Diagnosis of adrenal insufficiency in adults

INTRODUCTION — The clinical presentation of adrenal insufficiency is variable, depending on whether the onset is acute, leading to adrenal crisis, or chronic, with symptoms that are insidious and vague. The diagnosis of adrenal insufficiency depends upon a critical level of clinical suspicion. Measurements of serum cortisol within three hours after wakening (eg, 6 to 9 AM) can suggest the diagnosis of adrenal insufficiency. However, these measurements often cannot definitively establish the presence or absence of adrenal insufficiency. The ACTH stimulation test is the best test to diagnose or exclude adrenal insufficiency when the baseline cortisol value is indeterminate.

This topic provides an overview of testing for diagnosis of adrenal insufficiency in adults. The clinical manifestations, evaluation to determine the etiology, and management of adrenal insufficiency are reviewed in detail separately.

●(See "Clinical manifestations of adrenal insufficiency in adults".)

●(See "Determining the etiology of adrenal insufficiency in adults".)

●(See "Treatment of adrenal insufficiency in adults".)

WHOM TO TEST — Diagnostic testing should be performed in adults with signs and symptoms of adrenal insufficiency. The symptoms of adrenal insufficiency may be subtle, and the diagnosis of adrenal insufficiency depends upon a critical level of clinical suspicion. (See "Clinical manifestations of adrenal insufficiency in adults".)

●Adrenal crisis – Adrenal crisis should be suspected in patients who present with peripheral vascular collapse (vasodilatory shock), whether or not the patient is known to have adrenal insufficiency (table 1). (See "Clinical manifestations of adrenal insufficiency in adults", section on 'Adrenal crisis'.)

●Chronic adrenal insufficiency – Chronic adrenal insufficiency (primary or central) should be suspected in patients with a combination of fatigue, weakness, myalgias, arthralgias, anorexia, and/or weight loss (table 2). Hyponatremia may be present in patients with primary or central adrenal insufficiency. Hyperkalemia is only seen in primary adrenal insufficiency, as the renin-angiotensin-aldosterone system is intact in secondary adrenal insufficiency. Additionally, postural hypotension, salt craving, and hyperpigmentation may be present only in chronic primary adrenal insufficiency. (See "Clinical manifestations of adrenal insufficiency in adults", section on 'Chronic adrenal insufficiency'.)

●Isolated corticotropin (ACTH) deficiency – Isolated ACTH deficiency, although rare, should be considered in any adult patient who has unexplained severe hypoglycemia or hyponatremia. (See "Causes of secondary and tertiary adrenal insufficiency in adults", section on 'Isolated ACTH deficiency'.)

●Patients taking chronic glucocorticoid therapy – Administration of exogenous glucocorticoids (eg, for rheumatic diseases) can suppress the hypothalamic-pituitary-adrenal (HPA) axis.

•Glucocorticoid withdrawal – Abrupt cessation or reduction of therapeutic doses of glucocorticoids to subphysiologic replacement doses may cause symptoms of adrenal insufficiency. Identifying central adrenal insufficiency in this population is reviewed separately. (See "Glucocorticoid withdrawal", section on 'Identifying patients with HPA suppression'.)

•Preoperative evaluation – Chronic glucocorticoid treatment may prevent the adrenal glands from responding appropriately during times of stress (eg, surgery). The approach to preoperative testing for HPA axis suppression should be individualized. Most patients who receive therapeutic dose of glucocorticoids do not routinely require evaluation of the HPA axis (or perioperative stress doses of glucocorticoids), provided they continue to receive their usual daily dose of glucocorticoid (or its equivalent parenteral dose). The decision to perform preoperative evaluation of the HPA axis and to give additional perioperative glucocorticoid coverage is based upon the patient's history of glucocorticoid intake and likelihood of HPA axis suppression, as well as the type and duration of surgery. This issue is reviewed in detail separately. (See "The management of the surgical patient taking glucocorticoids", section on 'Approach based upon HPA axis suppression' and "The management of the surgical patient taking glucocorticoids", section on 'Intermediate patients'.)

Perioperative management of patients taking chronic glucocorticoids for the treatment of primary or central adrenal insufficiency is also reviewed separately. (See "Treatment of adrenal insufficiency in adults", section on 'Surgery'.)

DIAGNOSTIC TESTING

Biochemical tests

●Basal and corticotropin (ACTH)-stimulated serum cortisol – The diagnosis of adrenal insufficiency depends on the demonstration of inappropriately low cortisol production. Total serum cortisol is the preferred measurement, except in patients with abnormalities of corticosteroid-binding globulin (CBG) or albumin (eg, patients with decreased CBG due to advanced cirrhosis or nephrotic syndrome, or those with increased CBG due to oral estrogens) [1]. In the setting of abnormal CBG, the usual serum cortisol thresholds may lead to an incorrect diagnosis. Measurements of salivary cortisol may be useful in patients with abnormalities of CBG, although few laboratories have established ACTH-stimulated salivary cortisol thresholds. (See 'Abnormal CBG' below and 'Basal salivary cortisol' below and 'ACTH-stimulated salivary cortisol' below.)

