INTRODUCTION — Measurements of serum cortisol soon after wakening (eg, 6 to 8 AM in people who sleep at night) can suggest the diagnosis of adrenal insufficiency. However, these measurements often cannot establish the presence or absence of adrenal insufficiency with certainty and cannot distinguish among the three disturbances in the hypothalamic-pituitary-adrenal (HPA) axis that can cause deficient cortisol secretion [1,2]:
●A primary adrenal disorder resulting in deficiency of cortisol (primary adrenal insufficiency or Addison's disease; plasma corticotropin [ACTH] is increased due to the loss of cortisol negative feedback)
●A pituitary disorder resulting in deficiency of ACTH secretion (secondary adrenal insufficiency; plasma ACTH not increased or low due to the failure of the pituitary corticotrophs to increase ACTH secretion in the absence of glucocorticoid negative feedback)
●A hypothalamic disorder resulting in deficiency of corticotropin-releasing hormone (CRH) and secondarily of ACTH (tertiary adrenal insufficiency; plasma ACTH not increased or low due to the failure of the CRH neurons in the hypothalamus to increase activity in the absence of glucocorticoid negative feedback)
This topic will provide an overview of available laboratory tests for patients with suspected cortisol deficiency. The evaluation and management of patients with suspected adrenal crisis or chronic adrenal insufficiency are covered separately. (See "Diagnosis of adrenal insufficiency in adults" and "Treatment of adrenal insufficiency in adults".)
Possible adrenal crisis — It is essential that treatment of patients who present in possible adrenal crisis not be delayed while diagnostic tests are performed. Blood for plasma ACTH, serum cortisol, plasma renin activity or concentration, aldosterone, and basic or comprehensive metabolic panel should be drawn and therapy initiated immediately with intravenous (IV) saline and glucocorticoids. Hydrocortisone is the glucocorticoid of choice where it is available. Alternatives are dexamethasone or methylprednisolone if hydrocortisone is not available (see "Treatment of adrenal insufficiency in adults", section on 'Adrenal crisis'). If these results do not provide a clear diagnosis, further testing can be performed at a later date once the patient has recovered.
BASAL SERUM CORTISOL AND ACTH — An early morning serum cortisol value can be helpful in excluding chronic adrenal insufficiency. As shown in a study comparing results of three different immunoassays, no single criterion is applicable for this purpose. In that study 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). In contrast, a value below 3.6 to 4.4 mcg/dL (102 to 123 nmol/L) makes adrenal insufficiency very likely [3].
In addition to assay variability, a major problem with relying on unstimulated serum cortisol values as the basis for the diagnosis of insufficiency (particularly in ambulatory patients) is that cortisol secretion is episodic [4]. A single early morning serum cortisol value, if it falls within the reference range, is inconclusive, although a patient with adrenal insufficiency is very unlikely to have a high-normal value [5]. Furthermore, the morning cortisol reference range is broad, and a patient can have pituitary or adrenal insufficiency but maintain basal corticotropin (ACTH) and/or cortisol secretion within the range of normal. For these reasons, dynamic function tests should be performed when there is doubt about the status of hypothalamic-pituitary-adrenal (HPA) function.
ROLE OF DYNAMIC TESTING
Establish diagnosis — Dynamic testing is performed to establish the diagnosis in patients with equivocal serum cortisol values in whom hypoadrenalism is suspected. With simultaneous measurement of the basal plasma corticotropin (ACTH), these tests can usually distinguish between primary and secondary or tertiary adrenal insufficiency. Distinction between hypothalamic and pituitary causes of hypoadrenalism is made by assessing the response to corticotropin-releasing hormone (CRH), although this is rarely necessary since their treatment is similar. (See "Corticotropin-releasing hormone stimulation test".)
While the history and symptoms may provide clues to the possible etiology of adrenal deficiency, measurement of plasma ACTH and the responses of the adrenal cortex and the anterior pituitary gland to a variety of stimuli will confirm the diagnosis and define the site at which the defect exists. The initial approach is typically assessing the adrenocortical response to the injection of synthetic ACTH(1-24).
