratio of 0.47 in prospective studies, which the authors read as the observational association being partly confounded rather than conferred directly by circulating SHBG. Across 43 studies the pooled prospective relative risks were 0.48 for men above 28.3 nmol/L and 0.20 for women above 60.14,15,12Grade A

The broadest genetic analysis found the same thing with a caution. Using 2,571 variants in 425,097 UK Biobank participants, genetically higher SHBG protected against type 2 diabetes across every model, with an odds ratio of 0.65 per standard deviation (about 30.3 nmol/L), and against fasting insulin and central fat; the genetic architecture of SHBG was nearly identical in men and women, and the liver was the most enriched tissue. The authors note that their SHBG instruments were not independent of testosterone, so the analysis cannot settle whether SHBG acts on its own or through the lower free testosterone that comes with it.16Grade B

For mortality the direction reverses, at least in older men. In 149,436 UK Biobank men aged 40 to 69 followed for 11 years (10,053 deaths), men in the lowest SHBG quintile had a lower rate of death from any cause than men in the highest (hazard ratio 0.68), from cardiovascular disease (0.70) and from cancer (0.80), after adjustment for age, waist circumference, medical conditions and testosterone; excluding deaths in the first two years did not change the result. In individual data from nine prospective cohorts with mass-spectrometry hormones (255,830 participant-years), lower SHBG again went with lower all-cause mortality (hazard ratio 0.85 for the lowest versus the highest quintile, medians 20.6 versus 68.3 nmol/L) and lower cardiovascular mortality (0.81). Both papers call for mechanistic studies and trials before anyone reads cause into it.7,13Grade B

In women the mortality signal points the same way and is thinner. A 2025 systematic review of 53 publications and 359,047 participants pooled the top against the bottom tertile of each hormone: higher SHBG was associated with higher all-cause mortality in women (hazard ratio 1.25, from three studies) and showed no association in men, while total testosterone had a U-shaped relationship with death in men. Three studies is a small base for the women's estimate, and the men's null result sits beside the two larger analyses above that found an association.17Grade C

SHBG's practical weight is in what it does to a testosterone result. Among 3,334 community-dwelling European men aged 40 to 79, 277 had a normal total testosterone (at least 10.5 nmol/L) with a low calculated free testosterone (below 220 pmol/L, about 64 pg/mL): they were older, in poorer health, had higher SHBG and LH, reported more sexual and physical symptoms, and had lower hemoglobin and bone ultrasound values than men normal on both. The 96 men with a low total but a normal free testosterone were younger and more obese, had lower SHBG and normal LH, and lacked the features of androgen deficiency. The authors concluded that free testosterone should be assessed in men with suspected hypogonadal symptoms; the Endocrine Society guideline makes that recommendation for a borderline total or an altered SHBG.18,2Grade B

In women, low SHBG is the biochemical face of polycystic ovary syndrome and its insulin resistance. Across 62 articles, SHBG was lower in women with PCOS than in controls (standardized mean difference -0.83) with high heterogeneity, and lower SHBG tracked the syndrome's metabolic disturbances. The 2023 international PCOS guideline asks clinicians to use total and free testosterone to assess biochemical hyperandrogenism, with free testosterone estimated by the calculated free androgen index (which divides testosterone by SHBG) or measured by equilibrium dialysis, on mass-spectrometry assays rather than direct immunoassays; it notes that assessment is very difficult on the combined oral contraceptive pill because the pill raises SHBG, and that the pill should be withdrawn for at least three months if biochemical androgens must be measured. In the UK Biobank genetics, a genetically higher testosterone raised the odds of PCOS in women (1.51 per standard deviation) and of type 2 diabetes (1.37).19,20,16Grade A

Two further associations are consistent but observational. In pregnancy, SHBG measured before conception or in early pregnancy predicted later gestational diabetes across seven studies of 1,947 women (247 with the diagnosis): a diagnostic odds ratio of 6.68, sensitivity 0.70, specificity 0.74, and an area under the curve of 0.78, with heterogeneity the authors said warrants more research. In fatty liver disease, across 16 studies of 13,721 men and 5,840 women, SHBG was lower in people with nonalcoholic fatty liver disease than in controls in both sexes, more strongly in women, and higher SHBG went with lower odds of the disease; total testosterone was lower in men with fatty liver (by 2.78 nmol/L) and the reverse in women.21,22Grade C

06 · What moves it

The levers with evidence behind them.