Although basal cortisol levels can suggest the diagnosis of adrenal insufficiency, these measurements often cannot definitively establish the presence or absence of adrenal insufficiency, particularly in the outpatient setting. Cortisol secretion is episodic [2]. A single early morning serum cortisol value may be inconclusive [3]. When the baseline cortisol value is inconclusive or if the clinical assessment is discordant with the cortisol findings, an ACTH stimulation test is needed to diagnose or exclude adrenal insufficiency. (See 'Basal serum cortisol testing' below and 'ACTH stimulation tests' below.)

●Aldosterone and renin – Some UpToDate contributors measure aldosterone and renin levels after a diagnosis of adrenal insufficiency is established to help distinguish between primary and central adrenal insufficiency. (See "Determining the etiology of adrenal insufficiency in adults", section on 'ACTH interpretation'.)

Other UpToDate contributors measure aldosterone and renin as part of the diagnostic evaluation of suspected primary adrenal insufficiency, in part because cortisol results can be equivocal. The morning cortisol reference range is broad, and a patient can have adrenal insufficiency but maintain basal ACTH and/or cortisol secretion within the range of normal. In the early evolution of primary adrenal insufficiency in particular, a low aldosterone and elevated renin level can precede overt cortisol deficiency and elevation in ACTH [3].

●Dehydroepiandrosterone sulfate (DHEAS) – We typically measure DHEAS when initial biochemical testing with cortisol and ACTH is indeterminate. However, some UpToDate contributors measure DHEAS in all patients with suspected adrenal insufficiency. (See 'DHEAS' below.)

Influence of cortisol assay technique — The interpretation of the cortisol result depends on the type of assay that is used to measure it [4-10]. Available techniques include the following:

●Immunoassays using polyclonal antibodies – These antibodies cross-react with non-cortisol glucocorticoids, and, therefore, these older assays give a higher cortisol result than assays using more specific monoclonal antibodies or liquid chromatography-tandem mass spectrometry (LC-MS/MS).

●Immunoassays using a more specific monoclonal antibody to cortisol – When using these more specific (less cross-reactive) assays, the morning cortisol threshold to exclude adrenal insufficiency is approximately 25 to 30 percent lower than that used with older immunoassays.

●LC-MS/MS – LC-MS/MS is a structurally based technique that does not use antibodies. It is considered to provide the most accurate measurement of cortisol. The threshold to exclude adrenal insufficiency is approximately 25 to 30 percent lower than that used with older, less specific immunoassays. LC-MS/MS is not routinely used for serum cortisol analysis.

The importance of the type of cortisol assay employed for diagnostic testing has not been evaluated in the context of adrenal crisis (see 'Suspected adrenal crisis' below). However, in patients with suspected chronic adrenal insufficiency, the interpretation of the serum cortisol result depends on the assay used to measure it. (See 'Suspected chronic adrenal insufficiency' below.)

BASAL SERUM CORTISOL TESTING — Basal serum cortisol levels can be used to exclude the diagnosis of adrenal insufficiency and to select patients for additional evaluation with corticotropin (ACTH) stimulation testing [11]. Test results must always be interpreted in conjunction with clinical assessment of the likelihood of adrenal insufficiency in an individual patient.

Suspected adrenal crisis — For patients with suspected adrenal crisis, we measure the serum cortisol and plasma ACTH level after inserting an intravenous (IV) line. The cortisol does not need to be measured in the early morning in the setting of suspected adrenal crisis. Therapy with IV saline and glucocorticoid should be initiated immediately. Treatment of patients with suspected adrenal crisis should not be delayed pending results. (See "Treatment of adrenal insufficiency in adults", section on 'Adrenal crisis'.)

●Cortisol <18 mcg/dL – In hypotensive patients considered to be in adrenal crisis, a basal serum cortisol level <18 mcg/dL (<500 nmol/L) suggests the diagnosis of adrenal insufficiency. To confirm the diagnosis of adrenal insufficiency, ACTH stimulation testing should be performed at a later date. If supraphysiologic doses of glucocorticoids (40 to 200 mg of hydrocortisone daily) have not been given for more than a few days, an ACTH stimulation test can be performed after initiation of glucocorticoid treatment. Withhold glucocorticoids on the morning of the test and resume after the test is completed while awaiting results. (See 'ACTH stimulation tests' below.)

The plasma ACTH level is useful in determining the etiology of adrenal insufficiency. (See "Determining the etiology of adrenal insufficiency in adults", section on 'Establish the level of defect'.)

●Cortisol ≥18 mcg/dL – If the basal serum cortisol level is ≥18 mcg/dL (≥500 nmol/L), adrenal insufficiency is unlikely. However, caution should be used when interpreting the cortisol results in patients with increased corticosteroid-binding globulin (CBG) levels, as may occur in pregnant individuals and those taking oral estrogens (eg, oral contraceptives, especially at a dose of 30 mcg or higher). Patients with CBG excess may have adrenal insufficiency even with apparently normal cortisol values. In the setting of CBG excess, therefore, the diagnosis of adrenal insufficiency may be missed. (See 'Abnormal CBG' below.)