Rarely, hypothalamic-pituitary-adrenal (HPA) stimuli are needed, such as the CRH stimulation test, insulin-induced hypoglycemia, or metyrapone-induced hypocortisolemia. The high-dose ACTH stimulation tests have supplanted insulin-induced hypoglycemia in clinical practice except in rare circumstances [6]. In addition, ACTH stimulation testing is inexpensive and safe. In contrast to insulin-induced hypoglycemia, close monitoring by an advanced health care provider is not required. A low-dose ACTH stimulation test is more physiologic and may detect a more subtle deficiency in cortisol secretion than the high-dose test because plasma ACTH does not increase as much [7]. However, its utility in clinical practice is limited by the difficulty in precisely obtaining diluted ACTH(1-24) for injection, the possible loss of critical amounts of ACTH(1-24) in the injection apparatus, and the unreliability of the aldosterone response [8-11]. (See 'Low-dose ACTH stimulation test' below.)
ACTH STIMULATION TESTS — The physiologic basis of these tests is that adrenocortical destruction (in primary adrenal insufficiency) or adrenocortical atrophy (due to loss of the trophic action of endogenous corticotropin [ACTH]) will lead to a subnormal response to a pharmacologic dose of synthetic ACTH.
Several protocols have been used to assess the response to exogenous ACTH. The agent used is synthetic ACTH(1-24) (cosyntropin; tetracosactrin), which has the full biologic potency of native ACTH(1-39). Short (one hour or less) tests involve administration of a single "high" or "low" dose and can be performed on an outpatient basis. Both tests result in supraphysiologic plasma ACTH concentrations: approximately 60,000 pg/mL (1320 pmol/L) after the standard high-dose ACTH test and approximately 1900 pg/mL (41.8 pmol/L) after the low-dose test [12]. There are no untoward side effects. Allergic reactions are extraordinarily rare.
In healthy individuals, cortisol responses are greatest in the morning, but in patients with adrenal insufficiency, the response to cosyntropin is subnormal in the morning and afternoon [13]. Thus, we recommend that the test be done in the morning to minimize the risk of a false-positive result in an unaffected individual. This dose stimulates maximal adrenocortical secretion up to 30 to 60 minutes post injection and, in normal subjects, results in a peak plasma ACTH concentration approximately twice that of insulin-induced hypoglycemia [14].
Basal blood sample — It is often useful and most efficient to measure plasma ACTH in the basal sample because once glucocorticoid therapy is initiated, plasma ACTH levels can be difficult to interpret. Furthermore, measurement of serum electrolytes and plasma renin activity can be helpful in the basal sample because they may help to interpret measurement of basal and ACTH-stimulated aldosterone.
Standard high-dose ACTH stimulation test — This pharmacologic test evaluates the maximal cortisol secretory capacity. Typically, serum cortisol is first measured immediately before an intravenous (IV) or intramuscular (IM) injection of 250 mcg (85 nmol, or 40 international units) of cosyntropin, which represents the entire contents of the vial. Serum cortisol is drawn again at 30 minutes post-injection, or at both 30 and 60 minutes post-injection.
This dose of cosyntropin results in pharmacologic plasma ACTH concentrations for the 60-minute duration of the test. Measurement of basal and stimulated aldosterone concentrations can be helpful to distinguish subtle primary versus secondary adrenal insufficiency because, in the former, all adrenal zones are typically affected while in the latter, the renin-angiotensin system is intact and aldosterone responses are usually preserved [15].
The advantage of the high-dose test is that the cosyntropin can be injected IM as well as IV because pharmacologic plasma ACTH concentrations are still achieved [16]. The 1 mcg low-dose test described below has not been validated after IM injection and may not provide valid results by this route [17,18].