What is listed here has been tested. The size of each effect, and the size of the study, are in the sentence, so a small effect reads as a small effect.

  • Body weight and liver fat

    weight loss raises it in proportion; the rise follows liver fat more than total fat

    In 2,395 European men followed 4.4 years, weight loss went with a proportional rise and weight gain with a proportional fall in both testosterone and SHBG, and the authors concluded that obesity-associated changes in these hormones are reversible with weight reduction. In 225 adults through a 9-month lifestyle program, the people who lost the most liver fat gained the most SHBG (r = -0.37 for the two changes), and once liver fat was in the model, changes in total and visceral fat no longer predicted the change in SHBG. The lever is the liver; the scale is how most people move it.6,3Grade B

  • Sugar and refined carbohydrate

    glucose and fructose lower SHBG production in the laboratory; the human evidence is indirect

    In liver cells and transgenic mice, glucose and fructose reduced human SHBG production in a dose-dependent way by driving fat synthesis, which lowered the transcription factor HNF-4alpha that keeps the SHBG gene switched on; blocking fat synthesis prevented the effect. In the 225-adult lifestyle study, SHBG correlated negatively with total energy intake at baseline (r = -0.28) but not with the share of carbohydrate, fat or protein, and the trend toward a larger SHBG rise with a larger fall in carbohydrate intake did not reach significance. The mechanism is clear and the human trial is missing; what is established is that the diet changes that shrink a fatty liver raise SHBG.4,3Grade C

  • Oral estrogen and combined oral contraceptives

    raise it, often several-fold, which lowers free testosterone more than total

    Across 42 studies of 1,495 healthy women aged 18 to 40, combined oral contraceptives raised SHBG by a mean of 99.08 nmol/L, lowered total testosterone by 0.49 nmol/L and lowered free testosterone by 61 percent, about twice the fall in total, because of the SHBG rise; lower estrogen doses and second-generation progestins raised SHBG less. The Endocrine Society lists estrogens among the causes of a high SHBG, and the PCOS guideline treats androgen testing on the pill as unreliable for the same reason. The effect is a property of oral estrogen as a class, passing through the liver before the circulation; whether a given preparation is right for a given woman is a conversation with her physician, not a lab finding.23,2,20Grade A

  • Thyroid status

    thyroid excess raises it; thyroid deficiency lowers it

    The Endocrine Society's guideline lists hyperthyroidism among the conditions that raise SHBG and hypothyroidism among those that lower it, and the laboratory study that validated the free testosterone calculation chose hyperthyroid sera as its example of extremely high SHBG. An SHBG that has risen without explanation is a reason to look at the thyroid, and a thyroid result is part of reading any SHBG.2,24Grade B

  • Androgens, glucocorticoids and some progestins

    lower it

    The same guideline table lists glucocorticoids, some progestins and androgenic steroids among the causes of a low SHBG, together with obesity, diabetes, nephrotic syndrome, hypothyroidism and acromegaly, and lists some anticonvulsants, HIV disease, cirrhosis and hepatitis among the causes of a high one. A man taking testosterone or an anabolic steroid will have a lowered SHBG that reflects the drug; nothing on this page applies to his number.2Grade B

07 · Reading it well

Caveats, and what belongs with a physician.

There is no single range. Mean SHBG was 24.8 nmol/L in men and 56.2 in women in one metabolic cohort; it climbs about 0.56 nmol/L a year through middle age; combined oral contraceptives add about 99 nmol/L; and in a laboratory series spanning women off the pill, on the pill and pregnant, SHBG ran from 18.5 to 633 nmol/L. The same number is a different finding in a 30-year-old woman, a 45-year-old man with a fatty liver and a 75-year-old man, and a laboratory flag that ignores that is not a diagnosis.3,6,23,25Grade B