Suspected chronic adrenal insufficiency — For patients with suspected chronic adrenal insufficiency, we measure a serum cortisol level within three hours after wakening (eg, 6 to 9 AM) (algorithm 1). When the clinical suspicion for chronic adrenal insufficiency is high, we obtain both serum cortisol and plasma ACTH levels simultaneously. If a diagnosis of adrenal insufficiency is established, the ACTH level is used to distinguish between primary and central adrenal insufficiency (algorithm 2). If the ACTH is indeterminate, measurement of aldosterone and renin may help distinguish between the two possibilities. (See 'Biochemical tests' above and "Determining the etiology of adrenal insufficiency in adults", section on 'ACTH interpretation'.)

●Morning cortisol thresholds – The interpretation of the serum cortisol result depends on the assay used to measure it. (See 'Influence of cortisol assay technique' above.)

When using the more specific (less cross-reactive) assays, the morning cortisol threshold to exclude chronic adrenal insufficiency is approximately 25 to 30 percent lower than the older cortisol assays (eg, approximately 13.0 to 14.0 mcg/dL instead of 18 mcg/dL [360 to 390 nmol/L instead of 500 nmol/L]) [4,9,12]. As an example, in a study comparing results of three different cortisol immunoassays, the lowest morning serum cortisol concentration that excluded adrenal insufficiency ranged from 12.1 mcg/dL to 18.2 mcg/dL (336 nmol/L to 506 nmol/L) [4]. A value below 2 to 3.7 mcg/dL (56 to 102 nmol/L) was consistent with adrenal insufficiency. In another study using two different cortisol assays, the morning serum cortisol threshold to exclude adrenal insufficiency ranged from >12.4 to >14.5 mcg/dL (340 to 400 nmol/L) [12].

We use the following thresholds for interpretation of early morning (6 to 9 AM or within three hours of morning waking) serum cortisol values (algorithm 1). Clinical judgement is used in conjunction with the cortisol results to determine whether ACTH stimulation testing should be performed. As an example, if there is discordance between symptoms and early morning cortisol levels, an ACTH test should always be performed. (See 'ACTH stimulation tests' below.)

•Cortisol ≤3 mcg/dL – A low early morning serum cortisol concentration (≤3 mcg/dL [80 nmol/L]) in the absence of CBG deficiency is consistent with adrenal insufficiency [13-15]. When multiple symptoms correlate with the low cortisol value, ACTH stimulation testing typically is not needed to confirm the diagnosis. (Related Lab Interpretation Monograph(s): "Low cortisol in adults".)

If symptoms do not correlate, perform an ACTH stimulation test. Normal serum cortisol responses to ACTH and insulin tolerance test (ITT) have been demonstrated for baseline serum cortisol concentration as low as approximately 1.2 to 5 mcg/dL (34 to 125 nmol/L) [1,7,9,16-19].

•Cortisol >3 mcg/dL but <18 mcg/dL – If the morning serum cortisol concentration is >3 mcg/dL (>80 nmol/L) but <18 mcg/dL (<500 nmol/L) using an older cortisol assay, perform a standard ACTH stimulation test. With less cross-reactive cortisol assays, the upper threshold is 13 to 14 mcg/dL (360 to 390 nmol/L) instead of 18 mcg/dL (ie, perform ACTH stimulation test if cortisol is >3 mcg/dL but <13 to 14 mcg/dL). (See 'ACTH stimulation tests' below.)

•Cortisol ≥18 mcg/dL – An early morning serum cortisol concentration ≥18 mcg/dL (≥500 nmol/L) in the absence of CBG excess (eg, due to oral estrogen use) excludes adrenal insufficiency [20]. When cortisol is measured using less cross-reactive assays (eg, using more specific immunoassays or liquid chromatography-tandem mass spectrometry [LC-MS/MS]), the morning cortisol threshold to exclude adrenal insufficiency is approximately 25 to 30 percent lower (eg, 13 to 14 mcg/dL instead of 18 mcg/dL [360 to 390 nmol/L instead of 500 nmol/L]). (See 'Abnormal CBG' below.)

DHEAS — We typically measure dehydroepiandrosterone sulfate (DHEAS) when biochemical testing with cortisol and corticotropin (ACTH) is indeterminate. However, some UpToDate contributors measure DHEAS in all patients with suspected adrenal insufficiency.

A normal DHEAS (using an age- and sex-specific reference range) makes adrenal insufficiency unlikely. DHEAS levels are low in people with primary or central adrenal insufficiency [21]. Although a low serum DHEAS may raise the clinical suspicion of adrenal insufficiency or support the diagnosis, it is not used as a single test to diagnose adrenal insufficiency. Patients with indeterminate cortisol levels and low (or indeterminate) serum DHEAS should undergo ACTH stimulation testing. (See 'Standard high-dose ACTH stimulation test' below.)

ACTH STIMULATION TESTS

Standard high-dose ACTH stimulation test — A standard high-dose (250 mcg) corticotropin (ACTH) stimulation test should be performed to establish the diagnosis of adrenal insufficiency in patients with equivocal serum cortisol values in whom hypoadrenalism is suspected. Patients taking glucocorticoids must withhold them on the morning of the test and resume after the test is completed while awaiting results. (See 'Suspected adrenal crisis' above.)

