Testosterone

Blood test · early-morning draw

Also known as: Total testosterone, Free testosterone, Bioavailable testosterone

264 ng/dL Endocrine Society: low below this line
300 ng/dL AUA: low below this line

For men. These lines mark where two guidelines call testosterone low. They come from the spread of levels in healthy men, not from how men feel. Symptoms can begin above them, at different levels for different symptoms, so a result above the lines is not proof that testosterone is adequate for you. A low result counts only if it is confirmed on a second early-morning test and there are symptoms.

Units and reference ranges differ between labs.

Cutoffs: Endocrine Society, 2018; American Urological Association, 2018

Testosterone is the main male sex hormone, and women make it in smaller amounts. Most of it travels in the blood bound to proteins, so a lab can report total, free or bioavailable testosterone. In men, it is read together with symptoms to diagnose testosterone deficiency.

What it tells you

About 44% of testosterone is tightly bound to a protein called SHBG, about half is loosely bound to albumin, and about 2% is free. Total testosterone counts all of it and is the first test guidelines use.
Free testosterone is the unbound part. It helps when total testosterone is borderline or SHBG is unusually high or low. Guidelines set no single free testosterone cutoff.
Bioavailable testosterone is the free part plus the part loosely bound to albumin. No guideline uses it for diagnosis, and it has no guideline cutoff.
In men, both main guidelines require two low early-morning results plus symptoms or signs. They draw the line in slightly different places.
The lines are not a goal. In a large European study, each sexual symptom began to become more likely at a different level.
In women, no testosterone level diagnoses low desire or any other condition. An international consensus says no cutoff separates women with and without sexual problems.

What can throw it off

Time of day matters. Levels peak in the early morning. In one study, half the men with a low afternoon result were not low at any of three morning visits.
Levels vary from day to day. About 30% of men with one low result are not low on a repeat test.
Sugar and food can lower it. A glucose drink lowered testosterone by about 25% in one study, so one guideline advises a fasting draw.
A recent illness lowers it. Testing waits until you have recovered.
Opioid pain medicines lower it, and so can steroid medicines used for inflammation.
Changes in SHBG move total testosterone without changing free testosterone. Obesity and diabetes lower SHBG; aging and liver disease raise it. Free testosterone sorts this out.
Birth control pills raise SHBG and lower androgen production, so testosterone is hard to judge in women taking them.
Lab method matters. Many immunoassays read women's levels too high and men's slightly low compared with mass spectrometry. Compare results from the same lab and method.

Evidence

Diagnosing low testosterone in men

Some human evidenceCEBM Level 2b

No gold-standard test exists for low testosterone. So the guideline lines come from the spread of levels in healthy men and from expert agreement, not from proof of where symptoms start. Free testosterone, calculated or measured by a reference method, helps when total testosterone is borderline.

Predicting death and heart disease in men

Strong human evidenceCEBM Level 1a

Across 9 studies of about 20,600 men, only very low testosterone went with more deaths. The thresholds were below 213 ng/dL for any cause and below 153 ng/dL for heart deaths. Levels were not linked to new heart disease. Low testosterone may mark illness rather than cause it.

Diagnosing low desire in women

Tested: no benefit shownCEBM Level 2b

In 1,021 women, no single androgen level predicted low sexual function. An international consensus agrees that no testosterone level can tell women with and without sexual problems apart.

Limits

The number is not the diagnosis. Both guidelines require symptoms or signs plus repeated low morning results, and symptoms like tiredness and low mood have many causes.
The two lines disagree. A man between 264 and 299 ng/dL counts as low under one guideline but not the other.
A result above the lines is not proof that testosterone is adequate. The lines describe the spread of levels in healthy men, not how any one man feels.
Free testosterone has no agreed range, and results depend on the method. Direct free testosterone immunoassays are inaccurate.
Bioavailable testosterone has no guideline cutoff and is mainly a research measure.
In women it diagnoses nothing. It is used to check that levels are not already high before treatment and do not go too high during it.
The link with death is an association, seen only at very low levels.

References

Reviewed by Shubham Gupta, MD — Founder & President · November 5, 2026

This page is for education only. It is not medical advice. Decisions about any treatment belong with a physician who knows your history.