A calculated free testosterone is an estimate with assumptions inside it. The widely used formula (Vermeulen) solves the binding equations from total testosterone, SHBG and albumin; in the original validation it matched equilibrium dialysis in sera from hirsute women to hyperthyroid patients, except in pregnancy, where estradiol occupies SHBG and the calculation reads low. When 146 men and 183 women were remeasured by direct equilibrium dialysis with mass spectrometry, the Vermeulen calculation overestimated the measured value by a median ratio of 1.19 (range 0.90 to 1.47) but was the most robust of three methods, largely independent of SHBG, testosterone and albumin, while a newer allosteric model overestimated by a ratio of 2.05 and drifted with SHBG. The Endocrine Society says direct analog free testosterone immunoassays should not be used because they are inaccurate, and the free androgen index is unreliable for the same reason the original study showed: its ratio to the true free fraction changes with SHBG.24,26,2Grade A

There is no harmonized reference range for free testosterone. The guideline says so and encourages laboratories to establish their own; the 220 pmol/L (about 64 pg/mL) figure it quotes is the level below which sexual symptoms clustered in the European Male Aging Study, not a treatment threshold. In 145 healthy nonobese men measured by a standardized equilibrium dialysis method, the 2.5th, 50th and 97.5th percentiles of free testosterone were 66, 141 and 309 pg/mL across all ages and 120, 190 and 368 pg/mL at ages 19 to 39, and free testosterone fell with age, body mass index and SHBG; the authors ask for validation against outcomes before the intervals are used for diagnosis.2,27,18Grade B

SHBG is a steadier number than most hormones, but not a fixed one. In 20 men sampled over 12 months, the within-person coefficient of variation was 12.1 percent for SHBG against 10.9 percent for testosterone, while the between-person variation was 42.7 percent for SHBG against 21.3 percent for testosterone, which means people differ from each other far more than they differ from themselves, and a person's own trend carries more information than the population interval. In 132 Boston men sampled over 6 months, biological variation outweighed assay variation across the reproductive hormones including SHBG, and the authors concluded that one sample is generally not enough and more than three probably not warranted.28,29Grade B

SHBG can distort the testosterone it is meant to interpret. When 90 samples from women off the pill, on the pill and pregnant were run on six testosterone immunoassays against mass spectrometry, the four first-generation assays read inaccurately (correlations 0.61 to 0.86), and their error ran inversely with SHBG, so samples with very high or very low SHBG were the ones most likely to be wrong; the second-generation assays did not show the effect. A free testosterone calculated from a biased total inherits the bias. The guideline's preference for mass spectrometry applies with extra force when SHBG is abnormal.25,2Grade C

Most of what is known about SHBG is association, and the causal estimates are smaller. The Finnish analysis put the observational hazard ratio for diabetes at 0.47 per standard deviation and the genetic estimate at 0.83, and read the gap as confounding by the liver fat and insulin resistance that lower SHBG in the first place; the UK Biobank mortality paper says confirmation and determination of causality require mechanistic studies and prospective trials. A low SHBG is a signal to look at the liver, insulin and triglycerides; it is not itself a diagnosis, and raising it is not itself a goal that any trial has tested.15,7Grade B

See a physician

  • A low SHBG together with a fasting insulin, triglycerides, waist size or ALT that are also moving in the wrong direction, since that cluster is the liver-fat and insulin pattern that predicted type 2 diabetes in the cohorts.
  • A normal total testosterone with low desire, poor morning erections, erectile dysfunction or unexplained fatigue in a man whose SHBG is high, which is the pattern the guideline asks to be rechecked on a second fasting morning with a free testosterone by equilibrium dialysis or a validated calculation, SHBG and LH.
  • A low total testosterone with a low SHBG in a heavier man, before any conclusion about deficiency, because in the European cohort the free fraction was usually normal and the features of deficiency were absent.
  • In a woman, irregular or absent periods, acne, excess facial or body hair or scalp hair thinning together with a low SHBG or a high free androgen index, which is a PCOS assessment under the 2023 guideline; androgen testing on the combined pill is unreliable, and withdrawal is a physician's decision.
  • An SHBG that has risen without an explanation, especially with weight loss, palpitations, tremor, heat intolerance, a known liver disease or heavy alcohol use, since hyperthyroidism, hepatitis and cirrhosis raise it.
  • An SHBG result in pregnancy or while planning one; early-pregnancy SHBG predicted gestational diabetes in the studies but is not a screening test, and glucose testing remains the standard.
  • Anyone taking or considering testosterone, anabolic steroids, estrogen, a combined oral contraceptive, an anticonvulsant or a glucocorticoid, whose SHBG reflects the medication and needs the prescriber's reading, not a page's.

