How to Diagnose High Cortisol: 3 Screening Tests

Three screening tests are used to detect high cortisol (hypercortisolism): the late-night salivary cortisol test, the overnight 1-mg dexamethasone suppression test, and the 24-hour urinary free cortisol collection. International guidelines recommend starting with at least two of these, because no single test provides a definitive answer on its own.¹ Each test captures cortisol from a different angle and comes with its own collection quirks, timing requirements, and false-positive traps, so understanding what each one actually measures gives you a much better sense of what your results mean.

Late-Night Salivary Cortisol

Cortisol follows a daily rhythm: it peaks in the early morning and drops to its lowest point around midnight. In people with Cushing’s syndrome, that midnight dip disappears or becomes much shallower. The late-night salivary cortisol test (often abbreviated LNSC) exploits this pattern. You collect a saliva sample close to midnight, typically between 11 p.m. and midnight, and the lab checks whether your cortisol level is higher than it should be at that hour.

The appeal of this test is convenience. Saliva collection is non-invasive, painless, and can be done at home with a simple cotton swab or collection tube. You do not need to visit a lab or a hospital at midnight. Research consistently shows the test performs well: one study found a sensitivity of about 92%, meaning it correctly flagged the vast majority of confirmed Cushing’s patients, while another reported sensitivity and specificity both in the mid-90s when measured by modern mass spectrometry methods.²³ Because of that combination of accuracy and ease, many endocrinologists have started pushing for LNSC as the go-to first-line screen.

A newer refinement involves measuring not just salivary cortisol but also salivary cortisone at the same time. Cortisone is a closely related hormone that the body converts from cortisol in the salivary glands. Adding cortisone to the panel can boost specificity dramatically. One study found that using an optimal cortisone cutoff achieved 100% specificity while keeping sensitivity above 95%.³ That dual-measurement approach is not yet standard everywhere, but it is gaining traction in centers that use mass spectrometry-based assays.

There are practical rules that matter for accuracy. Smoking, chewing tobacco, and licorice can all artificially raise salivary cortisol. Brushing your teeth shortly before collection can introduce blood contamination. Night-shift workers have a shifted circadian rhythm, so their cortisol nadir occurs at a different time of day, making the standard midnight collection misleading. Guidelines generally recommend collecting two separate samples on different nights, because a single elevated reading could reflect an off night rather than a chronic problem.

The Overnight Dexamethasone Suppression Test

This test works on a feedback principle. Dexamethasone is a synthetic steroid that mimics cortisol’s signal to the brain. When you take a small dose (1 mg) at bedtime, a healthy pituitary gland responds by cutting its cortisol-production signal. The next morning, your blood cortisol should be low. If it stays elevated, something is overriding the normal feedback loop.

You swallow a 1-mg dexamethasone tablet at 11 p.m., then have blood drawn the following morning, usually between 8 and 9 a.m. A morning serum cortisol at or below 50 nmol/L (roughly 1.8 µg/dL) is considered a normal suppression, effectively ruling out autonomous cortisol excess.⁴ A value above that threshold warrants further investigation.

The test is highly sensitive, meaning it catches most true cases, but its specificity is moderate, meaning it also flags some people who do not actually have Cushing’s syndrome.⁵ The clinical consequence is that false positives are a real concern, particularly if the test is used in isolation. One large series found the DST had a specificity of about 97% when a higher cortisol cutoff was used, but specificity dropped to around 20% at a lower cutoff that swept in too many borderline results.⁶

A major source of false positives is inadequate absorption of dexamethasone itself. People metabolize dexamethasone at different rates. Drugs that rev up a liver enzyme called CYP3A4, including certain anti-seizure medications, rifampin, and some older diabetes drugs, can break down dexamethasone so fast that not enough reaches the bloodstream to suppress cortisol.⁷⁸ Some clinics now measure serum dexamethasone alongside cortisol the morning after the test. Research shows that when dexamethasone levels are adequate (above roughly 5 nmol/L), there is no meaningful relationship between dexamethasone concentration and cortisol outcome, but low dexamethasone levels clearly correlate with falsely unsuppressed cortisol.⁹ In other words, measuring dexamethasone is most useful for catching those false positives rather than improving accuracy across the board. One study found that simultaneous dexamethasone measurement reduced false positives by about 20%.⁵

