← Reference library

REFERENCE LIBRARY

The HPA axishypothalamic–pituitary–adrenal

The body's slow channel for handling threat: the hypothalamus signals, the pituitary relays, the adrenal cortex releases cortisol — three stages, running on minutes to hours. Its most important design feature is that it shuts itself off. And failing to shut off is exactly where chronic stress does its damage.

Two channels: one in seconds, one in minutes

A threat opens two routes. The fast one: sympathetic nerves run straight to the adrenal medulla, which releases adrenaline within seconds — heart racing, pupils wide, blood sugar up. The second is the HPA axis, much slower. The paraventricular nucleus of the hypothalamus releases CRH, carried down a short dedicated blood vessel to the pituitary; the pituitary releases ACTH into the general circulation; ACTH reaches the adrenal cortex, which releases cortisol. From event to peak cortisol usually takes fifteen to twenty-odd minutes.

Their jobs differ too. The fast channel handles "move now"; the slow one handles "how to hold out over the next stretch" — mobilising glucose, suppressing expenses that can wait (digestion, reproduction, parts of immune function), and feeding back into the brain: hippocampus, amygdala and prefrontal cortex are all dense with cortisol receptors. So stress isn't only a bodily response — it directly rewrites the memory and judgement you're making at that moment.

What matters isn't switching on, it's switching off

Once cortisol rises, it binds receptors back on the hypothalamus and pituitary and presses the upstream switches down — negative feedback, the same logic as a thermostat cutting out at temperature. The hippocampus helps push that brake too: it carries a high density of glucocorticoid receptors and is an important source of the inhibitory signal.

Which is where the trouble starts: the hippocampus is also the structure most vulnerable to cortisol. Chronically high levels thin its dendritic branches and suppress neurogenesis in the dentate gyrus. That makes a slide: high cortisol → damaged hippocampus → blunter brake → higher cortisol. On top of that, receptors marinated in high hormone become less responsive (glucocorticoid resistance): the blood level isn't low, but fewer cells still "hear" it. Medicine calls the accumulated wear of adapting like this allostatic load.

It's a clock first and a stress gauge second

Treating cortisol purely as "the stress hormone" misses half of it. It has a strong circadian rhythm: climbing before dawn and peaking 30–45 minutes after you wake (that bump is the cortisol awakening response), then falling all day to a trough around midnight. That isn't because mornings are stressful — cortisol's day job is to push you out of sleep and into wakefulness by raising blood sugar and blood pressure and mobilising energy.

So a single blood draw says little about the state of the axis; what you want is the shape — is the morning peak there, is the daytime slope steep enough. Chronic stress, accumulated sleep debt and shift work characteristically flatten the curve: it can't climb in the morning and won't come down at night.

"Stress illness = high cortisol" is an over-simplification

Depression: often elevated, and hard to suppress

A subset of depressed patients run high cortisol and fail to suppress it when given dexamethasone (a synthetic glucocorticoid that shuts off a healthy person's own output) — evidence that the negative feedback is blunted. Last century this was hoped to become a diagnostic test, but sensitivity and specificity both fell short and it never became one. Its value now is as a mechanistic clue, not a lab result.

PTSD: often the opposite, running low

The typical PTSD pattern is low basal cortisol with unusually strong negative feedback — the reverse of depression. Both are "stress disorders," yet the axis can be displaced in opposite directions.

Burnout: no stable cortisol signature

"Adrenal fatigue" has no standing in endocrinology and is not recognised by any mainstream guideline. Genuine adrenal insufficiency (Addison's disease) is a different thing entirely, with clear laboratory criteria.

What actually moves it

Sleep and light: the axis takes its phase from the master clock, so a consistent wake time and morning light are the most direct way to put that morning peak back where it belongs. Exercise: a single session acutely raises cortisol — that's normal mobilisation, not harm; what regular training does over months is make the axis respond less and recover faster to the same stressor. Slow-wave sleep: deep sleep is the strongest inhibitory window this axis gets, which is why sleep debt writes itself straight onto evening cortisol.

And the slowest-acting, largest-effect item is perceived control: the same objective stressor produces a differently shaped hormone curve depending on whether you can do something about it. That's a large part of why psychological interventions leave marks on physiological measures.