Low-dose ACTH stimulation test — A test involving more physiologic plasma concentrations of ACTH theoretically provides a more sensitive index of adrenocortical responsiveness. The low-dose ACTH test is performed by measuring serum cortisol immediately before and at both 20 and 30 minutes [13] after IV injection of cosyntropin (synthetic ACTH[1-24]) in a dose either of 1 mcg (160 milli-international units) [19] or 0.5 mcg (80 milli-international units) per 1.73 m2 [20].
The 0.5 mcg per 1.73 m2 dose has the advantage of compensating for plasma volume, but it is approximately one-half the 1 mcg dose in most patients and may not result in the same peak serum cortisol concentrations at 30 minutes because the peak tends to be reached earlier and is 30 percent lower than after insulin-induced hypoglycemia [14,20,21].
Additionally, criteria for interpretation are not available for patients whose body surface area is significantly larger or smaller than 1.73 m2.
There are no commercially available preparations of "low-dose" cosyntropin. Accurate preparation of the dose is essential because 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 [11,22,23].
The vials of cosyntropin typically used contain 250 mcg and come with sterile normal saline solution to be used as diluent. Instructions are as follows:
●Inject 1 mL of the diluent into the vial of cosyntropin to produce a 250 mcg/mL solution and shake thoroughly.
●Using a 1 mL tuberculin syringe, withdraw 0.2 mL (ie, 50 mcg cosyntropin) and inject it into a vial containing 24.8 mL of sterile normal saline solution to produce a 2 mcg/mL solution.
●After shaking thoroughly, again using a 1 mL syringe, withdraw 0.5 mL (1 mcg cosyntropin) or the appropriate volume for the patient's surface area, and inject the entire volume immediately intravenously.
An alternate dilution method that may be more accurate is to inject 250 mcg of cosyntropin (the entire contents of the vial) into a 250 mL bag of isotonic saline, mix well, and then inject 1 mL (1 mcg) of this diluted cosyntropin into the patient. This method requires fewer dilution steps and results in less error as a result.
We suggest that the low-dose test not be routinely used. Although early small studies suggested that the low-dose test was more sensitive for diagnosing partial adrenal insufficiency [7,24-26], two systematic reviews and meta-analyses reported that the high- and low-dose tests have similar diagnostic accuracy for identifying secondary adrenal insufficiency.
Interpretation of results
Variability in cortisol assays: Impact on cutoff criteria — Like all hormone assays, there is variability between assay methods and laboratories. Commonly used immunoassays typically give higher serum cortisol values than liquid chromatography/tandem mass spectrometry (LC-MS/MS) and the new platform methods (validated against LC-MS/MS), so clinicians should be aware which cutoff is appropriate for their laboratory [27]. For most assays, however, a peak serum cortisol concentration ≥18 mcg/dL (500 nmol/L) indicates normal adrenal function (table 1). The peak cortisol value typically occurs 60 minutes after administration but may occur at 30 minutes.
●Established immunoassays – The criteria used to indicate normal adrenal function are a minimum serum cortisol concentration ≥18 to 20 mcg/dL (500 to 550 nmol/L) before or after corticotropin (ACTH) injection. (See 'Absolute cutoffs' below.)
●New platform immunoassay – A new version of a platform cortisol immunoassay (Roche Diagnostics Elecsys Cortisol II) is now commonly used that has a significantly lower basal and ACTH-stimulated serum cortisol cutoff [28-30]. (See 'Absolute cutoffs' below.)
●LC-MS/MS – Some reference laboratories are now using LC-MS/MS for which the serum cortisol cutoff is similar to the new platform immunoassay described above and lower than the routinely used cortisol immunoassays [31]. (See 'Absolute cutoffs' below.)
Criteria for normal response to ACTH
Absolute cutoffs — The criterion for serum cortisol on the standard high-dose ACTH stimulation test is a minimum concentration of 18 mcg/dL (500 nmol/L) before or after IV ACTH injection (table 1) [32,33]. A lower cutoff is used (≥16 mcg/dL [440 nmol/L]) if ACTH is administered IM rather than IV [16].
