Approach Considerations
Clinical suspicion is important because the presentation of patients with adrenal insufficiency (Addison disease) may be insidious and subtle. The current tools for the diagnosis of adrenal insufficiency are likely inadequate, because they rely on measurement of total cortisol levels rather than free or unbound cortisol.
Subjects with critical illness, particularly premature infants, often have low serum albumin and transcortin concentrations, leading to low total serum cortisol concentration. This issue needs to be revisited when sound methods for measurement of free cortisol become available.
Provision of stress steroids in critically ill patients should be reserved for those who have a preexisting or concurrent reason for adrenal insufficiency (ie, history of adrenal insufficiency, previous chronic glucocorticoid exposure, etomidate exposure) or for those who have hypotension that is unresponsive to adequate fluid administration and catecholamines. [44, 45]
Wolman disease (OMIM 278000), an autosomal recessive disorder caused by a deficiency of lysosomal acid lipase, may present with adrenal calcifications that may be seen on plain radiography or computed tomography (CT) scanning of the adrenal glands.
Electrolyte levels
Hyponatremia with or without hyperkalemia is common in patients with primary adrenal insufficiency (Addison disease), and it is due to deficient aldosterone secretion. Hyponatremia is occasionally found in patients with central or secondary adrenal insufficiency. The presumed cause is water retention due to increased secretion of vasopressin. [46]
When hyponatremia or hyperkalemia is present, a simultaneous serum sample and spot urine or 24-hour urine measurement of sodium, potassium, and creatinine concentrations can be used to calculate the fractional excretion of sodium to determine whether inappropriate natriuresis is occurring (see Medscape Reference Laboratory Medicine articles Serum Sodium, Urine Sodium, Potassium, and Creatinine). A plasma renin activity (PRA)–to–aldosterone ratio of more than 30 is suggestive of inadequate mineralocorticoid production.
Serum Cortisol Testing
Interpret random serum cortisol concentrations in the context in which they were obtained. For example, adrenal insufficiency (Addison disease) is unlikely in an otherwise healthy individual whose 8:00 am serum cortisol concentration is more than 10 mcg/dL. By contrast, a serum cortisol concentration less than 18 mcg/dL in a sick and stressed patient is suggestive of adrenal insufficiency, although some critically ill patients may have such cortisol concentrations due to lack of protein binding to cortisol (see Relative adrenal insufficiency under Etiology).
Criteria for diagnosis
A diagnosis of adrenal insufficiency is confirmed if the serum cortisol level is less than 18 mcg/dL in the presence of a markedly elevated serum adrenocorticotropic hormone (ACTH) concentration and plasma renin activity. Based on normative data of children of various ages, adrenal insufficiency is likely if the serum cortisol concentration is less than 18 mcg/dL 30-60 minutes after intravenous (IV) administration of 250 mcg of cosyntropin (synthetic ACTH 1-24) in children over age 2 years. The cosyntropin test dose may be decreased to 15 mcg/kg for infants and 125 mcg for children under age 2 years. [2, 3, 47]
These criteria may not apply to premature or low-birth-weight infants, who have low cortisol secretion and, most likely, decreased cortisol binding to carrier proteins. [4] Therefore, the diagnosis of adrenal insufficiency in premature infants remains problematic.
If the serum cortisol level is low and the ACTH value is elevated, measurement of antiadrenal antibodies may be informative. Antibodies to one or more steroidogenic enzymes, particularly 21-hydroxylase, are often found in patients with autoimmune adrenal disease.