This page is education, not individual medical advice, and reading it creates no physician-patient relationship.

08 · Open questions

What the literature does not settle.

  • No guideline or trial defines an optimal SHBG, and no trial has raised or lowered SHBG on purpose to see whether anything improves; every line on this page marks where risk sat in cohorts, and any preferred zone inside the range is a judgment.
  • The two largest findings point in opposite directions: low SHBG predicts type 2 diabetes with genetic support for causation, while in older men the lowest SHBG quintile had the lowest mortality and the highest quintile the highest. Whether high SHBG in old age marks frailty, liver disease, low free hormone or something protective about low SHBG is unresolved, and the women's mortality data rest on three studies.
  • How much SHBG does on its own is open. The genetic estimates for diabetes (odds ratios of 0.83 and 0.65 per standard deviation) are far smaller than the observational ones (hazard ratio 0.47), the SHBG instruments are not independent of testosterone, and whether SHBG acts through a cell receptor or on hepatic glucose production, rather than only as a carrier, has not been settled in people.
  • Calculated free testosterone methods disagree by 20 to 100 percent against direct dialysis, there is no harmonized free testosterone reference range, and the 64 pg/mL (220 pmol/L) figure is a symptom association in one European cohort, not a tested treatment threshold; whether free testosterone should replace the total as the first test is argued.
  • Whether diet composition, sugar and fructose in particular, changes SHBG in people independently of weight and liver fat has been shown in cells and mice and not in a human trial.

10

Questions

11 · Sources

29 sources, read in full.

Numbered in the order they appear. Each line says what this page relies on from the paper; the link opens the record at PubMed or the publisher.

  1. 1.

    Hammond GL. Plasma steroid-binding proteins: primary gatekeepers of steroid hormone action. Journal of Endocrinology. 2016. PMID 27113851. DOI 10.1530/JOE-16-0070.

    Review · SHBG and corticosteroid-binding globulin bind steroids with nanomolar affinity and specificity, albumin loosely in bulk; together they set the free fraction considered biologically active, and SHBG production by the liver varies with physiological and disease states.

  2. 2.

    Bhasin S, Brito JP, Cunningham GR, Hayes FJ, Hodis HN, Matsumoto AM, Snyder PJ, Swerdloff RS, Wu FC, Yialamas MA. Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline. Journal of Clinical Endocrinology and Metabolism. 2018. PMID 29562364. DOI 10.1210/jc.2018-00229.

    Guideline · Obtain a free testosterone by equilibrium dialysis or an accurate formula when total testosterone is near the lower limit of normal or a condition alters SHBG; direct analog free testosterone immunoassays should not be used; no harmonized free testosterone range exists; the guideline's table lists conditions that lower SHBG (obesity, diabetes, glucocorticoids, some progestins, androgens, nephrotic syndrome, hypothyroidism, acromegaly) and raise it (aging, HIV, cirrhosis and hepatitis, hyperthyroidism, some anticonvulsants, estrogens).

  3. 3.

    Peter A, Kantartzis K, Machann J, Schick F, Staiger H, Machicao F, Schleicher E, Fritsche A, Häring HU, Stefan N. Relationships of circulating sex hormone-binding globulin with metabolic traits in humans. Diabetes. 2010. PMID 20841609. DOI 10.2337/db10-0179.

    Prospective cohort · In 225 adults measured by MR before and during a 9-month lifestyle intervention, liver fat was the strongest determinant of SHBG (total and visceral fat were not independent), change in liver fat tracked change in SHBG (r = -0.37), SHBG correlated with insulin sensitivity until liver fat was adjusted for, and with fasting glucose even after; mean SHBG 24.8 nmol/L in men and 56.2 in women.

  4. 4.