Salivary cortisol can also be measured during this test. A conservative cutoff for salivary cortisol at one hour after injection is 0.9 mcg/dL (26 nmol/L) [34]. (See "Measurement of cortisol in serum and saliva".)
On the low-dose ACTH stimulation test, the criterion for serum cortisol is a minimum value ≥18 mcg/dL (500 nmol/L) before or after ACTH injection [19,20]. As noted, there are several caveats with this use of this test, and therefore, great caution must be used if a subnormal response is encountered. (See 'Low-dose ACTH stimulation test' above.)
Incremental increase in cortisol — Incremental or percent increases in cortisol should not be used to assess the response to ACTH [35], 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 normal subjects with high-normal basal serum cortisol concentration have little or no rise after ACTH. It is therefore important to obtain both baseline and post-ACTH cortisol samples for accurate assessment of adrenal function.
Effect of increased or decreased corticosteroid-binding globulin values — In women, the response to ACTH may be affected by the use of oral contraceptives (OCs), which increase cortisol-binding globulin (CBG) levels, but data are limited. In one study evaluating the normal response to ACTH in 100 healthy adults (50 men and 50 women) and 13 women taking OCs, 30-minute total cortisol values were two-fold higher and free cortisol values were lower in the OC group compared with controls [36]. Typically, clinicians are aware of this effect and can interpret the responses to ACTH appropriately. Furthermore, the use of salivary cortisol response can be useful as their measurement is a surrogate for serum free cortisol and are not affected significantly by OCs [37]. A frankly subnormal response in the presence of OCs is useful to make the diagnosis.
Similarly, patients with nephrotic syndrome who have low CBG values may have an apparently "subnormal" cortisol response to ACTH in the absence of adrenal insufficiency [36].
The impact of high CBG levels on cortisol during pregnancy are reviewed separately. (See "Diagnosis of adrenal insufficiency in adults" and "Diagnosis and management of Cushing's syndrome during pregnancy".)
Primary versus secondary adrenal insufficiency — The response to ACTH varies with the underlying disorder.
●In a patient with clinically significant hypopituitarism and secondary adrenal insufficiency, the resultant adrenal atrophy will lead to a subnormal cortisol response. However, there may still be a detectable response, particularly if the hypopituitarism is partial and/or very recent.
●If, on the other hand, the patient has well established primary adrenal insufficiency, endogenous ACTH secretion is already elevated and there should be little or no adrenal response to exogenous ACTH.
Thus, a clearly subnormal response to the high-dose ACTH stimulation test is diagnostic of primary or secondary adrenal insufficiency, whereas a normal response excludes both disorders [33]. Cortisol values between 18 to 25.4 mcg/dL (500 to 700 nmol/L) in established assays and 12.7 to 14.9 microgram/dL (351 to 412 nmol/L) in newer generation assays represent a range of uncertainty for the diagnosis of secondary adrenal insufficiency. Higher concentrations represent a normal response in the non-intensive care unit setting. (See "Glucocorticoid therapy in septic shock in adults", section on 'Critical illness-related corticosteroid insufficiency'.)
An inadequate serum cortisol response to ACTH stimulation establishes the diagnosis of adrenal insufficiency but does not distinguish between the primary and secondary forms. (See "Causes of secondary and tertiary adrenal insufficiency in adults".)
There are two ways to distinguish among them:
●Measurement of plasma ACTH in the basal sample – If it is higher than normal, the patient has primary adrenal insufficiency, whereas if it is low, the diagnosis is secondary or tertiary adrenal insufficiency. A normal value may represent evolving primary adrenal insufficiency or a maximal pituitary response of secondary adrenal insufficiency. ACTH must be measured before injection of cosyntropin because the latter reduces the measured levels of intact ACTH in modern two-site ACTH immunometric assays [38].
●With the availability of accurate measurements of basal plasma ACTH and the availability of corticotropin-releasing hormone (CRH) for stimulation testing, prolonged ACTH infusion tests have become unnecessary for determining the cause of adrenal insufficiency. Of note, CRH is available in many, but not all, countries. (See "Corticotropin-releasing hormone stimulation test".)
