Free Testosterone vs Total Testosterone. Why Your Number Might Be Lying to You.

A landmark bioavailability analysis in JCEM established that free testosterone, not total T, best correlates with androgen-related symptoms and cardiovascular endpoints in men over 40.

The testosterone number on your lab report is the total. It measures all testosterone in your blood: the testosterone bound to sex hormone binding globulin (SHBG), the testosterone bound to albumin, and the small fraction that is unbound. The unbound fraction is free testosterone. It is the only fraction that can enter your cells and produce the effects testosterone is responsible for.

A man can have a total testosterone of 520 ng/dL, firmly within the normal reference range, and have a free testosterone that is meaningfully low. He feels exactly like a man with low testosterone. His physician looks at the total T and says things look fine. They are not fine. The free fraction is the one that runs the system.

The Mechanism

Testosterone circulates in three forms. Approximately 44 percent is tightly bound to SHBG and essentially biologically unavailable. Approximately 54 percent is loosely bound to albumin and, to a small degree, other proteins. Albumin-bound testosterone is more readily released at the tissue level and is considered partially bioavailable. The remaining 2 to 3 percent circulates unbound: this is free testosterone, and it is the fraction that diffuses directly into cells, binds to androgen receptors, and exerts biological effects on muscle, bone, the central nervous system, and the cardiovascular system.

SHBG is the primary regulator of this equilibrium. SHBG is a glycoprotein produced by the liver that binds sex hormones with high affinity. When SHBG is high, a larger proportion of total testosterone is sequestered in the biologically inactive bound form, and free testosterone falls even if total testosterone production is unchanged. This is the mechanism behind the most common scenario in which a man has a normal total T and still experiences the full clinical picture of low testosterone.

SHBG rises with several conditions relevant to middle-aged men. It rises with age: the Massachusetts Male Aging Study, which followed a population-based cohort of men aged 40 to 70 over a decade, documented that free testosterone declines at approximately 2 percent per year in middle-aged men, a rate faster than the decline in total testosterone, because SHBG rises simultaneously. It rises with hepatic dysfunction, because SHBG is synthesized by the liver and liver disease can paradoxically elevate circulating levels. It rises with hyperthyroidism, because thyroid hormone stimulates SHBG production. It rises with use of certain medications, including some antiepileptic drugs.

The direction reverses with insulin resistance and visceral adiposity, which suppress SHBG. This creates a different but equally problematic scenario: the obese man with insulin resistance may have low SHBG, apparently normal free testosterone by calculation, and still have functional androgen insufficiency because of peripheral factors including aromatase activity, which converts testosterone to estrogen within adipose tissue.

The 1999 bioavailability analysis by Vermeulen, Verdonck, and Kaufman, published in the Journal of Clinical Endocrinology and Metabolism, provided the equations most commonly used to calculate free testosterone from total testosterone and SHBG. The study also established that calculated free testosterone correlates reasonably well with directly measured free testosterone in most clinical ranges, but diverges at extremes of SHBG or total T. For men with substantially elevated SHBG, direct measurement of free testosterone is more accurate than the calculated estimate. 4 / Promising

The HPG axis connects this testosterone biology to brain function and systemic metabolic health. The hypothalamus releases gonadotropin-releasing hormone (GnRH) in pulses. The pituitary responds by releasing luteinizing hormone (LH), which signals the Leydig cells of the testes to produce testosterone. This feedback loop is exquisitely sensitive to cortisol. Chronic cortisol elevation, the physiological result of sustained psychological or occupational stress, suppresses GnRH release at the hypothalamic level. LH falls. Testosterone production falls. Total T decreases. If SHBG rises simultaneously in response to cortisol’s metabolic effects, free T decreases proportionally more. The result is the chronically stressed, high-functioning man with a total testosterone of 410, an SHBG of 65, a free testosterone in the bottom quartile of the reference range, and a clinical picture that his standard labs do not explain.