The agent used is synthetic ACTH(1-24) (cosyntropin; tetracosactin), which has the full biologic potency of native ACTH(1-39). Administration of a 250 mcg dose of ACTH(1-24) results in supraphysiologic plasma ACTH concentrations of approximately 60,000 pg/mL (1320 pmol/L) [22,23] and thus evaluates the maximal cortisol secretory capacity. The physiologic basis of the test is that adrenocortical destruction (in primary adrenal insufficiency) or adrenocortical atrophy (due to loss of the trophic action of endogenous ACTH) will lead to a subnormal cortisol response to a pharmacologic dose of synthetic ACTH.

If the loss of ACTH trophic action is recent (eg, within four to six weeks after pituitary surgery), the ACTH stimulation test may be falsely normal if adrenal atrophy has not occurred. Other diagnostic tests (eg, insulin-induced hypoglycemia, metyrapone-induced hypocortisolemia) may be warranted in this setting. (See 'Recent ACTH deficiency' below and "Diagnostic testing for hypopituitarism", section on 'Corticotropin'.)

Generally, there are no adverse effects of the stimulation test. Allergic reactions are extraordinarily rare [11].

Procedure — The standard high-dose (250 mcg) ACTH stimulation test can be performed at any time of day when using the serum cortisol response to ACTH injection to establish the diagnosis of adrenal insufficiency. The testing is performed as follows:

●Obtain baseline serum cortisol.

●Administer cosyntropin 250 mcg (85 nmol, or 40 international units [the entire contents of the vial]) via intravenous (IV) or intramuscular (IM) injection [24].

●Obtain serum cortisol again at 30 and 60 minutes post-injection.

If baseline ACTH, renin, and aldosterone levels are being obtained with the stimulation test (to determine the etiology of adrenal insufficiency), the test must be performed in the morning, between 6 and 9 AM, as the normal decrease in hormone levels later in the day overlaps with values observed in central adrenal insufficiency. Measurement of serum electrolytes and plasma renin activity in the basal sample can be useful because they may help to interpret measurement of basal aldosterone. In primary adrenal insufficiency, all adrenal zones are typically affected, and aldosterone will be low with an elevated renin level. In central adrenal insufficiency, the renin-angiotensin system is intact, and plasma renin and aldosterone are usually unaffected [25]. (See "Determining the etiology of adrenal insufficiency in adults", section on 'ACTH interpretation'.)

Interpretation of results — Only the absolute thresholds are used to exclude or diagnose adrenal insufficiency. Incremental or percent increases in cortisol should not be used to assess the response to ACTH [26], because individuals who have a high basal serum cortisol concentration, due either to normal circadian rhythmicity or acute stress, may be nearly maximally stimulated and unable to increase cortisol secretion further. In addition, approximately 20 percent of healthy individuals with high-normal basal serum cortisol concentration have little or no rise after ACTH.

●Serum cortisol – Interpretation of the ACTH stimulation test is based on the peak cortisol value, which typically occurs 60 minutes after administration (but may occur at 30 minutes, particularly when cosyntropin is administered intravenously) (algorithm 1) [27,28]. In general, the 30-minute cutoff is lower than the 60-minute value regardless of the cortisol assay method [29,30].

The ACTH-stimulated cortisol thresholds to exclude adrenal insufficiency vary with the assay used to measure cortisol (table 3) [8,16,31] (see 'Influence of cortisol assay technique' above). In studies evaluating the less cross-reactive cortisol assays to establish diagnostic thresholds for ACTH stimulation testing, peak cortisol levels that excluded adrenal insufficiency ranged from 14 to 18 mcg/dL (390 to 500 nmol/L) [16,18,29,30,32]. Using a peak ACTH-stimulated cortisol threshold of 17 to 18 mcg/dL (470 to 500 nmol/L) to exclude adrenal insufficiency is unlikely to miss a diagnosis of adrenal insufficiency, but it may lead to overdiagnosis in the more specific assays.

Our approach to interpreting the peak serum cortisol value after ACTH administration is as follows (algorithm 1) [16,18,29,30,32]:

•<14 mcg/dL (390 nmol/L) – Adrenal insufficiency likely.

•≥14 to <18 mcg/dL (390 to 500 nmol/L) – The interpretation of the cortisol result depends on the type of assay that is used to measure it (table 3) and the clinical likelihood of adrenal insufficiency in an individual patient. If the assay type is unknown or diagnosis is uncertain, a DHEAS level may be useful. A normal DHEAS (using an age- and sex-specific reference range) makes adrenal insufficiency unlikely. (See 'DHEAS' above.)

•≥18 mcg/dL (500 nmol/L) – Adrenal insufficiency generally excluded regardless of assay used to measure cortisol.

Exceptions include measurements obtained within four to six weeks of interventions (such as pituitary surgery or radiation) that may damage the hypothalamus or pituitary, or measurements obtained in patients with abnormal corticosteroid-binding globulin (CBG) levels. (See 'Recent ACTH deficiency' below and 'Abnormal CBG' below.)