When a patient's serum cortisol response to cosyntropin is subnormal but his or her serum ACTH level is not elevated, the possibility of central adrenal insufficiency should be considered. Other indications of pituitary dysfunction, such as previous glucocorticoid exposure (suggesting a suppressed hypothalamic-pituitary-adrenal axis) or evidence of other pituitary dysfunction (suggesting hypopituitarism) are helpful. Because the adrenal gland may not yet be hypoplastic, a normal cortisol response to cosyntropin may be seen in patients who recently began suffering from ACTH or CRH deficiency. A smaller dose of cosyntropin (1 µg) may be more sensitive in this setting. [48, 49]
In central adrenal insufficiency, a 3-day stimulation with ACTH produces a normal cortisol response, indicating intact adrenal glands and implying that the initial low cortisol response to cosyntropin was related to chronic ACTH deficiency. ACTH gel (ACTHar Gel) is administered at 25 U/m2 every 12 hours for 3 days. Plasma cortisol levels should increase to more than 40 mcg/dL in response. This procedure is now seldom performed since plasma ACTH concentrations can be measured.
Salivary cortisol testing
A study by Chao et al indicated that salivary cortisol may be used to confirm or replace serum cortisol testing in the diagnosis of adrenal insufficiency in children. The investigators measured salivary cortisol via liquid chromatography–tandem mass spectrometry, employing a cutoff value of 500 ng/dL for salivary cortisol and 18 μg/dL for serum cortisol. They found that during high-dose adrenocorticotropic hormone (ACTH) stimulation testing, salivary and serum cortisol each had 100% specificity and sensitivity in the detection of adrenal insufficiency in pediatric patients. Recognize that there may be limitations to the approach in a pediatric setting, as compliance with the collection procedure in patients below age 8 years may be suboptimal. [50]
ACTH Testing
The standard ovine corticotropin-releasing hormone (CRH) stimulation test (1 mcg/kg over 1 min) may be helpful in the differential diagnosis of adrenal insufficiency. A lack of a 2-fold increase in serum adrenocorticotropic hormone (ACTH) concentration indicates pituitary dysfunction. A 2-fold or greater rise in ACTH without a concomitant rise in serum cortisol to more than 18-20 mcg/dL implies primary adrenal insufficiency. [51] Ovine CRH is difficult to obtain, and this test is mainly performed for research purposes.
A study by Iwanaga et al indicated that CRH stimulation tests can be effectively used in the diagnosis of relative adrenal insufficiency in preterm infants. Administering CRH stimulation tests to preterm infants with relative adrenal insufficiency and to those without it, the investigators found that neither base nor peak serum cortisol levels differed between the two groups. However, in the group with relative adrenal insufficiency, significant reductions were seen in delta cortisol levels and in the responsive ratio (peak-to-base ratio). [52]
The best dose of cosyntropin to administer for a cosyntropin stimulation test remains controversial, [53, 54, 55, 56] and this issue remains unresolved in the pediatric age group.
The standard dose is 250 mcg intravenously; some pediatric endocrinologists reduce the cosyntropin dose to 50-125 mcg for infants. Very low cosyntropin doses (1 mcg or 0.5 mcg/m2) have been used in the belief that the low-dose test is more sensitive for central adrenal insufficiency. [48] Meta-analysis suggested that the low-dose cosyntropin stimulation test may be superior, but the difference was small. [53] There actually may be a higher rate of false-positive test results with the low-dose ACTH stimulation test. [57]
Alternatives to ACTH Testing
Because the adrenal glands may not have had sufficient time to atrophy in the absence of adrenocorticotropic hormone (ACTH) stimulation, the relatively cumbersome and risky insulin-tolerance test or metyrapone stimulation test may be preferable to a cosyntropin challenge if the patient has recent-onset (ie, < 10 d) central adrenal insufficiency (Addison disease) (eg, a patient who recently underwent surgery of the hypothalamus or pituitary regions). The insulin-tolerance test is still considered the criterion standard.
Insulin-tolerance test
An insulin-tolerance test requires an intravenous administration of insulin (usually regular insulin 0.05-0.15 U/kg) to induce a 50% reduction in blood sugar concentration. Cortisol and glucose concentrations are measured every 15 minutes for 60 minutes. The test is considered adequate if the blood sugars level decreases by at least 50%. In response to the hypoglycemic stimulus, serum or plasma cortisol concentrations should rise to more than 18-20 mcg/dL.
The insulin-tolerance test poses some risk of hypoglycemic seizure. Therefore, closely monitor the patient and reverse the hypoglycemia if the patient becomes overtly symptomatic.
