INTRODUCTION — Dexamethasone suppression tests (DSTs) are used to screen for excess cortisol production (from an adrenal incidentaloma) and Cushing's syndrome and to evaluate the differential diagnosis of corticotropin (ACTH)-dependent Cushing's syndrome. Dexamethasone is approximately 30 to 40 times more potent than cortisol and has no appreciable mineralocorticoid activity. The DSTs assess the hypothalamic and pituitary corticotroph cell responses to glucocorticoid negative feedback inhibition of corticotropin-releasing hormone (CRH) and ACTH secretion.

This topic will review the basic principles of the DSTs. Additional information on when to choose them to determine the diagnosis and the cause of Cushing's syndrome is discussed separately. The diagnosis of autonomous cortisol hypersecretion in adrenal incidentalomas is also reviewed separately.

●(See "Establishing the diagnosis of Cushing's syndrome".)

●(See "Establishing the cause of Cushing's syndrome".)

●(See "Evaluation and management of the adrenal incidentaloma".)

LOW-DOSE DSTs — The low-dose dexamethasone suppression tests (DSTs) are standard screening tests to differentiate patients with endogenous Cushing's syndrome of any cause from patients who do not have Cushing's syndrome. The high-dose test described below should not be used for this purpose. (See 'High-dose DSTs' below and "Establishing the diagnosis of Cushing's syndrome".)

The binding of dexamethasone to glucocorticoid receptors in hypothalamic paraventricular nuclei and in pituitary corticotroph cells inhibits corticotropin-releasing hormone (CRH) and corticotropin (ACTH) secretion. In humans, dexamethasone does not directly inhibit adrenal steroid production [1].

If the hypothalamic-pituitary-adrenal (HPA) axis is functioning normally, any supraphysiologic dose of dexamethasone is sufficient to suppress pituitary ACTH secretion. This should reduce cortisol production, with concomitant decreases in serum, saliva, and urine concentrations.

Two main protocols are used: the overnight 1 mg screening test and the two-day, low-dose test. Side effects are extremely rare, and either test can be conducted on an outpatient basis. Obviously, the test should not be conducted if the patient is receiving exogenous ACTH or any type of glucocorticoid therapy.

Overnight screening test — The overnight 1 mg DST is a quick screening test [2,3]. Dexamethasone (1 mg) is taken orally between 11 PM and midnight, and a single blood sample is drawn at 8 AM the next morning for assay of serum cortisol and, if available, serum dexamethasone. A dose of 0.3 mg/m² surface area can be used in children [4].

Dose adjustments for obese adults — We continue to suggest the standard 1 mg dose for the overnight screening DST. Although there have been concerns that obese individuals may require higher dexamethasone doses, data are conflicting:

●In one study of 34 normal weight (13 men and 21 women) and 87 obese (36 men and 51 women) adults, pituitary sensitivity to feedback inhibition by dexamethasone in the obese group was preserved, even at doses lower than 1 mg [5].

●However, in a second series of 100 consecutive obese adults (body mass index [BMI] >30 kg/m²), a false-positive rate of morning plasma cortisol >1.8 mcg/dL was found in 8 percent of individuals receiving 1 mg overnight dexamethasone, while it was only 2 percent in those receiving 2 mg overnight dexamethasone [6].

Normal results — Using current immunoassays (which are more specific for cortisol than older assays), most normal individuals suppress their 8 AM cortisol value to less than 2 mcg/dL (55 nmol/L) [7-9]. An occasional patient with Cushing's disease will suppress to 1.8 mcg/dL (50 nmol/L), so this cutoff point is used to maximize sensitivity [10,11]; it also appears to be valid when using liquid chromatography-tandem mass spectrometry (LC-MS/MS) to measure serum cortisol [12]. While structurally based assays such as LC-MS/MS fully discriminate between dexamethasone and cortisol, because serum dexamethasone levels are low and have limited crossreactivity in immunoassays, structurally based assays are not required.

Falsely normal results — Despite use of a diagnostic criterion chosen to improve sensitivity, some patients with Cushing's disease suppress the 8 AM serum cortisol concentration to less than 1.8 mcg/dL (50 nmol/L) after the overnight DST [8]. Because of this, it is important to obtain at least two screening tests. (See "Establishing the diagnosis of Cushing's syndrome".)

●A review of a single practitioner's experience in 103 patients with Cushing's syndrome showed that 6 of 80 (8 percent) suppressed to less than 2 mcg/dL (55 nmol/L) [13].

●In another study of 97 patients with Cushing's syndrome using a slightly higher 1.5 mg dexamethasone dose, four suppressed to between 1.2 and 2.8 mcg/dL (34 and 77 nmol/L) [14].