    Selva DM, Hogeveen KN, Innis SM, Hammond GL. Monosaccharide-induced lipogenesis regulates the human hepatic sex hormone-binding globulin gene. Journal of Clinical Investigation. 2007. PMID 17992261. DOI 10.1172/JCI32249.

    Laboratory method study · In transgenic mice and HepG2 liver cells, glucose and fructose reduced human SHBG production dose-dependently by lowering HNF-4alpha through lipogenesis; inhibiting lipogenesis prevented it.

  5. 5.

    Simó R, Sáez-López C, Barbosa-Desongles A, Hernández C, Selva DM. Novel insights in SHBG regulation and clinical implications. Trends in Endocrinology and Metabolism. 2015. PMID 26044465. DOI 10.1016/j.tem.2015.05.001.

    Review · Liver fat content rather than body mass index is a strong determinant of circulating SHBG; evidence suggests insulin might not regulate SHBG production, while proinflammatory cytokines and diet do.

  6. 6.

    Camacho EM, Huhtaniemi IT, O'Neill TW, Finn JD, Pye SR, Lee DM, et al.; EMAS Group. Age-associated changes in hypothalamic-pituitary-testicular function in middle-aged and older men are modified by weight change and lifestyle factors: longitudinal results from the European Male Ageing Study. European Journal of Endocrinology. 2013. PMID 23425925. DOI 10.1530/EJE-12-0890.

    Prospective cohort · In 2,395 men aged 40 to 79 followed 4.4 years, SHBG rose 0.56 nmol/L a year while testosterone fell 0.1 nmol/L a year; weight loss went with a proportional rise and weight gain with a proportional fall in testosterone and SHBG.

  7. 7.

    Yeap BB, Marriott RJ, Antonio L, Chan YX, Raj S, Dwivedi G, et al.. Serum Testosterone is Inversely and Sex Hormone-binding Globulin is Directly Associated with All-cause Mortality in Men. Journal of Clinical Endocrinology and Metabolism. 2021. PMID 33059368. DOI 10.1210/clinem/dgaa743.

    Prospective cohort · In 149,436 UK Biobank men aged 40 to 69 followed 11 years (10,053 deaths), the lowest SHBG quintile had lower all-cause (hazard ratio 0.68), cardiovascular (0.70) and cancer (0.80) mortality than the highest, adjusted for age, waist, conditions and testosterone; causality requires mechanistic studies and trials.

  8. 8.

    Hammond GL. Diverse roles for sex hormone-binding globulin in reproduction. Biology of Reproduction. 2011. PMID 21613632. DOI 10.1095/biolreprod.111.092593.

    Review · SHBG controls plasma distribution, clearance and bioavailability of sex steroids and accumulates in extravascular compartments and epithelial cells; subprimate Sertoli cells make it as androgen-binding protein under FSH, human germ cells make an acrosomal isoform, fish make it in liver with a gill function, and salmon keep a second gene expressed in ovary, muscle and gill.

  9. 9.

    Miguel-Queralt S, Hammond GL. Sex hormone-binding globulin in fish gills is a portal for sex steroids breached by xenobiotics. Endocrinology. 2008. PMID 18483148. DOI 10.1210/en.2008-0384.

    Laboratory method study · Zebrafish SHBG, made in the liver and abundant in gill filaments, controls the flux of sex steroids across the gills, rapidly removing its ligands from water; synthetic steroids such as ethinylestradiol and progestins bind it with high affinity and hijack the portal.

  10. 10.

    Plymate SR, Matej LA, Jones RE, Friedl KE. Inhibition of sex hormone-binding globulin production in the human hepatoma (Hep G2) cell line by insulin and prolactin. Journal of Clinical Endocrinology and Metabolism. 1988. PMID 2842359. DOI 10.1210/jcem-67-3-460.

    Laboratory method study · In Hep G2 human liver cells, insulin and prolactin each cut SHBG production from 65.0 to 46.8 nmol per million cells and insulin blunted estradiol- and thyroxine-stimulated production; testosterone, estradiol and thyroxine stimulated it.

  11. 11.

    Ding EL, Song Y, Manson JE, Hunter DJ, Lee CC, Rifai N, Buring JE, Gaziano JM, Liu S. Sex hormone-binding globulin and risk of type 2 diabetes in women and men. New England Journal of Medicine. 2009. PMID 19657112. DOI 10.1056/NEJMoa0804381.

