Adrenal glands

Two organs in one

The outer layer, the cortex, is a hormone factory arranged in three bands. Each band carries a different set of enzymes, and the enzymes decide the product. The outermost makes aldosterone, which tells the kidney how much salt and water to keep. The middle makes cortisol. The innermost makes the adrenal androgens, DHEA and its sulphated form.

All three start from cholesterol. Three destinations from one raw material, decided by nothing more than which enzymes a particular cell happens to carry. That is the same principle the Sex hormones section describes for aromatase and 5-alpha reductase.

The core is not glandular tissue at all. The medulla at the centre is nervous tissue. It is made of sympathetic nerve cells that lost their projections and secrete into the bloodstream instead. It makes adrenaline. So an adrenal gland is a piece of the nervous system with an endocrine gland wrapped around it.

That explains the timing. Adrenaline acts in seconds and is gone in minutes, because it is a nerve signal delivered by blood. Cortisol acts over hours, because it works the way the Thyroid section describes, entering the nucleus and changing which genes get read. One system, two speeds, one gland.

The loop, and why the rhythm is the signal

The control loop has the same shape as the thyroid and sex hormone loops. The hypothalamus signals the pituitary, the pituitary releases ACTH, ACTH drives the cortisol and androgen bands of the cortex. Cortisol produced feedback to quieten both. The aldosterone band sits partly outside that loop, answering mainly to the kidney's renin signal and to blood potassium, with ACTH a lesser influence.

What this axis has that the others do not is a clock sitting on top of it. Cortisol is not held at a steady level and corrected when it drifts. It is driven up and down on a daily cycle. Lowest around midnight, rising through the early hours, and peaking shortly after you wake, climbing roughly half as much again within the first 30 minutes. It then falls through the day.

So the rhythm is not noise sitting around some true value. The rhythm is what the system produces, and flattening it is itself a finding.

Two consequences. A cortisol result without a collection time attached is close to uninterpretable, because the reference range assumes a morning sample and an afternoon one is being measured against a scale it does not fit. And a single value describes one point on a curve rather than the curve. What to do about that before a test is in the Biomarker Guide.

What cortisol is actually for

Calling it the stress hormone is true and not very useful, because it is doing the same job all day whether anything stressful is happening or not. That job is making fuel available and keeping the rest of the system responsive.

It raises blood glucose, both by driving the liver to manufacture it and by opposing insulin in muscle and fat. That is the link to the Blood sugar section, and the reason steroid treatment can push someone into diabetic territory. It breaks down protein, mostly muscle, to supply the raw material for that manufacture. It suppresses inflammation and dampens the immune response, which is why the Inflammation section and this one are connected, and why steroid drugs work at all. They are cortisol analogues doing the same thing harder. And it keeps blood vessels responsive to adrenaline, which is a permissive role rather than a direct one. Cortisol does not raise blood pressure so much as allow adrenaline to.

Sustained excess is that list running without a brake. High blood sugar, muscle wasting, bone loss, thin and easily bruised skin, raised blood pressure, and a suppressed immune response. Every one of them is the ordinary job continuing for too long, which is why they are the recognised effects of long-term steroid treatment as well as of the disease.

One effect shows up somewhere that surprises people. Steroids raise the white cell count, and mostly not by producing more cells. They detach neutrophils from the vessel walls they were rolling along. The Blood cells section explains that mechanism, and it is why a raised white cell count on steroid treatment is usually redistribution rather than infection.

The androgens, and who they actually matter to

The innermost band of the cortex produces DHEA and DHEA-sulphate, in greater quantity by weight than any other steroid the body makes. They do very little themselves. They are precursors, converted in peripheral tissue into testosterone and oestradiol.

How much that matters depends entirely on who you are, and the difference is larger than most people expect.

In a man with normally functioning testes, adrenal androgens account for under 5% of circulating testosterone. Physiologically it is close to negligible.

In a woman of reproductive age, roughly a quarter of circulating testosterone comes directly from the adrenal glands and another quarter from the ovaries, with the remaining half made by conversion in peripheral tissue from precursors that are themselves largely adrenal. So the adrenal contribution to a woman's androgens is substantial, and to a man's it is not. That single asymmetry is why DHEA-sulphate gets looked at when androgen excess in a woman is being investigated, and is rarely informative in men. Guidance puts it behind testosterone rather than beside it, to be measured where testosterone is not already raised, because it is the less specific of the two and falls steeply with age.