Falsely abnormal results — The 2008 Endocrine Society clinical guidelines suggest a diagnostic cortisol criterion of 1.8 mcg/dL (50 nmol/L), recognizing that this choice decreases specificity [11]. The use of progressively higher cutoff points will reduce falsely abnormal results at the expense of failing to identify patients with Cushing's syndrome.

●This was illustrated by an Italian investigation of 137 individuals suspected of Cushing's syndrome [15]. Using the cutoff of 1.8 mcg/dL (50 nmol/L) identified the 38 cases of Cushing's syndrome with 100 percent sensitivity and a specificity of 91 percent. When using a cutoff of 5 mcg/dL (137 nmol/L), the sensitivity decreased to 87 percent, but the specificity increased to 97 percent.

●Using a serum cortisol criterion of less than 3.6 mcg/dL (99 nmol/L), the test has a 12 to 15 percent false-positive rate [2,3,16]. If, however, the criterion for suppression of serum cortisol is increased to less than 7.2 mcg/dL (200 nmol/L), the false-positive rate decreases to 7 percent [3].

Morning serum cortisol values between 1.8 and 7.2 mcg/dL could be false positive due to a variety of factors including failure to take or absorb the dexamethasone properly [17]. Simultaneous measurement of serum dexamethasone can identify such cases [18] (see 'Use of dexamethasone measurements' below). Because of the possibility of a false result, the 1 mg low-dose dexamethasone test should not be used as the sole criterion for excluding or establishing the diagnosis of Cushing's syndrome. (See "Establishing the diagnosis of Cushing's syndrome", section on 'Initial testing'.)

Interpretation based upon assay used — Older immunoassays with antibodies that crossreact with other glucocorticoids in addition to cortisol may have a cutoff of 8 AM normal serum cortisol concentration less than 5 mcg/dL (138 nmol/L) [17]. However, modern immunoassays using monoclonal antibodies that have minimal crossreactivity to endogenous steroids are likely to yield results similar to LC-MS/MS and have a cutoff of <1.8 mcg/dL [10,19,20]. (See "Initial testing for adrenal insufficiency: Basal cortisol and the ACTH stimulation test".)

We do not recommend the routine use of salivary cortisol measurements for the overnight 1 mg DST, because results may vary between assays (see "Measurement of cortisol in serum and saliva"). Although most salivary cortisol immunoassays are not affected by dexamethasone, a commonly used US Food and Drug Administration (FDA)-cleared salivary cortisol enzyme immunoassay has 19 percent crossreactivity with dexamethasone so it cannot be used for the DST [21]. If used, the assay should be validated for this purpose in patients with and without Cushing's syndrome to provide an appropriate criterion for its interpretation [22].

A criterion of less than 1 ng/mL (2.8 nmol/L) achieved 100 percent diagnostic accuracy for distinguishing between 27 patients with Cushing's syndrome and 64 normal controls [16]. Using an automated electrochemiluminescence immunoassay for salivary cortisol and a criterion of 0.71 ng/mL (2 nmol/L) to evaluate 40 patients with confirmed hypercortisolism, 45 patients with various adrenal masses, and 115 healthy subjects, the text had a sensitivity of 97 percent and specificity of 86 percent [23].

Two-day, low-dose test — The two-day test is used to assess suppressibility in patients with an equivocal overnight test or as an alternative to the overnight test. Dexamethasone 0.5 mg is taken orally every six hours, usually at 8 AM, 2 PM, 8 PM, and 2 AM on each day, for a total of eight doses. In this protocol, the total dose given during the test is 4 mg. This test is sometimes referred to as the two-day, 2 mg test, which refers to the number of days and the total daily dose (2 mg) (see "Establishing the diagnosis of Cushing's syndrome"). The dose can be modified in children who weigh less than approximately 45 kg [24].

Blood is drawn two or six hours after the last dose for measurement of cortisol and dexamethasone (and ACTH, if desired).

The normal response to the low-dose, two-day test consists of the following:

●When using specific assays to measure serum cortisol, the recommended criterion for suppression is a serum cortisol <1.8 mcg/dL (<50 nmol/L). Using a serum cortisol concentration <1.8 mcg/dL (50 nmol/L) at 24 or 48 hours as a criterion of normal suppression [25], one retrospective study correctly identified 98 percent of 245 patients with Cushing's syndrome [25]. (See "Measurement of cortisol in serum and saliva".)

●Smaller prospective studies using a serum cortisol concentration of 1.4 or 2.2 mcg/dL (38 or 60 nmol/L) at 48 hours as the exclusion criterion reported 90 to 100 percent sensitivity and 97 to 100 percent specificity in patients suspected of having Cushing's syndrome [26-28].