What the Evidence Shows

The Vermeulen et al. analysis in JCEM (1999) established the foundational clinical framework: free testosterone, not total T, best correlates with androgen-related symptoms and outcomes in men over 40. This has been replicated across multiple subsequent observational studies. 4 / Promising

The European Male Aging Study (EMAS), which enrolled 3,369 men aged 40 to 79 across eight European centers and published primary findings in the New England Journal of Medicine in 2010 (Wu and colleagues), examined the relationship between testosterone levels and symptoms. A critical finding: the sexual symptoms most specific to androgen deficiency, including reduced frequency of morning erections and reduced sexual desire, correlated most strongly with free testosterone levels, not total testosterone. Men with total testosterone in the lower normal range had substantially different symptom burdens depending on their SHBG and free testosterone values. The EMAS study identified a free testosterone threshold below approximately 220 pmol/L (or roughly 63 pg/mL on most assays) as the level at which androgen-deficiency symptoms became more prevalent. 4 / Promising

The cardiovascular associations of low testosterone are observational but consistent across large datasets. A meta-analysis by Ruige and colleagues, published in the Journal of the American College of Cardiology in 2011, reviewed 11 prospective studies and found that men with low testosterone had approximately 36 percent higher all-cause mortality and significantly elevated rates of cardiovascular events compared to men with normal testosterone, after adjustment for traditional cardiovascular risk factors. Critically, the studies that measured free testosterone showed stronger associations than those measuring total testosterone only, consistent with the EMAS finding that free T is the more biologically relevant variable. 3 / Early

The causal direction of the testosterone-cardiovascular relationship is not fully established. Low testosterone is associated with insulin resistance, visceral adiposity, endothelial dysfunction, and dyslipidemia, all of which independently drive cardiovascular disease. Whether the low testosterone causes these conditions, or whether a shared upstream cause produces both, remains the central unresolved question. The testosterone replacement therapy trial evidence, including the TRAVERSE trial published in the New England Journal of Medicine in 2023 by Lincoff and colleagues, found that testosterone replacement in men with hypogonadism and high cardiovascular risk was noninferior to placebo for major adverse cardiovascular events over a median 33 months of follow-up. The TRAVERSE trial did not demonstrate cardiovascular benefit of testosterone replacement, but also did not show harm in a population that had historically raised concern. What the trial evidence does not yet support is using testosterone replacement specifically as a cardiovascular risk reduction strategy. 4 / Promising

What is more clearly established is that the upstream metabolic state, including insulin resistance, visceral adiposity, sleep disruption, and sustained cortisol load, suppresses testosterone through the mechanisms described above and independently drives cardiovascular risk. Addressing those upstream factors is both the most evidence-based cardiovascular intervention and the most evidence-based approach to improving free testosterone in men without primary testicular failure.

SHBG: Why the Binding Protein Changes the Clinical Picture

Sex hormone-binding globulin is a liver-produced protein that binds testosterone with high affinity and transports it in the bloodstream. Bound testosterone is biologically inactive; only the unbound fraction can enter cells and activate androgen receptors. Understanding what raises and lowers SHBG is as clinically important as the total testosterone level itself, because two men with identical total testosterone values can have dramatically different free testosterone concentrations depending on their SHBG.

SHBG increases with age, rising approximately 1 to 2 percent per year after age 40 in most studies. The mechanism involves reduced hepatic clearance of SHBG and reduced suppression of its production as testosterone naturally declines. The practical consequence: a man of 35 and a man of 57 can have the same total testosterone of 480 ng/dL, yet the older man’s free fraction may be 40 percent lower because his SHBG is substantially elevated. His androgen receptor exposure is not equivalent. His androgen-related symptoms are not equivalent. Treating both men identically because their total T matches is a clinical error that the EMAS study (Wu et al., NEJM 2010) documented explicitly, the free testosterone threshold, not the total, was what differentiated symptomatic from asymptomatic men.

Several medical conditions predictably elevate SHBG: hyperthyroidism (thyroid hormone directly stimulates hepatic SHBG synthesis), early liver disease, anticonvulsant medications including phenytoin and carbamazepine, and HIV-associated inflammation. For a man whose SHBG is elevated by one of these conditions, free testosterone will be low even when total testosterone falls within the reference range.