    Prospective cohort · In nested case-control studies of 359 women (Women's Health Study) and 170 men (Physicians' Health Study II), the highest SHBG quartile had odds ratios for type 2 diabetes of 0.09 and 0.10 versus the lowest; SHBG gene variants shifting the level by 10 percent shifted risk accordingly; Mendelian randomization odds ratios 0.28 (women) and 0.29 (men) per standard deviation; quartile medians 17.1 to 55.8 nmol/L in women and 15.2 to 38.0 in men.

  12. 12.

    Ding EL, Song Y, Malik VS, Liu S. Sex differences of endogenous sex hormones and risk of type 2 diabetes: a systematic review and meta-analysis. JAMA. 2006. PMID 16537739. DOI 10.1001/jama.295.11.1288.

    Systematic review and meta-analysis · Across 43 studies (6,974 women, 6,427 men), women with SHBG above 60 nmol/L had an 80 percent lower risk of type 2 diabetes (relative risk 0.20) and men above 28.3 nmol/L a 52 percent lower risk (0.48); the inverse association was stronger in women.

  13. 13.

    Yeap BB, Marriott RJ, Dwivedi G, Adams RJ, Antonio L, Ballantyne CM, et al.. Associations of Testosterone and Related Hormones With All-Cause and Cardiovascular Mortality and Incident Cardiovascular Disease in Men: Individual Participant Data Meta-analyses. Annals of Internal Medicine. 2024. PMID 38739921. DOI 10.7326/M23-2781.

    Systematic review and meta-analysis · In individual data from nine prospective cohorts with mass-spectrometry hormones (255,830 participant-years), lower SHBG was associated with lower all-cause mortality (quintile 1 versus 5 medians 20.6 versus 68.3 nmol/L; hazard ratio 0.85) and lower cardiovascular mortality (0.81).

  14. 14.

    Perry JR, Weedon MN, Langenberg C, Jackson AU, Lyssenko V, Sparsø T, et al.; MAGIC. Genetic evidence that raised sex hormone binding globulin (SHBG) levels reduce the risk of type 2 diabetes. Human Molecular Genetics. 2010. PMID 19933169. DOI 10.1093/hmg/ddp522.

    Mendelian randomization · In 27,657 type 2 diabetes patients and 58,481 controls from 15 studies, the SHBG-raising allele at rs1799941 lowered diabetes odds (0.94 per copy), matching the expected 0.92 given 0.2 SD higher SHBG per allele and 0.23 SD lower SHBG in patients; similar in men and women.

  15. 15.

    Wang Q, Kangas AJ, Soininen P, Tiainen M, Tynkkynen T, Puukka K, et al.. Sex hormone-binding globulin associations with circulating lipids and metabolites and the risk for type 2 diabetes: observational and causal effect estimates. International Journal of Epidemiology. 2015. PMID 26050255. DOI 10.1093/ije/dyv093.

    Mendelian randomization · In 6,475 young Finnish adults, higher SHBG went with a better cardiometabolic profile and predicted insulin resistance at 6 years; the causal estimate for type 2 diabetes (41,439 cases, 103,870 controls) was an odds ratio of 0.83 per SD, much weaker than the observational hazard ratio of 0.47, suggesting partial confounding.

  16. 16.

    Ruth KS, Day FR, Tyrrell J, Thompson DJ, Wood AR, Mahajan A, et al.. Using human genetics to understand the disease impacts of testosterone in men and women. Nature Medicine. 2020. PMID 32042192. DOI 10.1038/s41591-020-0751-5.

    Mendelian randomization · In 425,097 UK Biobank participants with 2,571 variants, genetically higher SHBG lowered type 2 diabetes odds (0.65 per SD, about 30.3 nmol/L) across all models; SHBG genetics were concordant between sexes with the liver the most enriched tissue; genetically higher testosterone raised PCOS (1.51) and diabetes (1.37) in women and lowered diabetes in men (0.86); SHBG instruments were not independent of testosterone.

  17. 17.

