High cortisol symptoms: the signs that count, and the ones that do not

The symptoms of high cortisol that circulate online, which are tiredness, weight gain around the middle, poor sleep, anxiety and a puffy face, are real features of cortisol excess. They are also features of being 45, stressed and short of sleep. Endocrinologists separate them into two groups, and only one group tells you anything: the signs that discriminate, and the signs that are simply common.

What are the symptoms of high cortisol, and which ones actually mean anything?

The Endocrine Society's clinical practice guideline on diagnosing Cushing's syndrome, written by Lynnette Nieman and colleagues in 2008, sorts the features of cortisol excess into exactly that split.

The features with high discriminatory value, meaning they are unusual in the general population and so point toward cortisol excess, are easy bruising, facial plethora, which is a persistent deep flush across the cheeks, proximal myopathy, which is weakness in the big muscles close to the trunk, and reddish-purple stretch marks, particularly ones wider than a centimetre. Thin skin in a young person belongs on the list. In children, weight going up while height velocity falls off is the classic pairing.

The features that are common and do not discriminate are depression, fatigue, weight gain, obesity, a dorsocervical fat pad, facial fullness, high blood pressure, an incidentally found adrenal mass and spinal osteoporosis.

Look at those two lists again and notice which one the internet talks about.

Why do the symptoms I keep reading about not count for much?

Because they are too common to be informative. If a sign appears in a quarter of the adult population and the condition appears in about two people per million per year, then almost everyone with the sign does not have the condition. This is the base-rate problem, and no amount of symptom-matching gets around it.

Tiredness, weight gain, broken sleep and anxiety travel together in enormous numbers of people for reasons that have nothing to do with the adrenal glands: poor sleep, alcohol, pain, thyroid disease, anaemia, depression, perimenopause, a new medication, or simply a hard year.

The discriminatory signs work differently. Thin skin that bruises from nothing, a deep facial flush, struggling to get out of a low chair or climb stairs because the thigh muscles have wasted, and wide purple striae are all unusual enough that a clinician takes them seriously as a set.

The guideline's own caution is worth repeating: most of the discriminating features do not have high sensitivity. They point toward the condition when present, but their absence does not rule it out.

How rare is genuinely high cortisol?

Rare enough that the numbers are worth seeing written down.

A Danish population study by Jens Lindholm and colleagues, covering every case diagnosed in the country over 11 years, found an incidence of 1.2 to 1.7 cases per million people per year for Cushing's disease, 0.6 per million for a cortisol-producing adrenal adenoma and 0.2 per million for adrenal carcinoma. A later nationwide study in Korea put adrenal Cushing's syndrome at 1.27 cases per million person-years after age standardisation.

So the whole category runs at roughly two to three new cases per million people a year. For comparison, the search term "how to lower cortisol" is typed tens of thousands of times a month in the United States alone.

That gap is the entire story of cortisol as a wellness category. Millions of people recognise a symptom list; a few hundred a year in a country the size of Britain have the condition it describes.

Is the first question really about medication?

Yes, and this is the single most useful thing in the guideline. Before any blood, urine or saliva test, the recommendation is to take a thorough drug history to rule out cortisol excess caused by the patient's own prescriptions.

Steroid tablets are the obvious case. But inhaled steroids for asthma, steroid creams used heavily or on large areas, steroid nasal sprays, joint injections and steroid eye drops can all contribute, and people do not think of them as drugs that reach the whole body. Some over-the-counter and imported skin creams contain potent steroids that are not declared.

If the cause is a prescription, the picture can look identical to the disease, and the answer is a conversation with the prescriber rather than a test.

Can ordinary stress give you high cortisol symptoms?

Stress does raise cortisol. That is not in dispute, and it is why the hormone became famous.

What is not established is the next step in the popular argument: that ordinary life stress produces a sustained elevation large enough to cause the physical changes on the discriminatory list, and that supplements reverse it. Those physical changes come from months to years of a substantial cortisol excess, which is a different quantity of hormone from a stressful fortnight.

There is also a measurement problem. Cortisol swings across the day by design, peaks within an hour of waking and responds to the stress of having blood taken. A single reading is close to meaningless, which is why the guideline recommends tests built around timing or suppression instead. We have written separately about how cortisol is actually tested and about what the supplements do and do not do.

Which signs make a doctor take this seriously?

Testing is recommended for people with multiple and progressive features, particularly the ones with high discriminatory value, and for people who turn out to have an adrenal mass found for some other reason.

The word "progressive" is doing real work there. A set of features that has appeared and worsened over months is a different proposition from a stable set you have had for years. Old photographs are genuinely useful here, and clinicians ask for them.

Also significant: these features arriving together rather than one at a time. Bruising plus muscle weakness plus new purple striae plus rising blood pressure plus new diabetes is a pattern. Tiredness on its own is not.

How would I actually get tested?

The guideline recommends starting with one test of high diagnostic accuracy, and names four: 24-hour urinary free cortisol, late-night salivary cortisol, a 1mg overnight dexamethasone suppression test, or a 2mg test over 48 hours.

An abnormal result means a referral to an endocrinologist and a second test, not a diagnosis. Two concordant abnormal results move you on to finding the cause. Two normal results mean no further investigation is needed.

The 2021 Pituitary Society consensus update, led by Maria Fleseriu with more than 50 contributors, kept the same broad approach and gave more detail on the role of salivary cortisol assays. None of this is a kit you buy and interpret yourself. The tests are timing-sensitive and the interpretation depends on things like shift work, pregnancy, oral contraceptives, licorice and tobacco.

What about "cortisol face" and "cortisol belly"?

Facial fullness and weight around the middle are both on the guideline's list, but both are in the non-discriminatory group. Facial plethora, that fixed deep flush, is the discriminatory version, and it is not the same thing as a rounder face.

The useful distinction, if you want one: in cortisol excess the fat redistributes rather than simply accumulating, so the face, the back of the neck and the abdomen fill while the arms and legs thin and weaken. A rounder face with unchanged limb strength is a very different picture from a rounder face with thighs that cannot push you out of a chair. We have looked at the cortisol face claim in more detail.

What should I do if I recognise myself here?

If you are matching the common list only, the honest answer is that cortisol is probably not your explanation, and the things worth checking are sleep, alcohol, thyroid function, iron, mood and medication. A GP can run most of that.

If you are matching the discriminatory list, especially several items that have appeared and worsened over months, book an appointment and say so in those words. Bring a complete list of every steroid you take in any form, including creams and inhalers, and bring photographs from a few years ago.

And if a product promises to lower your cortisol, remember the arithmetic. The condition it is invoking affects about two people per million per year, and the test for it is not something you can buy.

This article is general information and not medical advice. Speak to a qualified clinician about your own symptoms.