After cortisol deficiency is diagnosed, determine the level of the defect (algorithm 2) and the etiology, if unknown. Determining the etiology of adrenal insufficiency is reviewed in detail separately. (See "Determining the etiology of adrenal insufficiency in adults" and "Causes of secondary and tertiary adrenal insufficiency in adults" and "Causes of primary adrenal insufficiency (Addison disease)".)

Low-dose ACTH stimulation test not recommended — Although the low-dose (1 mcg) corticotropin (ACTH) stimulation test had been proposed as a better test for central adrenal insufficiency, we do not typically use the low-dose stimulation test in clinical practice for a number of reasons:

●It does not provide better diagnostic accuracy than the standard test [3].

●There are no commercially available preparations of "low-dose" cosyntropin, and it is difficult to precisely measure diluted ACTH(1-24) for injection. A small error in dilution or injection volume or a small loss of the drug during the injection can result in a physiologically significant variability in the plasma ACTH levels achieved [33-35].

●The aldosterone response is considered to be unreliable [33,36-38].

●The 1 mcg low-dose test has not been validated after IM injection and may not provide valid results by this route [39,40]. Additionally, criteria for interpretation are not available for patients whose body surface area is significantly larger or smaller than 1.73 m².

SPECIAL SITUATIONS WHEN STANDARD TESTING MAY BE UNRELIABLE

There are several situations when standard testing may be unreliable.

Abnormal CBG — In patients with abnormalities of corticosteroid-binding globulin (CBG) or albumin, such as patients taking oral estrogens (high CBG) or those with cirrhosis or nephrotic syndrome (low CBG), the usual cortisol thresholds may lead to an incorrect diagnosis [41-43]. Serum free cortisol has been suggested as an alternative, but the assay is not widely available, and criteria for response have not been well developed [44]. Salivary cortisol correlates well with serum free cortisol [41]. Measurements of salivary cortisol are not affected significantly by oral contraceptives or in the presence of low CBG (eg, cirrhosis), but assay-specific cutoff points for morning and stimulated values must be used [41,45,46]. (See 'Basal salivary cortisol' below and 'ACTH-stimulated salivary cortisol' below and "Measurement of cortisol in serum and saliva", section on 'Measurement of salivary cortisol concentration'.)

If the clinical suspicion for adrenal insufficiency is high and available tests to diagnose adrenal insufficiency are indeterminate due to the CBG abnormality, we generally treat with glucocorticoids until definitive biochemical testing can be performed.

High CBG — In the setting of possible high corticosteroid-binding globulin (CBG; eg, oral contraceptive use, pregnancy), it is useful to measure CBG. If normal, the usual testing can be pursued. If CBG is elevated, we measure an early morning serum cortisol. Only a frankly subnormal cortisol value (≤3 mcg/dL [83 nmol/L]) is useful to make the diagnosis of adrenal insufficiency. Values >3 mcg/dL neither confirm nor exclude a diagnosis of adrenal insufficiency. If the serum cortisol is >3 mcg/dL, salivary cortisol testing (if available) can be used to assess for adrenal insufficiency. Assay-specific cutoff points for basal or corticotropin (ACTH)-stimulated values must be used [45,46]. (See 'Basal salivary cortisol' below and 'ACTH-stimulated salivary cortisol' below.)

For women taking oral contraceptives, another option is to discontinue oral contraceptives for six to eight weeks (if able) and perform an ACTH stimulation test. (See 'Standard high-dose ACTH stimulation test' above.)

The influence of estrogens on serum total cortisol measurement is illustrated by the following findings [8,43,46,47]:

●In women taking oral contraceptives, the serum total cortisol is approximately 1.5 times higher than women not taking oral contraceptives [47]. In one study evaluating the normal response to ACTH in 100 healthy adults (50 men and 50 women) and 13 women taking oral contraceptives, 30-minute total cortisol values were two-fold higher and free cortisol values were lower in the oral contraceptive group compared with controls [43]. A second study of healthy women reported a 1.7 to 2.2- and a 1.3 to 1.6-fold elevation in basal and stimulated total cortisol levels, respectively, in women treated with oral contraceptives compared with those who were not [8] and suggested that women stop oral contraceptive agents before undergoing ACTH stimulation testing.

●In contrast, transdermal estrogen has less of an influence on CBG and does not appear to increase serum total cortisol [47]. Therefore, it may be unnecessary to discontinue transdermal estrogen prior to evaluation with a morning serum cortisol or ACTH stimulation test.

●During pregnancy, there is a progressive rise in total cortisol, free cortisol, and CBG. During the third trimester, morning serum total cortisol concentrations may be 50 mcg/dL (1400 nmol/L) or higher. The impact of high CBG levels on cortisol assessment during pregnancy is reviewed separately. (See "Diagnosis and management of Cushing syndrome during pregnancy", section on 'HPA changes in normal pregnancy'.)

