Testosterone Blood Testing: Getting It Right

A single afternoon testosterone reading tells you very little, and a substantial number of prescriptions are written on exactly that. The workup that supports a real diagnosis is not complicated, but it has to be done properly.

Testosterone testing looks simple. It is a blood draw. The complications are all in the timing, the repetition, and the surrounding measurements — and those are the parts that get skipped when a service is built for speed rather than accuracy.

Timing: why the morning matters

Testosterone secretion follows a circadian pattern. In most men, particularly younger men, levels peak in the early morning and fall through the day. The size of that daily swing is large enough that the same man can produce a result in the normal range at eight in the morning and a result below the range in the late afternoon, with nothing having changed except the clock.

Standard practice is therefore to draw the sample in the morning, typically within a few hours of waking. Fasting is usually requested, because eating — particularly a carbohydrate-containing meal — transiently lowers measured testosterone. If your provider does not tell you when to attend or whether to fast, that is a signal about how the rest of the assessment will be run.

Acute illness also matters. Testosterone falls during significant illness, after surgery and during periods of severe physiological stress. Testing during or shortly after any of those can produce a result that has nothing to do with your baseline. Postponing a fortnight is usually the right call.

One result is not a diagnosis

Testosterone varies day to day in the same individual, and laboratory assays carry their own variability. Guidelines from the major endocrine bodies consistently call for a low result to be confirmed on a second morning sample taken on a different day before any diagnosis is made.

This is not bureaucratic caution. A meaningful proportion of men with one low reading produce a normal reading on repeat testing. Prescribing on the first number means treating some men who never had the condition — with a Schedule III controlled substance, indefinitely, with all the consequences that carries. Repeat testing is cheap. Unnecessary lifelong treatment is not.

Total, free, and what SHBG has to do with it

Most circulating testosterone is bound to proteins. A large fraction is bound tightly to sex hormone-binding globulin (SHBG), a smaller fraction is bound loosely to albumin, and a small remainder circulates unbound. Total testosterone measures all of it. Free testosterone measures the unbound portion.

Total testosterone is the appropriate first test in most men. The complication is that anything which changes SHBG changes total testosterone without necessarily changing the biologically available amount. SHBG tends to be lower in obesity, insulin resistance, type 2 diabetes, hypothyroidism, and with corticosteroid use. It tends to be higher with ageing, hyperthyroidism, liver disease, HIV and some anticonvulsants and oestrogens.

The practical consequence: an obese man with low SHBG may show a low total testosterone while his free testosterone is unremarkable. An older man with high SHBG may show a reassuring total while his free level is genuinely low. Where the clinical picture and the total testosterone disagree, or where an SHBG-altering condition is present, free testosterone is measured or calculated.

How free testosterone is obtained matters. Equilibrium dialysis is the reference method but is not universally available. Calculated free testosterone from total testosterone, SHBG and albumin is widely used and reasonable. Direct analogue immunoassays for free testosterone are regarded as unreliable and are not recommended for this purpose. It is fair to ask which method your laboratory used.

The rest of the panel

Testosterone alone answers one question. It does not tell you why the level is low, whether treatment would be safe, or whether something else explains the symptoms. A workup that stops at a testosterone number is not a workup.

Locating the problem

LH and FSH. These pituitary hormones separate primary testicular failure — where LH and FSH are elevated because the pituitary is pushing against an unresponsive testis — from secondary hypogonadism, where LH and FSH are low or inappropriately normal and the problem lies higher up. The distinction changes the investigation and sometimes changes the treatment entirely. It is also central to the fertility conversation.

Prolactin. Elevated prolactin suppresses the axis and can indicate a pituitary adenoma. It is a small addition to the panel and occasionally the single most important result on it.

Ruling out the alternatives

Thyroid function, because hypothyroidism reproduces the entire symptom picture. A full blood count, which identifies anaemia and establishes the baseline haematocrit that any future monitoring depends on. Ferritin where iron deficiency is plausible. HbA1c and a lipid panel, because metabolic disease and low testosterone travel together and both need addressing. Liver and kidney function. Vitamin D where relevant.

Safety baselines

Haematocrit before starting anything, because testosterone raises red cell mass and you cannot interpret a later value without the starting one. PSA in men of an age where prostate assessment is appropriate — typically discussed from around forty in higher-risk men and from the mid-forties to fifty more generally, alongside a conversation about what the result does and does not mean. Blood pressure. Where sleep apnoea is suspected, a sleep assessment rather than a blood test. What each of these baselines is for is set out in TRT risks and the monitoring that should come with it.

Reference ranges are not universal

Different laboratories use different assay platforms, different calibration and different reference populations. The lower limit of normal is not a single agreed number across all of them, and a result near a threshold can sit on either side of it depending on where the sample was analysed. Some ranges are also age-stratified and some are not, which is its own source of confusion.

What this means for you: compare a result only to the range printed by the laboratory that produced it, be sceptical of any provider quoting a universal cut-off, and try to have serial testing done at the same laboratory so the numbers are comparable. Mass spectrometry-based assays are generally regarded as more accurate than immunoassays, particularly at lower concentrations.

Questions worth asking before you pay

• What time will the sample be drawn, and should I fast?
• Will a low result be confirmed on a second sample before anything is prescribed?
• Which additional tests are included, and are LH, FSH, prolactin and thyroid among them?
• Which laboratory is used, and will repeat tests go to the same one?
• If my levels come back normal, what happens next?

That last question is the most revealing. A clinical service has an answer for it — investigate the symptoms further, look at sleep, look at mood, look at metabolic health. A service built to sell testosterone often does not. Our full guide to testosterone replacement therapy sets out who the treatment genuinely helps, our piece on why TRT is not an anti-ageing treatment covers the men it is most often sold to inappropriately, and you can read how our hormone health service handles assessment before booking anything.