The three fractions of testosterone
Testosterone in the bloodstream exists in three forms:
Tightly bound testosterone (approximately 44–57%): bound to sex hormone-binding globulin (SHBG), a protein produced primarily by the liver. This fraction is biologically inactive — SHBG binds testosterone so firmly that it cannot interact with cellular androgen receptors.
Loosely bound testosterone (approximately 40–54%): bound to albumin. This binding is weak and reversible; albumin-bound testosterone is partially available to tissues. Together with free testosterone, it constitutes "bioavailable testosterone."
Free testosterone (approximately 2–3%): unbound. This is the fraction that crosses cell membranes, binds to androgen receptors, and produces the physiological effects attributed to testosterone — libido, energy, muscle protein synthesis, bone density, and mood regulation.
Total testosterone measures all three fractions combined. It does not distinguish between what is pharmacologically active and what is sequestered.
Why SHBG changes the picture
SHBG concentration varies substantially between individuals and is affected by multiple conditions:
Conditions that raise SHBG:
- Aging (SHBG rises approximately 1–2% per year after age 40)
- Liver disease (SHBG is synthesised by the liver; hepatic inflammation increases SHBG)
- Hyperthyroidism
- Oestrogen exposure
- Certain anticonvulsants (phenytoin, carbamazepine)
Conditions that lower SHBG:
- Obesity and insulin resistance (insulin directly suppresses SHBG production)
- Hypothyroidism
- Elevated androgens
- Glucocorticoid excess
The practical consequence: two men with identical total testosterone levels may have dramatically different free testosterone levels depending on their SHBG. A man with aging-related SHBG elevation may have a total testosterone of 500 ng/dL but free testosterone of 8 pg/mL — well below the threshold associated with symptoms. A man with obesity and low SHBG may have a total testosterone of 320 ng/dL but free testosterone of 14 pg/mL, which is biologically normal.
The standard "is it above the reference range or below it?" question asked of total testosterone alone is insufficient.
What the reference ranges actually represent
Laboratory reference ranges for total testosterone are generated from large population samples. In India, most laboratories use ranges derived from Western populations or from limited Indian data. Several issues arise:
The lower bound of "normal" includes symptomatic men. Reference ranges are set to include 95% of the reference population, which itself may include men with early testosterone deficiency who have not yet been diagnosed. A total testosterone of 300 ng/dL may be labelled "within normal limits" while being clinically insufficient for a given patient.
Morning sampling is required. Testosterone follows a diurnal rhythm — levels peak in the early morning and fall substantially through the day. An afternoon total testosterone may be 20–30% lower than a morning value in the same patient, causing a borderline result to appear normal or abnormal depending on when the blood was drawn.
Single measurements are insufficient. Testosterone levels fluctuate daily. A single borderline result should be confirmed with a repeat morning measurement.
Indian population data. The Yadav et al. 2019 study from Sir Ganga Ram Hospital found testosterone deficiency syndrome (TDS) in 48.18% of men aged 40 and above using clinical and biochemical criteria — a figure substantially higher than Western estimates. The lower mean testosterone levels in Indian men may reflect differences in metabolic burden, lifestyle factors, and possibly reference population selection.
How free testosterone is measured and interpreted
Calculated free testosterone: The most widely used clinical approach. The Vermeulen formula uses total testosterone, SHBG, and albumin to calculate free testosterone. Most Indian laboratories can provide all three measurements needed. This method is validated and correlates well with measured free testosterone in most clinical contexts.
Measured free testosterone by equilibrium dialysis: The gold standard. Dialysis separates bound from free testosterone directly. Rarely available in routine Indian laboratory practice; typically used in research settings or ambiguous cases.
Direct immunoassay for free testosterone: Commercial kits exist but have poor accuracy and should not be used for clinical decision-making.
In a man with symptoms consistent with testosterone deficiency and a borderline total testosterone, the next step is to measure SHBG and calculate free testosterone — not to accept the total testosterone result at face value.
What this means for clinical decisions
A man whose total testosterone is 420 ng/dL (within the reference range) but whose SHBG is 55 nmol/L (elevated due to aging) and calculated free testosterone is 9.2 pg/mL (below the threshold for most symptoms) has a clinical picture of testosterone deficiency — regardless of the total testosterone number. A decision about management based only on total testosterone would miss his condition entirely.
Conversely, a symptomatic man whose total testosterone is 280 ng/dL (below many reference ranges) but whose SHBG is low at 15 nmol/L and calculated free testosterone is 16 pg/mL does not have testosterone deficiency by bioavailability criteria — and testosterone supplementation would be inappropriate.
This is why a testosterone evaluation that measures only total testosterone, without SHBG and free testosterone, is an incomplete evaluation.