Special situations
Stress and surgery — An area in which there continues to be controversy is whether the high-dose ACTH test accurately predicts the ability of a patient to respond adequately to stress, such as major surgery. Rare patients who respond normally to the high-dose ACTH test have a subnormal serum cortisol response to surgery [35]; however, these patients tolerate surgery normally [39]. The reason patients do well during major surgery, even though their serum cortisol concentrations may not rise normally in response to ACTH or the surgery, may be that more than basal daily cortisol secretion is not needed to survive surgery [40]. This issue is reviewed in detail separately. (See "The management of the surgical patient taking glucocorticoids".)
Recent ACTH deficiency — Outside of the critical care setting, with one exception, any patient who responds normally to the high-dose ACTH test does not require glucocorticoid supplementation for stress, surgical or otherwise. The exception is the patient who is acutely ACTH deficient, as might occur soon after pituitary surgery or brain or pituitary stalk trauma [41].
During the several days after cessation of ACTH secretion, the zonae fasciculata and reticularis of the adrenals undergo functional and anatomical atrophy. However, during part of that interval, the adrenals may respond normally to pharmacologic doses of ACTH, whereas the pituitary gland is unable to release ACTH in response to stress. Therefore, these patients will respond normally to the high-dose ACTH test, but fail to respond to insulin-induced hypoglycemia (since they cannot increase ACTH secretion). Such patients, who can usually be identified by their history, require glucocorticoid supplementation during surgical or other stresses, and may subsequently fail to respond to the ACTH stimulation test and require daily replacement glucocorticoid.
In patients who have undergone pituitary surgery, the cortisol response to standard-dose ACTH testing correlates well with the results of an insulin-induced hypoglycemia test if performed at least four weeks after surgery. In a study of 41 patients who were evaluated using high-dose ACTH and insulin-induced hypoglycemia testing at one and four to six weeks after pituitary surgery (Cushing's disease excluded), all patients with a 30-minute serum cortisol after ACTH stimulation of less than 12.7 mcg/dL (350 nmol/L) four to six weeks after surgery had a subnormal response to hypoglycemia, while those with a serum cortisol greater than 23.6 mcg/dL (650 nmol/L) had a normal response to hypoglycemia [42]. Patients with intermediate values required additional testing to determine whether they had adrenal insufficiency.
Some [7,19,21,24,43-46], but not all [15,23,47,48], studies have reported that patients with mild or recent secondary or tertiary adrenal insufficiency are more likely to have a subnormal cortisol response on a low-dose test than the high-dose ACTH test. A systematic review and meta-analysis of the literature did not find a difference in diagnostic accuracy between the low- and high-dose tests [22]. Neither test adequately ruled out secondary adrenal insufficiency.
●The issue is not whether patients with mild or recent secondary or tertiary adrenal insufficiency will have normal ACTH and cortisol responses during stress [49]; the issue is whether patients who respond normally to the low-dose ACTH test (or any other test) are at any greater risk for adverse effects of adrenal insufficiency during stress than a patient with perfectly normal hypothalamic-pituitary-adrenal (HPA) function.
●Until this issue is resolved, patients with possible acute ACTH deficiency, even if they have a normal response to ACTH, should be given glucocorticoid and saline during major stress or surgery (see "Treatment of adrenal insufficiency in adults"). In addition, repeat testing should be performed four to six weeks after the injury or surgical insult.
Critical illness — The HPA axis responses during critical illness have been of great interest. One would assume that critical illness (eg, due to sepsis) would be a stressful situation and that ACTH secretion would be increased leading to large increases in cortisol. However, critically ill patients have reduced cortisol metabolic clearance leading to suppression of ACTH [50]. The term relative adrenal insufficiency (ie, suboptimal cortisol production for total body demands) has been coined to take into account the high prevalence of HPA dysfunction in those who are critically ill. This condition has also been termed "critical illness-related corticosteroid insufficiency (CIRCI)." This topic is reviewed in detail separately (see "Glucocorticoid therapy in septic shock in adults", section on 'Critical illness-related corticosteroid insufficiency'). Guidelines suggest that glucocorticoid administration should be reserved for very specific patients with sepsis and septic shock.