The DST is especially favored when screening for mild autonomous cortisol secretion in people who have an adrenal nodule found incidentally on imaging. European guidelines specifically recommend the 1-mg overnight DST as the primary screen for these patients.⁴

24-Hour Urinary Free Cortisol

Rather than measuring cortisol at one point in time, the 24-hour urine collection captures your total cortisol output over an entire day and night. Because cortisol pulses throughout the day, a single blood draw can be misleading. The urine test smooths out that variation by integrating the full day’s production. The cortisol in your urine reflects the free (unbound) fraction in blood that is filtered by the kidneys, so it gives a direct measure of bioactive cortisol exposure.¹⁰

The main drawback is logistical. You collect every drop of urine over a 24-hour window in a large container, keep it refrigerated, and bring it to the lab. Missed collections, overfilled containers, and volume errors are common and can throw off results. An incomplete collection underestimates cortisol; excess fluid intake can produce a falsely high total. For this reason, labs usually measure creatinine in the same sample to verify that the collection was complete.

Kidney function also matters. If your glomerular filtration rate is significantly reduced, urinary cortisol drops even if blood cortisol is high, because less cortisol gets filtered into the urine. Guidelines generally consider this test unreliable in people with substantially impaired kidney function. On the other end, very high fluid intake can increase urine volume and push urinary free cortisol above the reference range in people who are perfectly healthy. When performed under proper conditions and measured with a sensitive assay, urinary free cortisol has a reported sensitivity around 96% and specificity around 91%.¹⁰ Both the DST and the urine test show high sensitivity and specificity when done correctly.¹¹

Why You Usually Need More Than One Test

Cushing’s syndrome is rare in the general population, but many of its symptoms, such as weight gain, high blood pressure, fatigue, and mood changes, are common. That mismatch means screening tests are applied to a lot of people who ultimately do not have the condition, which amplifies the practical impact of even a small false-positive rate. Guidelines from the Endocrine Society recommend that if the first screening test is abnormal, you should see an endocrinologist and undergo a second, different screening test before concluding that you have hypercortisolism.¹² If both tests agree that cortisol is elevated, workup proceeds. If both are normal, further testing is usually unnecessary. Discordant results, where one test is abnormal and the other normal, call for additional evaluation, sometimes including more specialized tests like a serum midnight cortisol or the dexamethasone-CRH test.

A consensus statement from Korean and Japanese endocrine societies echoed this approach, emphasizing that a multimodal strategy using at least two complementary tests is essential to maximize accuracy and minimize the chance of misdiagnosis.¹ Which two tests get paired depends on clinical context, what equipment the lab has, and patient-specific factors. For someone who works night shifts, the salivary cortisol test loses its advantage because the circadian reference range no longer applies. For someone on medications that speed up dexamethasone clearance, the DST may produce unreliable results. For someone with kidney disease, the urine test is compromised. Choosing the right combination is part of why initial screening is best done in consultation with a specialist.

The Assay Method Matters More Than You Might Think

Not all cortisol measurements are equal, even when the same sample type is used. Most hospital labs measure cortisol using immunoassays, which are fast and cheap but rely on antibodies that can cross-react with other steroid compounds in the body. Mass spectrometry, a more precise technique, separates cortisol from its look-alikes and gives a cleaner number. Research shows that immunoassays tend to measure cortisol higher than mass spectrometry, particularly in urine and saliva, because interfering glucocorticoid metabolites get counted alongside actual cortisol.¹³

Despite this, the clinical sensitivity of both methods appears to be broadly similar for detecting Cushing’s syndrome. The bigger issue is specificity and consistency across labs. Mass spectrometry offers substantially less variation between laboratories and fewer false-positive results.¹⁴ If you are getting tested at different centers over time, results may not be directly comparable if the labs use different assay platforms. Asking your doctor which method the lab uses is a reasonable step, especially when a borderline result is driving a clinical decision.