●Therefore, the use of the two-day, low-dose test has greater specificity at high sensitivity than the 1 mg overnight test. As a result, this is the better screening test, with the caveat that it requires more patient effort than the 1 mg test to achieve excellent diagnostic results. There is no single established criterion for interpretation of the 2-day test.

●Because the overnight test is easier to administer, it is often the preferred test in the United States, but the two-day test is used elsewhere because of its better performance. Although there are no formal studies validating this, the two-day, low-dose DST may be preferable to the overnight low-dose test when evaluating patients who are night-shift workers.

●Urinary corticosteroid endpoints provide suboptimal sensitivity and specificity [8,9,26]; serum cortisol endpoints should be used instead.

●Measuring plasma ACTH, if either of the tests is abnormal, gives an indication of the etiology of the hypercortisolism; it will usually be high normal or high in patients with the ectopic ACTH syndrome, within the normal range or elevated in those with Cushing's disease, and low or undetectable in those with a primary adrenal source. (See "Establishing the cause of Cushing's syndrome".)

HIGH-DOSE DSTs — The high-dose (8 mg) overnight dexamethasone suppression test (HDDST) can theoretically be used to determine the cause of corticotropin (ACTH)-dependent Cushing's syndrome (see "Establishing the cause of Cushing's syndrome"). The test is no longer used to identify patients with ACTH-independent causes of Cushing's syndrome. The HDDST takes advantage of the physiology of Cushing's disease, in which neoplastic ACTH secretion can retain partial responsiveness to glucocorticoid negative feedback inhibition so that cortisol does not suppress normally with either the overnight 1 mg or the two-day, low-dose test [29,30]. By increasing the dose of dexamethasone four- to eightfold (ie, 12 to 16 times the usual daily maintenance dose), ACTH secretion may be suppressed partially in most patients with Cushing's disease. In contrast, many nonpituitary tumors that produce ACTH ectopically (such as small cell lung carcinomas) are typically not responsive to glucocorticoid negative feedback; benign neuroendocrine tumors that secrete ACTH ectopically may suppress ACTH secretion following high-dose dexamethasone, similarly to pituitary corticotroph tumors.

Unfortunately, there are pituitary corticotroph adenomas that do not suppress with the HDDST (particularly those with high ACTH production), and ectopic tumors that do [31-33]. Therefore, the HDDST should be used with caution and only when inferior petrosal sinus sampling for ACTH is not available (see "Establishing the cause of Cushing's syndrome", section on 'Petrosal venous sinus catheterization'). The corticotropin-releasing hormone (CRH) stimulation test and clinical features (older age, male sex, rapid onset of symptoms, severe hypertension, hypokalemia, and the magnitude of the increases in urine free cortisol and plasma ACTH) also help to differentiate between pituitary and ectopic sources of ACTH if inferior petrosal sinus sampling is not available [31,34,35]. Therefore, the use of the HDDST as a primary test to differentiate Cushing's disease and ectopic ACTH is controversial.

There are several HDDSTs in use:

Overnight 8 mg test — Dexamethasone (8 mg) is taken orally between 11 PM and midnight. A single blood sample is drawn at 8:30 and 9 AM the day before and the next day for measurement of serum cortisol and, if one wishes, plasma ACTH and serum dexamethasone.

Interpretation — The test is interpreted by calculating the suppression of morning serum cortisol concentration on the day before and after dexamethasone administration, and considering a 50 to 80 percent or greater suppression to indicate Cushing's disease [36-39]. This approach yields variable results: 56 to 92 percent sensitivity and 57 to 100 percent specificity [36,37,39-45].

False-positive responses to the 8 mg DST occur in perhaps as many as one-half of patients with ACTH-secreting pulmonary neuroendocrine tumors (mostly benign) and occasionally other tumors respond to HDDST with decreased tumor secretion of ACTH and cortisol [34,46,47]. Such tumors present a difficult diagnostic problem because they are often occult and may mimic the clinical features of Cushing's disease without the short course, severe hypercortisolism and hypokalemia considered more typical of ectopic ACTH secretion from malignant tumors such as small cell lung carcinomas. (See "Lung neuroendocrine (carcinoid) tumors: Epidemiology, risk factors, classification, histology, diagnosis, and staging".)