Low CBG — Serum cortisol in patients with suspected low corticosteroid-binding globulin (CBG) due to nephrotic syndrome, sepsis, or cirrhosis should be interpreted with caution. In the appropriate clinical setting, a morning cortisol above the assay-specific threshold to exclude adrenal insufficiency can exclude adrenal insufficiency.

Patients with cirrhosis or nephrotic syndrome who have low CBG values may have "subnormal" morning cortisol [48] and cortisol response to ACTH [41] in the absence of adrenal insufficiency [43]. In this setting, we prefer using salivary cortisol levels to assess for adrenal insufficiency. Assay-specific cutoff points for basal or ACTH-stimulated values must be used [41]. (See 'Basal salivary cortisol' below and 'ACTH-stimulated salivary cortisol' below.)

Tests not affected by abnormal CBG

Basal salivary cortisol — Measurements of salivary cortisol may be useful in patients with abnormalities of corticosteroid-binding globulin (CBG) because salivary cortisol levels more accurately reflect serum free cortisol concentrations than do measurements of serum total cortisol. (See 'Abnormal CBG' above.)

Saliva is obtained after rinsing the mouth but before brushing the teeth. A cotton pledget is placed between the gum and cheek or under the tongue until it is saturated with saliva and feels "slimy." Patients should be cautioned not to use lip products that prevent chapping or are colored (gloss, lipstick), as some of these contain hydrocortisone and will contaminate the specimen. They should also not remove the pledget with their fingers if they use any type of lotion, which may also contain hydrocortisone. The saturated cotton pledget is placed in the container by protruding it from the mouth, rather than with the fingers. The container can be stored at room temperature for many days or frozen for extended periods.

Morning salivary cortisol concentrations are decreased in adrenal insufficiency. Each salivary cortisol assay has a different reference range. The clinician must be familiar with the normative values for the assay being used. In one study using liquid chromatography-tandem mass spectrometry (LC-MS/MS), the normal reference range for salivary cortisol was 0.06 to 1.25 mcg/dL (1.6 to 34.4 nmol/L) [49]. A morning value <0.04 mcg/dL (<1 nmol/L) was diagnostic of adrenal insufficiency, and a value >0.21 mcg/dL (>5.9 nmol/L) excluded it.

ACTH-stimulated salivary cortisol — Salivary cortisol can also be measured during this test and is particularly useful if venous access is problematic or if plasma proteins and presumably CBG levels are abnormal [2,45]. However, there is less experience with it than with serum cortisol [50,51], and many laboratories do not have firmly established corticotropin (ACTH)-stimulated salivary cortisol cutoffs. (See "Measurement of cortisol in serum and saliva".)

●In a study measuring salivary cortisol by immunoassay, a conservative cutoff to exclude adrenal insufficiency was >0.9 mcg/dL (>26 nmol/L) one hour after injection of ACTH [52].

●In a study using LC-MS/MS, the ACTH-stimulated salivary cortisol threshold to exclude adrenal insufficiency was a peak value >0.6 mcg/dL (>17.2 nmol/L) [49]. The thresholds to establish adrenal insufficiency were <0.1 and <0.3 mcg/dL (<3.2 and <9.5 nmol/L) at 30 and 60 minutes, respectively.

Recent ACTH deficiency

After pituitary surgery or trauma — Recent corticotropin (ACTH) deficiency, as may occur soon after pituitary surgery, pituitary apoplexy, or brain or pituitary stalk trauma, may result in a falsely reassuring ACTH stimulation test [53]. During the weeks after cessation of ACTH secretion, the zona fasciculata and reticularis of the adrenals undergo functional and anatomical atrophy. During this interval, the adrenals may respond normally to pharmacologic doses of ACTH. However, the pituitary gland is unable to release ACTH in response to stress. Therefore, patients with possible recent ACTH deficiency, even if they have a normal response to ACTH, should be given stress doses of glucocorticoids and saline during major stress or surgery until repeat ACTH stimulation testing four to six weeks after suspected onset of ACTH deficiency. (See "Treatment of adrenal insufficiency in adults" and "Diagnostic testing for hypopituitarism", section on 'Corticotropin'.)

After pituitary surgery, some experts routinely treat with glucocorticoid replacement until normal adrenal function is demonstrated four to six weeks after surgery. However, others measure serum cortisol on the third postoperative day, 24 hours after the previous dose of glucocorticoids, and only administer replacement glucocorticoids if the value is low (less than 5 mcg/dL [138 nmol/L]) or borderline (5 to 15 mcg/dL [138 to 469 nmol/L]). ACTH testing is performed four to six weeks after onset of adrenal insufficiency to determine if replacement glucocorticoids must be continued. (See "Transsphenoidal surgery for pituitary adenomas and other sellar masses", section on 'Perioperative management'.)

In rare patients for whom a confirmed diagnosis of adrenal insufficiency cannot be delayed, the insulin-induced hypoglycemia test may be performed to establish the diagnosis. This test is now rarely performed in clinical practice. (See "Insulin-induced hypoglycemia test protocol", section on 'Recent pituitary surgery'.)

