Free Testosterone, Nutrition, and Fasting: What Your Labs Actually Mean

What a Free Testosterone Result Actually Measures

Testosterone travels in blood mostly attached to proteins. The Depo-Testosterone label puts it at about 98% bound and about 2% free, and says the amount of binding globulin in plasma determines how testosterone splits between the bound and free forms [1]. Free and nonspecifically bound hormone levels generally reflect the clinical situation more accurately than total levels do [9].

Free testosterone on a lab report should say how it was measured, because the methods do not agree. In a comparison across serum samples with binding globulin ranging from low to very high, free testosterone calculated from total testosterone and SHBG came out almost identical to equilibrium dialysis, except during pregnancy [9]. The direct analog immunoassay for free testosterone did not: its values were only a fraction of the dialysis values, and that fraction changed with SHBG, so the authors called it an unreliable index [9].

What Counts as a Low Free Testosterone

The European Male Ageing Study assessed 3,369 men aged 40 to 79 at eight European centers and calculated free testosterone using the Vermeulen formula [4]. The authors defined late-onset hypogonadism as at least three sexual symptoms together with a total testosterone below 11 nmol/L (3.2 ng/mL) and a free testosterone below 220 pmol/L (64 pg/mL) [4]. The three sexual symptoms (poor morning erections, low sexual desire and erectile dysfunction) were the ones that clustered with low testosterone; depression, fatigue and reduced physical vigor were related to testosterone levels but did not form the same syndrome [4].

Total testosterone ranges were harmonized across four cohorts totaling 9,054 men by cross-calibrating assays to a CDC reference method. In healthy nonobese men aged 19 to 39, the 2.5th to 97.5th percentile range was 264 to 916 ng/dL [10]. That work is about total testosterone, so its figures should not be used to judge a free testosterone value.

When testosterone is used as treatment, the number followed is usually a total. In TRAVERSE, the dose of 1.62% testosterone gel was adjusted to keep total testosterone between 350 and 750 ng/dL [8]. TRAVERSE tested a gel, not injections [8].

Fasting, Timing, and the Label's Draw Standard

The Depo-Testosterone and testosterone enanthate labels call for the diagnosis of hypogonadism to be confirmed before treatment, with serum testosterone measured in the morning on at least two separate days and both results below the normal range [1][2]. The Endocrine Society guideline recommends a fasting morning total testosterone as the initial test, repeated to confirm the diagnosis [3].

Fasting appears in trial design as well. TRAVERSE enrolled men who had two fasting testosterone levels below 300 ng/dL [8].

What Diet Has Actually Been Shown to Do

A lower-fat, higher-fiber diet

In 1983, 30 healthy middle-aged men were moved from their usual diet to an experimental diet that was lower in fat, higher in polyunsaturated-to-saturated ratio and higher in fiber. Mean total testosterone fell from 22.7 to 19.3 nmol/L (P<0.001) and free testosterone from 0.23 to 0.20 nmol/L (P<0.01), and the changes were reversible [5]. Because fat quantity, fat type and fiber all changed at once, the study cannot say which of them drove the drop.

Zinc

Zinc status and serum testosterone track together. In a cross-sectional study of 40 men aged 20 to 80, lymphocyte zinc correlated with serum testosterone at r=0.43 (P=0.006) [11]. In four young men, serum testosterone fell from 39.9 to 10.6 nmol/L (P=0.005) after 20 weeks of zinc restriction [11]. Nine marginally zinc-deficient older men (mean age 64) who took oral zinc gluconate for 3 to 6 months went from a mean of 8.3 to 16.0 nmol/L (P=0.02) [11]. These were serum testosterone measurements; free testosterone was not reported separately.

Magnesium

In 399 men aged 65 and older in the CHIANTI cohort, serum magnesium was positively associated with total testosterone after adjustment for age, BMI, interleukin-6, DHEAS, SHBG, insulin, IGF-1, grip strength, Parkinson disease and heart failure [12]. This is an association, free testosterone was not the outcome, and it does not show that magnesium supplements raise testosterone.

Sugar, insulin, and SHBG

The carbohydrate link runs through SHBG, and it has been worked out in liver cells rather than in men eating different diets. In a human liver cell line, glucose and fructose reduced SHBG expression by driving lipogenesis and lowering hepatic HNF-4 alpha, a transcription factor that controls the SHBG promoter; the same cell line increased SHBG production when exposed to genistein or tamoxifen [13]. The authors describe SHBG as a sensitive marker of insulin resistance and the metabolic syndrome [13].

Alcohol

Eight healthy men aged 20 to 26 drank 1.5 g of ethanol per kg of body weight over three hours. Serum testosterone fell by an average of 23% between 10 and 16 hours after drinking started. LH and FSH levels and pulses were unchanged, LH biological activity was unchanged, and growth hormone secretion was suppressed [6]. This was serum testosterone in a small single-dose study, not free testosterone.

Body weight

The European Male Ageing Study followed 2,736 community-dwelling men aged 40 to 79 for a mean of about 4.4 years, with paired testosterone results available for 2,395 [7]. Weight loss was associated with higher total testosterone and SHBG, and weight gain with lower levels. Free testosterone showed a curvilinear relationship: it changed significantly only in men whose weight changed by 15% or more [7].

Reading a Result

A free testosterone result is easier to interpret alongside its assay method and the timing of the blood draw [9]. The product labels call for two separate morning serum testosterone measurements to confirm hypogonadism before treatment [1][2].