Pseudo-Cushing States and Why They Complicate Everything

Some conditions cause mildly elevated cortisol without actual Cushing’s syndrome. Doctors call these pseudo-Cushing states, and they are far more common than true Cushing’s because the conditions behind them, including obesity, depression, heavy alcohol use, poorly controlled diabetes, and polycystic ovary syndrome, are widespread.¹⁵ A person with severe obesity can have elevated urinary cortisol and a mildly abnormal dexamethasone suppression result without harboring a cortisol-secreting tumor. One study found that when screening obese patients for Cushing’s disease using late-night salivary cortisol, the sensitivity and specificity both dropped substantially compared to testing in lean individuals.¹⁶

When the clinical picture is ambiguous, a more specialized test called the dexamethasone-CRH test can help. In this two-step procedure, you take low-dose dexamethasone over two days, then receive an injection of corticotropin-releasing hormone (CRH). In true Cushing’s, the pituitary still responds to CRH with a cortisol surge despite the dexamethasone. In pseudo-Cushing states, the combination suppresses cortisol more effectively. An early study of this approach found 100% diagnostic accuracy in separating the two groups.¹⁷ Later research confirmed the test’s usefulness, though real-world accuracy in milder cases is not quite as clean as that initial report suggested.¹⁸

Screening During Pregnancy

Pregnancy naturally ramps up cortisol production. The placenta produces CRH, which stimulates the mother’s adrenal glands, and cortisol-binding protein levels rise, pushing total cortisol higher. Urinary free cortisol can climb to two or three times the normal upper limit by the third trimester in a completely healthy pregnancy. Additionally, dexamethasone does not fully suppress cortisol in pregnant women who do not have Cushing’s syndrome, which renders the standard DST unreliable.¹⁹ For these reasons, diagnosing true Cushing’s during pregnancy is one of the trickiest scenarios in endocrinology. Late-night salivary cortisol remains the most practical option in this setting, though even its reference ranges need to be adjusted for gestational age.

After Screening Comes the Search for a Cause

Confirming that cortisol is high is only the first step. The next question is why. Doctors typically measure plasma ACTH (adrenocorticotropic hormone), the pituitary hormone that tells the adrenal glands to produce cortisol. If ACTH is low (below roughly 10 pg/mL), the problem is in the adrenal glands themselves, often a tumor producing cortisol on its own. If ACTH is normal or high (above roughly 20 pg/mL), the signal is coming from elsewhere, usually a pituitary tumor or, more rarely, an ectopic source.²⁰ There is an overlap zone between 10 and 20 pg/mL where results are ambiguous, and further dynamic testing or imaging is needed to sort things out.²¹

This distinction matters because treatment differs completely. An adrenal tumor often means surgery on the adrenal gland. A pituitary tumor (the most common cause of Cushing’s disease) typically involves transsphenoidal surgery through the nose to remove the adenoma. An ectopic ACTH source, like a small lung tumor, requires locating and treating that primary tumor. Getting the cause right from the start avoids unnecessary procedures, so the diagnostic workup after positive screening tends to be detailed and sometimes frustrating in its pace.

Cyclic Cushing’s Syndrome

Some patients have cortisol levels that fluctuate between clearly elevated and entirely normal, sometimes over days, sometimes over weeks or months. This pattern, called cyclic Cushing’s syndrome, is among the most difficult diagnostic puzzles in endocrinology.²² A patient tested during a quiet phase can have completely normal screening results, only to develop florid symptoms weeks later when cortisol surges again. Standard practice guidelines account for this by recommending that patients with initially normal results but growing clinical suspicion undergo repeated testing over time.¹²

One emerging tool for catching cyclic cases is hair cortisol analysis. Because hair grows slowly and incorporates cortisol as it is produced, a segment of hair can provide a retrospective record of cortisol exposure over weeks or months. This approach is still mostly a research tool and not widely available in clinical labs, but it addresses a genuine blind spot in screening: capturing cortisol excess that is present but intermittent. The desmopressin stimulation test is another adjunct being explored for the same purpose, though clinical adoption remains limited.