Partial ACTH deficiency — Corticotropin (ACTH) deficiency occurring after pituitary surgery, hypothalamic-pituitary radiotherapy, trauma, or opioid use may be partial. Partial ACTH deficiency has also been reported after discontinuation of glucocorticoids, including inhaled preparations [54]. Partial ACTH deficiency usually causes symptoms during times of physical stress (eg, febrile illness) when corticotropin-releasing hormone (CRH) and/or ACTH secretion do not increase in response to the stress. However, basal ACTH production may be adequate to prevent adrenal gland atrophy and, therefore, ACTH-stimulated cortisol levels may be normal. (See "Diagnostic testing for hypopituitarism", section on 'Evaluate for ACTH deficiency'.)

A diagnosis of partial adrenal insufficiency is usually made on clinical grounds alone, typically in patients with a history of pituitary surgery or radiotherapy, who develop signs and symptoms of adrenal insufficiency exclusively during physiologic stress. However, if clinical evaluation is equivocal, the metyrapone test can be used to verify the diagnosis. (See "Diagnostic testing for hypopituitarism", section on 'Evaluate for ACTH deficiency' and "Metyrapone stimulation tests", section on 'Overnight, single-dose metyrapone test'.)

Opioid-induced adrenal insufficiency — Long-term opioid use for the treatment of chronic pain may cause chronic central adrenal insufficiency due to opioid suppression of the HPA axis [55,56]. Opioid-induced adrenal insufficiency should be suspected in people chronically taking opioids who have fatigue, postural hypotension, nausea, weight loss, and hypogonadism [57]. Opioid-induced central hypogonadism is more common than central adrenal insufficiency [56]. (See "Causes of secondary hypogonadism in males", section on 'Opioids'.)

The initial testing is the same as for any adrenal insufficiency. Opioid-induced adrenal insufficiency may be difficult to diagnose, particularly if the hypothalamic-pituitary suppression is recent as the ACTH stimulation test may show normal results. In this setting, other testing (eg, insulin-induced hypoglycemia test) or a trial of glucocorticoid replacement is warranted. (See 'Diagnostic testing' above and 'Standard high-dose ACTH stimulation test' above and "Insulin-induced hypoglycemia test protocol".)

Critical illness — Critically ill patients have reduced metabolic clearance of cortisol, leading to variable serum cortisol and ACTH concentrations [58,59]. There is little consensus about the definition of a "normal" or "appropriate" cortisol level during critical illness. In addition, critically ill patients often have abnormal binding globulins, and laboratory assays of cortisol concentration and cortisol response to ACTH stimulation are generally unreliable. Thus, most clinicians do not rely on laboratory testing to select glucocorticoid replacement therapy in patients with septic shock. This topic is reviewed in detail separately. (See "Glucocorticoid therapy in septic shock in adults".)

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: Adrenal insufficiency".)

SUMMARY AND RECOMMENDATIONS

●Whom to test – Diagnostic testing should be performed in patients with signs and symptoms of adrenal insufficiency (table 1 and table 2). The symptoms of adrenal insufficiency may be subtle, and the diagnosis of adrenal insufficiency depends upon a critical level of clinical suspicion. (See 'Whom to test' above and "Clinical manifestations of adrenal insufficiency in adults".)

●Basal serum cortisol testing

•Suspected adrenal crisis – For patients with suspected adrenal crisis, we obtain plasma corticotropin (ACTH) and serum cortisol levels (any time of day) and immediately begin therapy with intravenous (IV) saline and glucocorticoids. The treatment of patients who present in possible adrenal crisis should not be delayed while awaiting the results of diagnostic testing. (See 'Suspected adrenal crisis' above and "Treatment of adrenal insufficiency in adults", section on 'Adrenal crisis'.)

-Cortisol <18 mcg/dL – In hypotensive patients considered to be in adrenal crisis, a basal serum cortisol level <18 mcg/dL (<500 nmol/L) suggests the diagnosis of adrenal insufficiency. To confirm the diagnosis of adrenal insufficiency, ACTH stimulation testing should be performed at a later date. If supraphysiologic doses of glucocorticoids (40 to 200 mg of hydrocortisone daily) have not been given for more than a few days, an ACTH stimulation test can be performed after initiation of glucocorticoid treatment. Withhold glucocorticoids on the morning of the test, and resume after the test is completed while awaiting results.

-Cortisol ≥18 mcg/dL – In the absence of corticosteroid-binding globulin (CBG) excess, a basal serum cortisol level ≥18 mcg/dL (≥500 nmol/L) makes adrenal insufficiency unlikely.

•Suspected chronic adrenal insufficiency – For patients with suspected chronic adrenal insufficiency, we measure a serum cortisol level within three hours after wakening (eg, 6 to 9 AM) (algorithm 1). When the clinical suspicion for chronic adrenal insufficiency is high, we obtain both serum cortisol and plasma ACTH levels simultaneously. Once a diagnosis of adrenal insufficiency is established, the ACTH level is used to distinguish between primary and central adrenal insufficiency (algorithm 2). (See 'Suspected chronic adrenal insufficiency' above and "Determining the etiology of adrenal insufficiency in adults", section on 'ACTH interpretation'.)