    Raeisi-Dehkordi H, Amiri M, Beigrezaei S, Quezada-Pinedo HG, Khatami F, Alijla F, et al.. Endogenous Sex Steroid Hormones, Sex Hormone-Binding Globulin, and Risk of All-Cause and Cause-Specific Mortality: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies. Journal of Clinical Endocrinology and Metabolism. 2025. PMID 40333346. DOI 10.1210/clinem/dgaf262.

    Systematic review and meta-analysis · Across 53 publications and 359,047 participants, higher SHBG (top versus bottom tertile) was associated with higher all-cause mortality in women (hazard ratio 1.25, three studies) and showed no association in men; total testosterone had a U-shaped relationship with mortality in men.

  18. 18.

    Antonio L, Wu FC, O'Neill TW, Pye SR, Ahern TB, Laurent MR, et al.. Low Free Testosterone Is Associated with Hypogonadal Signs and Symptoms in Men with Normal Total Testosterone. Journal of Clinical Endocrinology and Metabolism. 2016. PMID 26909800. DOI 10.1210/jc.2015-4106.

    Cross-sectional study · In 3,334 men aged 40 to 79, 277 with normal total (at least 10.5 nmol/L) but low calculated free testosterone (below 220 pmol/L) were older, had higher SHBG and LH, more sexual and physical symptoms and lower hemoglobin and bone ultrasound values; 96 with low total but normal free testosterone were younger, more obese, had lower SHBG and lacked features of deficiency.

  19. 19.

    Deswal R, Yadav A, Dang AS. Sex hormone binding globulin - an important biomarker for predicting PCOS risk: A systematic review and meta-analysis. Systems Biology in Reproductive Medicine. 2018. PMID 29227165. DOI 10.1080/19396368.2017.1410591.

    Systematic review and meta-analysis · Across 62 articles, SHBG was lower in women with PCOS than in controls (standardized mean difference -0.83) with high heterogeneity; lower SHBG tracked the syndrome's metabolic disturbances.

  20. 20.

    Teede HJ, Tay CT, Laven J, Dokras A, Moran LJ, Piltonen TT, et al.; International PCOS Network. Recommendations from the 2023 International Evidence-based Guideline for the Assessment and Management of Polycystic Ovary Syndrome. Human Reproduction. 2023. PMID 37580037. DOI 10.1093/humrep/dead156.

    Guideline · Use total and free testosterone to assess biochemical hyperandrogenism, with free testosterone estimated by the calculated free androgen index or measured by equilibrium dialysis, on LC-MS/MS assays rather than direct immunoassays; assessment is very difficult on the combined oral contraceptive pill because it raises SHBG, and the pill should be withdrawn for at least three months if androgens must be measured.

  21. 21.

    Luo X, Cai WY, Song JY, Duan CC, Wu W, Man YJ, Wu XK, Xu J. Predictive value of circulating sex hormone-binding globulin for gestational diabetes: a meta-analysis. Biomarkers in Medicine. 2021. PMID 34289735. DOI 10.2217/bmm-2020-0487.

    Systematic review and meta-analysis · Across seven studies of 1,947 women (247 with gestational diabetes), preconception or early-pregnancy SHBG had a diagnostic odds ratio of 6.68, sensitivity 0.70, specificity 0.74 and area under the curve 0.78, with heterogeneity warranting more research.

  22. 22.

    Jaruvongvanich V, Sanguankeo A, Riangwiwat T, Upala S. Testosterone, Sex Hormone-Binding Globulin and Nonalcoholic Fatty Liver Disease: a Systematic Review and Meta-Analysis. Annals of Hepatology. 2017. PMID 28425408. DOI 10.5604/16652681.1235481.

    Systematic review and meta-analysis · Across 16 observational studies of 13,721 men and 5,840 women, SHBG was lower in people with nonalcoholic fatty liver disease in both sexes, more strongly in women, and higher SHBG went with lower odds of the disease; total testosterone was 2.78 nmol/L lower in men with the disease.

  23. 23.

    Zimmerman Y, Eijkemans MJ, Coelingh Bennink HJ, Blankenstein MA, Fauser BC. The effect of combined oral contraception on testosterone levels in healthy women: a systematic review and meta-analysis. Human Reproduction Update. 2014. PMID 24082040. DOI 10.1093/humupd/dmt038.