Metyrapone stimulation
Standard metyrapone stimulation tests involve administering metyrapone 300 mg/m2 in 6 divided doses over 24 hours. Because metyrapone inhibits 11-hydroxylase, which is involved in the last enzymatic step in cortisol synthesis, plasma levels of the cortisol precursor, 11-deoxycortisol, increase. A normal response is a rise in 11-deoxycortisol concentrations to more than 10.5 mcg/dL 4 hours after the last dose of metyrapone is given or a 2-fold to 3-fold increase in 24-hour urinary concentrations of 17-hydroxycorticosteroid (which include tetrahydro compound S, a urinary metabolite of 11-deoxycortisol) on the day of or the day after the administration of metyrapone.
This test is cumbersome and carries some risk of inducing an adrenal crisis.
Antiadrenal Antibody Testing
When primary adrenal insufficiency (Addison disease) is confirmed, antiadrenal antibodies, specifically anti-21-hydroxylase antibodies, can confirm an autoimmune cause for the disorder. If results for antiadrenal antibodies are negative, search for another etiology, such as tuberculosis (TB), adrenal hemorrhage, or adrenoleukodystrophy.
CT Scanning, MRI, and Nuclear Imaging
Computed tomography (CT) scanning is the imaging study of choice in the evaluation of adrenal insufficiency (Addison disease) and helps to identify adrenal hemorrhage, calcifications (see the following image), or infiltrative disease. Magnetic resonance imaging (MRI) is not as useful as CT scanning, and iodocholesterol scanning is also not particularly useful for adrenal insufficiency.
Computed tomography scan shows enlarged adrenal glands in a patient with early active autoimmune adrenalitis. Patients with chronic disease present with the opposite picture of hypotrophic adrenals.
Radiography and Ultrasonography
Abdominal radiography may reveal bilateral adrenal calcifications, which suggest a history of bilateral adrenal hemorrhage, tuberculosis (TB), or Wolman disease. Ultrasonography is a poor imaging modality for investigating the adrenal glands.
Histologic Features
Histologic findings in adrenal insufficiency (Addison disease) depend on the underlying cause. CT scan–guided fine-needle aspiration sometimes helps in diagnosing the etiology of infiltrative adrenal disease.
In cases of autoimmune adrenal failure, lymphocytic infiltration destroys the adrenal gland. Granulomatous changes in the adrenal glands indicate tuberculosis (TB)-related adrenal insufficiency (see the image below). Neoplastic infiltrations are caused by metastatic tumors. Hemorrhagic adrenal insufficiency results in hemorrhagic destruction of the adrenal glands. Fungal disease produces the typical picture of fungal infiltrates. Atrophy of the adrenals characterizes adrenocorticotropic hormone (ACTH) deficiency or resistance. Hyperplasia of the adrenals is characteristic of defects in steroidogenesis.
Left photomicrograph shows autoimmune adrenalitis. Right photomicrograph shows tuberculous adrenalitis. Note the caseous granuloma.
-
Regulation of the adrenal cortex. ACTH = adrenocorticotropic hormone; CRF = corticotropin-releasing factor; neg. = negative.
-
Left photograph shows hyperpigmentation on the dorsum of a patient's hand before the treatment of primary adrenal insufficiency. Right photograph shows normal pigmentation after treatment.
-
Left photograph shows a patient with Addison disease who has prominent pigmentation in areas not exposed to the sun, such as the palmar creases. Right photograph shows normal pigmentation after treatment.
-
Left photograph shows vitiligo in a patient with autoimmune adrenalitis. Right photograph shows an area of hyperpigmentation surrounding the vitiligo.
-
Left photomicrograph shows autoimmune adrenalitis. Right photomicrograph shows tuberculous adrenalitis. Note the caseous granuloma.
-
Computed tomography scan shows enlarged adrenal glands in a patient with early active autoimmune adrenalitis. Patients with chronic disease present with the opposite picture of hypotrophic adrenals.