Insulin resistance and obesity predictably lower SHBG. Elevated insulin suppresses hepatic SHBG synthesis through signaling pathways involving hepatic nuclear factor-4-alpha. This creates a clinical paradox: the metabolically dysregulated man with insulin resistance and visceral adiposity may have SHBG in the 15 to 25 nmol/L range, meaning a larger fraction of his total testosterone is theoretically bioavailable, yet the aromatase activity of visceral adipose tissue converts that bioavailable testosterone to estradiol simultaneously. The net clinical picture can be low androgen effect despite a total T in the low-normal range, because aromatization is consuming the free fraction faster than it can be replaced. Measuring free testosterone alongside SHBG makes this mechanism visible in a way that total testosterone alone cannot. 4 / Promising

What to Do This Week

1. If you have symptoms suggesting low testosterone and your total T has been measured as normal, ask specifically for free testosterone and SHBG at your next lab draw. If your SHBG is above 50 nmol/L in the context of a total testosterone below 500 ng/dL, your free fraction is almost certainly in the lower range of normal or below it.

2. Ask for direct measurement of free testosterone rather than the calculated estimate when clinically meaningful. Calculated free T uses the Vermeulen equation and is a reasonable estimate in most clinical ranges, but direct equilibrium dialysis measurement is more accurate at the extremes of SHBG and is worth requesting when the calculation falls near decision-relevant thresholds.

3. Address the upstream factors before any decision about testosterone replacement. Insulin resistance, chronic sleep deficit, and sustained cortisol elevation all suppress testosterone through the HPG axis. A man whose fasting insulin is 16, who sleeps 5.5 hours per night, and who is running at sustained occupational cortisol load has three modifiable suppressors of his testosterone axis that are treatable without medication. Addressing them changes the trajectory.

4. Reduce visceral adiposity as a direct testosterone intervention. Adipose tissue, particularly visceral fat, expresses aromatase, the enzyme that converts testosterone to estradiol. Waist circumference reduction has a documented and dose-dependent effect on free testosterone in men with obesity-associated hypogonadism, with more reliable and sustained results than many pharmacological approaches.

5. Do not start or stop testosterone replacement without a complete panel including free testosterone, SHBG, LH, and FSH. The decision about replacement therapy is more nuanced than a single total T number, and the wrong diagnosis, specifically starting replacement in a man with secondary hypogonadism from a reversible cause, produces partial results and eliminates the HPG feedback signal, which further suppresses endogenous production.

The total testosterone on your lab report is the beginning of the conversation, not the end of it. The number that explains your clinical picture is the free fraction, and that number requires a specific order and a clinician willing to interpret it in the context of SHBG, symptoms, and the upstream metabolic state.

Why the Symptom Picture Matters

Clinical guidelines for testosterone assessment, including the 2018 Endocrine Society Clinical Practice Guideline by Bhasin and colleagues, specify that biochemical testing should occur only in men with symptoms and signs that are consistent with androgen deficiency, not as a routine screening test in asymptomatic men. The symptoms most specific to low free testosterone in middle-aged men, based on the EMAS findings, are: reduced frequency of morning erections, reduced sexual desire, and erectile dysfunction. Less specific symptoms include fatigue, depressed mood, reduced muscle mass, difficulty concentrating, and increased body fat, all of which have multiple potential causes.

The specificity of this symptom cluster matters because a man who meets both the biochemical criterion (low free testosterone confirmed on two separate morning draws) and the symptomatic criterion has a different clinical situation than a man who has borderline low free testosterone and symptoms that are attributable to poor sleep, chronic stress, or insulin resistance. In the latter case, the upstream causes of both the low testosterone and the symptoms should be addressed before the question of testosterone replacement is raised.

Morning timing of the blood draw is not a minor detail. Testosterone follows a circadian rhythm with peak levels in the early morning, typically between 7 and 10 a.m., and trough levels in the afternoon and evening. A total testosterone measured at 2 p.m. can be 20 to 30 percent lower than the same man’s morning value. If there is clinical uncertainty about whether total T is low, the morning timing ensures the measurement represents the physiological peak rather than a trough. Both the Endocrine Society and the American Urological Association guidelines specify morning measurement, before 10 a.m., as the appropriate standard. This is a measurement detail that significantly affects the result and is not always followed in practice.

The Signal Check is fifteen questions mapping the male cardiovascular risk pattern, including the physiological domains most commonly missed in standard screenings. It produces a specific starting point for your next clinical conversation.