    Systematic review and meta-analysis · Across 42 studies of 1,495 healthy women aged 18 to 40, combined oral contraceptives raised SHBG by 99.08 nmol/L, lowered total testosterone by 0.49 nmol/L and free testosterone by 61 percent; lower estrogen doses and second-generation progestins raised SHBG less.

  24. 24.

    Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. Journal of Clinical Endocrinology and Metabolism. 1999. PMID 10523012. DOI 10.1210/jcem.84.10.6079.

    Laboratory method study · Free testosterone calculated from total testosterone and immunoassayed SHBG matched equilibrium dialysis across sera from hirsute women to hyperthyroid patients, except in pregnancy; analog free testosterone immunoassays and the free androgen index varied with SHBG and are not reliable indexes.

  25. 25.

    Heijboer AC, Savelkoul E, Kruit A, Endert E, Blankenstein MA. Inaccurate First-Generation Testosterone Assays Are Influenced by Sex Hormone-Binding Globulin Concentrations. Journal of Applied Laboratory Medicine. 2016. PMID 33626789. DOI 10.1373/jalm.2016.020065.

    Laboratory method study · In 90 samples from women off the pill, on the pill and pregnant (SHBG 18.5 to 633 nmol/L), four first-generation testosterone immunoassays were inaccurate against mass spectrometry (R 0.61 to 0.86) and their error ran inversely with SHBG; second-generation assays did not show the effect.

  26. 26.

    Fiers T, Wu F, Moghetti P, Vanderschueren D, Lapauw B, Kaufman JM. Reassessing Free-Testosterone Calculation by Liquid Chromatography-Tandem Mass Spectrometry Direct Equilibrium Dialysis. Journal of Clinical Endocrinology and Metabolism. 2018. PMID 29618085. DOI 10.1210/jc.2017-02360.

    Laboratory method study · Against direct equilibrium dialysis with mass spectrometry in 146 men and 183 women, the Vermeulen calculation overestimated free testosterone by a median ratio of 1.19 (0.90 to 1.47) but was the most robust and largely independent of SHBG, testosterone and albumin; the Zakharov allosteric model overestimated by 2.05 and drifted with SHBG.

  27. 27.

    Jasuja R, Pencina KM, Spencer DJ, Peng L, Privat F, Dhillo W, et al.. Reference intervals for free testosterone in adult men measured using a standardized equilibrium dialysis procedure. Andrology. 2023. PMID 36251328. DOI 10.1111/andr.13310.

    Laboratory method study · In 145 healthy nonobese men measured by standardized equilibrium dialysis with mass spectrometry, free testosterone percentiles (2.5th, 10th, 50th, 90th, 97.5th) were 66, 91, 141, 240 and 309 pg/mL in all men and 120, 128, 190, 274 and 368 pg/mL at ages 19 to 39; free testosterone fell with age, body mass index and SHBG.

  28. 28.

    Valero-Politi J, Fuentes-Arderiu X. Within- and between-subject biological variations of follitropin, lutropin, testosterone, and sex-hormone-binding globulin in men. Clinical Chemistry. 1993. PMID 8353962.

    Laboratory method study · In 20 men sampled over 12 months, within-subject coefficients of variation were 12.1 percent for SHBG and 10.9 percent for serum testosterone; between-subject coefficients were 42.7 and 21.3 percent.

  29. 29.

    Brambilla DJ, O'Donnell AB, Matsumoto AM, McKinlay JB. Intraindividual variation in levels of serum testosterone and other reproductive and adrenal hormones in men. Clinical Endocrinology. 2007. PMID 18052942. DOI 10.1111/j.1365-2265.2007.02976.x.

    Laboratory method study · In 132 men sampled up to six times over 6 months, biological variation in testosterone, SHBG and other reproductive hormones exceeded assay variation and day-to-day exceeded long-term variation; one sample is generally not enough to characterize a man's level, more than three probably unwarranted.

Related

Keep reading.

Reading a number well is the start of the work, not the end of it. If you would like a physician to read yours over time, Explore working with Dr. Tagge.