Mild Autonomous Cortisol Secretion

Not every case of excess cortisol looks like textbook Cushing’s syndrome with a round face, purple stretch marks, and muscle wasting. A large and growing category involves people with adrenal nodules found incidentally on imaging scans done for other reasons. These nodules sometimes secrete modest amounts of cortisol, not enough to produce dramatic symptoms but potentially enough to raise blood pressure, worsen blood sugar, or contribute to bone thinning over years. European guidelines define this as mild autonomous cortisol secretion (MACS) when the post-dexamethasone cortisol is above 50 nmol/L in someone without overt Cushing’s features.⁴

Diagnosing MACS is trickier than diagnosing overt Cushing’s because the cortisol excess is subtle and screening tests perform less cleanly at these low levels. One study found that about a third of patients with non-functioning adrenal nodules had cortisol above the 50 nmol/L suppression cutoff after dexamethasone, suggesting the test flags a lot of borderline cases that may or may not represent clinically meaningful cortisol excess.²³ Whether and when to treat MACS remains actively debated. Current guidelines lean toward monitoring cardiovascular risk factors and metabolic health rather than rushing to surgery, unless the cortisol excess is clearly worsening comorbidities that are resistant to standard treatment.

1
Endocrinology and Metabolism. Consensus on the Diagnosis of Cushing’s Disease: A Collaborative Statement from the Korean Endocrine Society and Japan Endocrine Society
2
PubMed. Late-night salivary cortisol as a screening test for Cushing’s syndrome
3
PubMed Central. Late-night salivary cortisol and cortisone should be the initial screening test for Cushing’s syndrome
4
European Journal of Endocrinology. European Society of Endocrinology clinical practice guidelines on the management of adrenal incidentalomas, in collaboration with the European Network for the Study of Adrenal Tumors
5
European Journal of Endocrinology. Simultaneous assay of cortisol and dexamethasone improved diagnostic accuracy of the dexamethasone suppression test
6
The Journal of Clinical Endocrinology & Metabolism. Specificity of First-Line Tests for the Diagnosis of Cushing’s Syndrome: Assessment in a Large Series
7
The Journal of Clinical Endocrinology & Metabolism. Troglitazone Induces CYP3A4 Activity Leading to Falsely Abnormal Dexamethasone Suppression Test
8
Endocrine Practice. Overnight 1-mg Dexamethasone Suppression Test for Screening Cushing Syndrome and Mild Autonomous Cortisol Secretion (MACS): What Happens when Serum Dexamethasone Is Below Cutoff? How Frequent Is it?
9
PubMed Central. Association between posttest dexamethasone and cortisol concentrations in the 1 mg overnight dexamethasone suppression test
10
Steroids. Clinical utility of an ultrasensitive urinary free cortisol assay by tandem mass spectrometry
11
PubMed. Overnight 1 mg dexamethasone suppression test and 24 h urine free cortisol-accuracy and pitfalls when screening for Cushing’s syndrome
12
The Journal of Clinical Endocrinology & Metabolism. The Diagnosis of Cushing’s Syndrome: An Endocrine Society Clinical Practice Guideline
13
PubMed. Performance of LC-MS/MS and immunoassay based 24-h urine free cortisol in the diagnosis of Cushing’s syndrome
14
PubMed Central. Cortisol Measurements in Cushing’s Syndrome: Immunoassay or Mass Spectrometry?
15
PubMed Central. Approach to patients with pseudo-Cushing’s states
16
PubMed Central. Specificity of late-night salivary cortisol measured by automated electrochemiluminescence immunoassay for Cushing’s disease in an obese population
17
PubMed. Corticotropin-releasing hormone stimulation following low-dose dexamethasone administration. A new test to distinguish Cushing’s syndrome from pseudo-Cushing’s states
18
PubMed Central. The dexamethasone-suppressed corticotropin-releasing hormone stimulation test differentiates mild Cushing’s disease from normal physiology
19
PubMed Central. The diagnosis and management of Cushing’s syndrome in pregnancy
20
The Journal of Clinical Endocrinology & Metabolism. Approach to the Patient: Diagnosis of Cushing Syndrome
21
PubMed Central. Limited Diagnostic Utility of Plasma Adrenocorticotropic Hormone for Differentiation between Adrenal Cushing Syndrome and Cushing Disease
22
PubMed Central. Cyclic Cushing’s Syndrome – A Diagnostic Challenge
23
Frontiers in Endocrinology. The diagnostic value of salivary cortisol and salivary cortisone in patients with suspected hypercortisolism