Two-day, high-dose test — The patient collects at least one baseline 24-hour urine specimen, usually beginning at 8 AM. After the baseline collection is completed, the patient begins taking 2 mg of dexamethasone orally every six hours for a total of eight doses (total dose = 16 mg), usually at 8 AM, 2 PM, 8 PM, and 2 AM, and the urine collections are continued during the two days that dexamethasone is given. In practice, this test is often performed immediately after completing the low-dose DST (if the test is positive), and no intervening baseline urine collection is obtained. The urine collections are assayed for urinary cortisol and creatinine. In addition, a blood specimen can be collected six hours after the last dose of dexamethasone for measurement of cortisol, dexamethasone, and ACTH.

Interpretation — The original description of the test used a criterion of more than 50 percent suppression of urinary 17-hydroxycorticosteroids to distinguish Cushing's disease from adrenal causes of Cushing's syndrome [29]; however, 17-hydroxycorticoteroids are no longer in use. Many studies used this criterion when measuring urinary free cortisol. Using the criterion of more than 50 percent suppression, 81 to 88 percent of patients with Cushing's disease showed suppression, and 31 to 57 percent of patients with ectopic ACTH secretion failed to suppress, yielding a high number of false-positive results [1,9,40].

A few studies evaluated use of a more stringent urinary free cortisol criterion (80 or 90 percent) to improve specificity to 100 percent and found a sensitivity of 47 to 83 percent [9,40,46].

It is prudent to perform at least one test in addition to the HDDST, even in patients who meet criteria for suppression, as the specificity of this test is less than 100 percent [31]. (See "Establishing the cause of Cushing's syndrome".)

Other types of DSTs

Intravenous DSTs — Several versions of an intravenous dexamethasone suppression test (DST) have been utilized in some centers and serve both in the initial confirmation and diagnosis of Cushing's syndrome within one day, while avoiding the potential difficulties of drug compliance and absorption with oral dexamethasone. However, this approach has been evaluated in a limited number of patients, and we do not suggest its routine use.

Dexamethasone is infused at 1 mg/hour intravenously for four [48,49], five [50], or seven [51] hours. Plasma cortisol is suppressed to levels <1.4 to 3.0 mcg/dL (<38 to 83 nmol/L) in the evening and the following morning in normal or obese individuals, and it is above these levels at 9 AM in patients with Cushing's syndrome (>20 percent of baseline value or >4.7 mcg/dL [130 nmol/L]), with a sensitivity and specificity of 100 and 90 percent, respectively [49,51]. Using cortisol and ACTH values during the 4 mg intravenous DST, differentiation of Cushing's disease (n = 32) from functional hypercortisolism (pseudo-Cushing's n = 36) was achieved with a sensitivity of 100 percent and specificity of 83.3 percent [52].

SOURCES OF ERROR — There are several common sources of error in dexamethasone suppression tests (DSTs):

●Acute stress or illness – In general, DSTs should not be performed in patients who are critically ill and/or in the postoperative period. Acute stress, infection, and the immediate postoperative period will activate the hypothalamic-pituitary-adrenal (HPA) axis and suppression with dexamethasone may not be normal in such circumstances.

●Elevated corticosteroid-binding globulin – Corticosteroid-binding globulin (CBG) levels are of concern in DSTs that use serum cortisol as an endpoint. Increased estrogen levels lead to elevated CBG levels with concomitantly increased serum cortisol values. This occurs with pregnancy or with high doses of oral estrogen (eg, ≥30 mcg ethinyl estradiol), as used in some oral contraceptives. This may result in an apparent lack of suppression to the low-dose DST [53]. In this setting, salivary cortisol values, which reflect unbound cortisol, will show normal suppression [54]. It may be helpful to measure CBG in women taking oral estrogens who have an abnormal response and to repeat the test after their discontinuation in patients who have a high CBG value. The time to normalization of CBG after discontinuation of oral estrogen is not established. Thus, it may be more practical to evaluate women taking exogenous estrogen using late-night salivary cortisol as a screening test [55,56]. Pregnancy is also a state of increased estrogen, CBG, and HPA axis activity [57-60]. In one study, CBG levels were still elevated in most women at two to three months postpartum [61]. The topic of the diagnosis of Cushing's syndrome in pregnancy is covered elsewhere. (See "Diagnosis and management of Cushing's syndrome during pregnancy".)

●Variation in dexamethasone intake or metabolism – Failure of the patient to take the dexamethasone or abnormal metabolism of the dexamethasone can interfere with interpretation of the test. Drugs that induce hepatic CYP3A4 enzymes, such as barbiturates, phenytoin, rifampin, and carbamazepine increase the metabolism of dexamethasone and other steroids [62,63]. Nearly one-half of the US Food and Drug Administration (FDA)-approved drugs in the United States interact with CYP3A4 and may confound the interpretation of any DST [64]. These effects can be detected by measuring serum dexamethasone at the appropriate interval after the last dose.