Clinical judgement is used in conjunction with the cortisol results to determine whether ACTH stimulation testing should be performed. The interpretation of the serum cortisol result depends on the assay used to measure it. (See 'Influence of cortisol assay technique' above.)

-Cortisol ≤3 mcg/dL – A low early morning serum cortisol concentration (≤3 mcg/dL [80 nmol/L]) in the absence of CBG deficiency is consistent with adrenal insufficiency. Typically, ACTH stimulation testing is not needed to confirm the diagnosis in these settings, unless there is discordance between symptoms and the cortisol levels. (Related Lab Interpretation Monograph(s): "Low cortisol in adults".)

-Cortisol >3 mcg/dL but <18 mcg/dL – If the morning serum cortisol concentration is >3 mcg/dL (>80 nmol/L) but <18 mcg/dL (<500 nmol/L), perform a standard ACTH stimulation test. With less cross-reactive cortisol assays, the upper threshold is 13 to 14 mcg/dL (360 to 390 nmol/L) instead of 18 mcg/dL (ie, perform ACTH stimulation test if cortisol is >3 mcg/dL but <13 to 14 mcg/dL). (See 'ACTH stimulation tests' above.)

-Cortisol ≥18 mcg/dL – An early morning serum cortisol concentration ≥18 mcg/dL (≥500 nmol/L) in the absence of CBG excess excludes primary adrenal insufficiency and most patients with central adrenal insufficiency using many immunoassays. When cortisol is measured using less cross-reactive assays, the morning cortisol threshold to exclude adrenal insufficiency is approximately 25 to 30 percent lower (eg, 13 to 14 mcg/dL instead of 18 mcg/dL [360 to 390 nmol/L instead of 500 nmol/L]).

●Role of DHEAS measurements – We typically measure dehydroepiandrosterone sulfate (DHEAS) when initial biochemical testing with cortisol and ACTH is indeterminate (algorithm 1). However, some UpToDate contributors measure DHEAS in all patients with suspected adrenal insufficiency.

A normal DHEAS (using an age- and sex-specific reference range) makes adrenal insufficiency unlikely. DHEAS levels are low in people with primary or central adrenal insufficiency. Although a low serum DHEAS may raise the clinical suspicion of adrenal insufficiency or support the diagnosis, it is not used as a single test to diagnose adrenal insufficiency. Patients with indeterminate cortisol levels and low (or indeterminate) serum DHEAS should undergo ACTH stimulation testing. (See 'DHEAS' above.)

●ACTH stimulation tests – A standard high-dose (250 mcg) ACTH stimulation test should be performed to establish the diagnosis of adrenal insufficiency in patients with equivocal serum cortisol values in whom hypoadrenalism is suspected. (See 'ACTH stimulation tests' above.)

Interpretation of the ACTH stimulation test is based on the peak serum cortisol value, which typically occurs 60 minutes after administration (but may occur at 30 minutes). The ACTH-stimulated cortisol thresholds to exclude adrenal insufficiency vary with the assay used to measure cortisol (table 3). (See 'Interpretation of results' above.)

Our approach to interpreting the peak serum cortisol value after ACTH administration is as follows (algorithm 1):

•<14 mcg/dL (390 nmol/L) – Adrenal insufficiency likely.

•≥14 to <18 mcg/dL (390 to 500 nmol/L) – The interpretation of the cortisol result depends on the type of assay that is used to measure it (table 3) and the clinical likelihood of adrenal insufficiency in an individual patient. If the assay type is unknown or diagnosis is uncertain, a DHEAS level may be useful. A normal DHEAS (using an age- and sex-specific reference range) makes adrenal insufficiency unlikely. (See 'DHEAS' above.)

•≥18 mcg/dL (500 nmol/L) – Adrenal insufficiency generally excluded regardless of assay used to measure cortisol.

Exceptions include measurements obtained within four to six weeks of interventions (such as pituitary surgery or radiation) that may damage the hypothalamus or pituitary, or measurements obtained in patients with abnormal CBG levels. (See 'Interpretation of results' above and 'Recent ACTH deficiency' above and 'Abnormal CBG' above.)

●Abnormal CBG level – High CBG levels in women using oral contraceptives or low CBG levels in patients with nephrotic syndrome may confound interpretation of serum cortisol levels (basal or during ACTH test). In patients with CBG excess, only a frankly subnormal basal value (≤3 mcg/dL [83 nmol/L]) is useful to make the diagnosis of adrenal insufficiency. Basal values >3 mcg/dL neither confirm nor exclude a diagnosis of adrenal insufficiency. In patients with suspected low CBG due to nephrotic syndrome, sepsis, or cirrhosis, a morning cortisol above the assay-specific threshold to exclude adrenal insufficiency can exclude adrenal insufficiency. However, "subnormal" morning cortisol and cortisol response to ACTH may occur in the absence of adrenal insufficiency. The use of salivary cortisol or serum free cortisol can be helpful when CBG is abnormal. For women taking oral contraceptives, another option is to discontinue oral contraceptives for six to eight weeks (if able) and perform an ACTH stimulation test. (See 'Abnormal CBG' above.)

ACKNOWLEDGMENT — 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.