After menopause the ovary's oestradiol production largely stops, and conversion of adrenal precursors in fat and other tissue becomes the main remaining source of oestrogen. The aromatase enzyme described in the Sex hormones section is what performs it, and it is one reason body composition and hormone status stay connected after the ovaries have stepped back.

DHEA-sulphate is the one measured, and the reason is stability. DHEA itself follows ACTH in pulses and is cleared from the blood quickly, so a single measurement catches whatever the pulse was doing. Attaching a sulphate group produces a molecule that circulates far longer, so the sulphated form behaves as a steady pool rather than a snapshot. And because it is made almost entirely by the adrenal glands and essentially not by the ovaries, it is used as a broad measure of adrenal androgen output, which points towards where an androgen excess is coming from rather than settling it.

It also declines steeply and continuously with age. It rises through childhood, peaks in the twenties, and falls from there for the rest of life. Which means an age-banded reference range is not a refinement here. A result that is unremarkable at 55 would be strikingly low at 25, and a range that is not banded by age is close to meaningless.

What a blood test can and cannot answer here

This is a section where the limits need stating plainly, because the gap between what people expect a cortisol test to do and what it does is wider than for anything else on this page.

The two conditions to name are too much cortisol, which is Cushing's syndrome, and too little, which is adrenal insufficiency or Addison's disease. Both are uncommon. Neither is diagnosed on a single morning blood cortisol.

Cushing's syndrome is screened with an overnight suppression test, a late-night salivary sample or a 24 hour urine collection. Adrenal insufficiency is confirmed with a stimulation test, which measures the response to a challenge rather than the resting level. We offer none of those.

Two of them you can arrange yourself. Late-night salivary cortisol is sold as a postal kit by some UK labs, and saliva is easy to collect at midnight whereas blood is not. What it has been validated for is suspected cortisol excess, where the normal midnight low is lost. It is not a general check on whether a rhythm has flattened, and an abnormal one is meant to go to a specialist and be followed by a second test rather than stand on its own. A 24 hour urine collection is also available privately. The suppression and stimulation tests both involve giving a drug, so those 2 genuinely need a doctor.

What a morning cortisol can do is point. A very low or very high result makes one of those conditions more or less likely, and that is a reason to be seen rather than an answer in itself.

Adrenal fatigue

This belongs here because it is the most common reason anyone buys a cortisol test, and because the answer is not the one the test is usually sold on.

The proposition is that prolonged stress exhausts the adrenal glands until they can no longer produce enough cortisol, and that this causes tiredness, low mood and poor recovery. No endocrinology society recognises it as a condition. A systematic review published in 2016 examined the whole body of evidence and found no substantiation for it, and went further than that and concluded that cortisol profile testing has no clinical utility in evaluating someone presenting with fatigue.

That review has been criticised, reasonably, for not defining the thing it set out to test. And the observation underneath it is real. Sustained stress does change how this axis behaves. What has not been shown is that it produces adrenal failure, that a cortisol measurement identifies it, or that anything useful follows from the number.

Which is uncomfortable to publish on a page that sells a cortisol test. If you are tired and want to know why, a cortisol result is unlikely to be the thing that tells you. The markers with a better claim are thyroid function, iron, B12 and folate, and blood sugar. Which is precisely why those are the ones in our Tired All The Time panel and cortisol is not.

What changes when this is disturbed

Too much cortisol, sustained, produces weight gain concentrated around the trunk and face, thin skin that bruises easily, purple stretch marks, muscle weakness in the thighs and shoulders, raised blood pressure and blood sugar, and bone loss.

Too little produces fatigue, weight loss, low blood pressure, dizziness on standing, salt craving, nausea and darkening of the skin. It can be gradual for a long time and then become an emergency quickly, which is the one thing about this system that is genuinely urgent.

Both lists overlap with a great deal else, and both conditions are rare. Dramatic symptoms, unremarkable causes far more often. That is the reason this section spends more space on what the test cannot do than on what it can.