●Glucocorticoid receptor polymorphisms – Several glucocorticoid receptor polymorphisms have been identified that confer increased glucocorticoid sensitivity. Individuals with N363S, GR9b, or BclI polymorphisms have greater serum cortisol reduction after dexamethasone administration [65]. In contrast, the ER22/23EK and A3669G polymorphisms are associated with decreased glucocorticoid sensitivity and relative glucocorticoid resistance [66]. These glucocorticoid receptor polymorphisms are not determined on a clinical basis but may be implicated in some false-positive or negative response to dexamethasone tests [65].

●False-negative responses using the high-dose DST (HDDST) to identify patients with Cushing's disease are more common with higher baseline corticotropin (ACTH) and cortisol secretion. In less than 5 percent of patients with Cushing's disease, for example, higher doses of dexamethasone (16 to 100 mg/day) were required to produce significant suppression. These patients tend to have large tumors and more severe hypercortisolism [67-69].

USE OF DEXAMETHASONE MEASUREMENTS — Measuring serum dexamethasone is suggested for all dexamethasone suppression tests (DSTs). It provides verification that the drug was taken and indicates whether the serum concentration is within the limits expected in an individual who metabolizes the drug normally. Laboratory nomograms are available that relate serum dexamethasone concentrations to serum cortisol concentrations in normal subjects and in patients with Cushing's disease [17,70].

Commercial laboratories often provide a range of expected values for a specific dose and interval until the blood draw. Finding an abnormally high or low serum dexamethasone concentration allows one to interpret the cause of an unusual serum or urinary cortisol response and to repeat the test, if necessary, with the same or another dexamethasone dose.

Using modern liquid chromatography-tandem mass spectrometry (LC-MS/MS) assays, a serum dexamethasone concentration of 3.3 to 3.6 nmol/L is sufficient to get appropriate suppression of morning serum cortisol [12]. The reference range for serum dexamethasone measured at 8 AM following the 1 mg dose taken at 11 PM the night before is 140 to 295 ng/dL (3.6 to 7.5 nmol/L) [71].

USE OF DSTs IN SPECIAL POPULATIONS — The use of dexamethasone suppression tests (DSTs) in each of these populations is discussed further in their specific topics.

Adrenal incidentalomas — The low-dose DST is used to identify dysregulated cortisol secretion in patients with incidentally found adrenal masses. (See "Evaluation and management of the adrenal incidentaloma", section on 'Subclinical Cushing's syndrome'.)

Pregnant women — The low-dose 1 mg overnight DST is not recommended to diagnose Cushing's syndrome during pregnancy, because of the risk for false-positive results. Instead, the initial evaluation includes a late-night salivary cortisol and a 24-hour urinary cortisol. The 8 mg high-dose DST is sometimes performed to help determine etiology. The criteria for interpretation are different in these patients. (See "Diagnosis and management of Cushing's syndrome during pregnancy".)

Primary pigmented nodular adrenocortical disease (PPNAD) — A paradoxical increase in urinary free cortisol during the sequential low-dose (2 mg) and high-dose (8 mg) six-day dexamethasone suppression testing [21] may be seen in patients with Cushing's syndrome due to primary pigmented nodular adrenocortical disease (PPNAD) [72]. This delayed "paradoxical" response can be useful to identify otherwise asymptomatic carriers in familial forms of PPNAD or to distinguish PPNAD from other adrenocortical tumors. (See "Cushing's syndrome due to primary pigmented nodular adrenocortical disease".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Diagnosis and treatment of Cushing's syndrome".)

SUMMARY AND RECOMMENDATIONS

●The low-dose dexamethasone suppression tests (DSTs; 1 mg overnight and two-day, low-dose) are used to differentiate patients with Cushing's syndrome of any cause from patients who do not have Cushing's syndrome.

●The high-dose (8 mg) DSTs (HDDSTs) are used for the differential diagnosis of corticotropin (ACTH)-dependent Cushing's syndrome. The HDDST may be used with caution when inferior petrosal sinus sampling for ACTH is not available.

●Simultaneous measure of serum dexamethasone concentration is very useful to ensure that the dexamethasone was taken, absorbed, and reached sufficient blood levels to exert glucocorticoid negative feedback.

●Because states of increased estrogen exposure (eg, pregnancy, oral contraceptives, and menopausal hormone therapy) may stimulate corticosteroid-binding globulin (CBG) production, results using serum cortisol as an endpoint may be falsely abnormal.

DISCLOSURE — The views expressed in this topic are those of the author(s) and do not reflect the official views or policy of the United States Government or its components.