Congenital adrenal hyperplasia — In the evaluation of patients with hirsutism, the possibility of nonclassic congenital adrenal hyperplasia (NCCAH) due to 21-hydroxylase deficiency must be considered. The use of the ACTH stimulation test to make this diagnosis is reviewed in detail separately. (See "Diagnosis and treatment of nonclassic (late-onset) congenital adrenal hyperplasia due to 21-hydroxylase deficiency", section on '17-hydroxyprogesterone'.)
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
●In general, the high-dose corticotropin (ACTH) stimulation test has supplanted insulin-induced hypoglycemia and metyrapone testing to evaluate hypothalamic-pituitary adrenal (HPA) function. ACTH tests can be performed without monitoring by an advanced health care provider and are less expensive. (See 'ACTH stimulation tests' above.)
●Dynamic testing should be performed to establish the diagnosis in patients with equivocal serum cortisol values in whom hypoadrenalism is suspected. With simultaneous measurement of the baseline plasma ACTH, it is possible to distinguish between primary and secondary or tertiary adrenal insufficiency. (See 'Role of dynamic testing' above.)
●ACTH(1-24) (cosyntropin), which has the full biologic potency of native ACTH (1-39), is used to perform these tests. (See 'ACTH stimulation tests' above.)
●A standard high-dose test consists of measuring serum cortisol before an intravenous (IV) or intramuscular (IM) injection of 250 mcg (85 nmol, or 40 international units) of cosyntropin and again at 30 minutes post-injection, or at both 30 and 60 minutes post-injection. (See 'Standard high-dose ACTH stimulation test' above.)
●Although it showed promise at first, the low-dose ACTH stimulation test may have an unacceptable false-positive rate and should be used with caution. It is performed by measuring serum cortisol before and at both 20 and 30 minutes after IV injection of 1 mcg cosyntropin. There are potential errors in preparation of the ACTH for injection. (See 'Low-dose ACTH stimulation test' above.)
●The criteria used to indicate normal adrenal function are a minimum serum cortisol concentration ≥18 mcg/dL (500 nmol/L) before or after cosyntropin injection. New generations of platform cortisol immunoassays and liquid chromatography/tandem mass spectrometry (LC-MS/MS) may give lower results, so clinicians must know the appropriate reference ranges for the method used in their laboratory. For most assays, however, a peak serum cortisol concentration ≥18 mcg/dL (500 nmol/L) indicates normal adrenal function (table 1). The peak cortisol value typically occurs 60 minutes after administration but may occur at 30 minutes. (See 'Interpretation of results' above.)
●A lower cutoff is used (≥16 mcg/dL [440 nmol/L]) if high-dose cosyntropin is administered IM. (See 'Standard high-dose ACTH stimulation test' above.)
●High cortisol-binding globulin (CBG) levels in women using oral contraceptives (OCs), or low CBG levels in patients with nephrotic syndrome may confound interpretation of ACTH test. The use of salivary cortisol or serum free cortisol can be helpful. A frankly low response to ACTH in the presence of OCs is helpful to establish the diagnosis of adrenal insufficiency. (See 'Effect of increased or decreased corticosteroid-binding globulin values' above.)
●ACTH stimulation tests are not valid to evaluate HPA function if there has been a recent pituitary injury. Metyrapone testing or insulin-induced hypoglycemia should be used in these circumstances. (See 'Recent ACTH deficiency' above.)
●Simultaneous determination of basal ACTH before injecting cosyntropin allows one to determine if adrenal insufficiency is a primary adrenal disease (elevated basal ACTH values) or secondary to hypothalamic-pituitary deficiency (low basal ACTH value). (See 'Primary versus secondary adrenal insufficiency' above.)
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.
