http://www.postgradmed.com/issues/1998/07_98/hasinski.htm
Assessment of adrenal glucocorticoid function
Which tests are appropriate for screening?
Stefan Hasinski, MD
VOL 104 / NO 7 / JULY 1998 / POSTGRADUATE MEDICINE
This is the second of three articles on on endocrine disorders
Preview: Overproduction or underproduction of adrenal hormones raises sticky diagnostic problems for primary care physicians. Fortunately, assessment of hypoadrenalism has been greatly simplified. On the other hand, evaluation of patients with suspected hyperadrenalism (Cushing's syndrome) can be difficult, confusing, and frustrating. Dr Hasinski reviews the options for testing these critical adrenal functions and provides up-to-date information on interpreting test outcomes and pursuing a diagnosis.
--------------------------------------------------------------------------------
The adrenal glands, the two small pyramidal structures on the superior pole of the kidneys that weigh just 4 g each, can raise diagnostic havoc when their functions go awry. A quick review of these glands seems appropriate in explaining how to assess whether or not they are responsible for a number of symptoms.
Adrenal function
About 90% of the adrenal gland tissue consists of adrenal cortex, which has three distinct zones: the glomerulosa, or outer layer, which produces aldosterone; the fasciculata, or middle layer, which makes glucocorticoid precursors and cortisol; and the reticularis, or inner layer, which synthesizes adrenal androgens (1,2).
Cortisol synthesis originates with cholesterol, about 80% of which is delivered to the adrenal glands by low-density lipoprotein (LDL) cholesterol. The number of LDL receptors is increased when the adrenal glands are stimulated by corticotropin (ACTH). The remaining cholesterol is made through hydroxymethylglutaryl coenzyme A reductase activity, which is not controlled by corticotropin. Cholesterol is modified by a series of enzymes of the P-450 system to produce cortisol.
The production of cortisol is regulated tightly by the hypothalamus and the pituitary gland, with classic feedback inhibition. Corticotropin-releasing hormone is released from the parvicellular division of the hypothalamic paraventricular nucleus. Binding of corticotropin-releasing hormone induces production of pro-opiomelanocortin, which is then cleaved into various fragments, including melanocyte-stimulating hormones, beta-lipotropins, beta-endorphins, and corticotropin.
Corticotropin, a 39-amino-acid hormone, is released from the anterior pituitary and enters the systemic circulation. Once bound to the adrenal cortex, corticotropin induces cortisol synthesis and secretion. Cortisol binds to receptors in the hypothalamus and pituitary to inhibit the release of corticotropin-releasing hormone and corticotropin, respectively.
Cortisol secreted by the adrenal cortex is transported mainly bound to plasma proteins, specifically corticosteroid-binding globulin (transcortin), but also albumin and sex hormone-binding globulin (1,2). Between 90% and 97% of cortisol is bound to proteins (1). It is only the "free fraction," or nonprotein-bound hormone, that is available to bind to specific tissue receptors. Cortisol metabolism occurs mainly in the liver, and products of cortisol metabolism can be detected in the urine as 17-hydroxycorticosteroids (17-OHCS) (2).
Adrenal insufficiency
The diagnosis of adrenal insufficiency is relatively straightforward. The chief difficulties are recognizing the constellation of symptoms and maintaining a high index of suspicion.
Adrenal insufficiency can be primary, secondary, or tertiary, depending on the location of the lesion. Primary adrenal insufficiency, or Addison's disease, is due to adrenal gland failure. In the past, the most common cause was tuberculosis (2), but now it is autoimmune adrenalitis, which may be isolated or part of a generalized autoimmune disorder. Other causes include metastatic disease (most commonly from lung and breast carcinomas), hemorrhage, infection, rare familial disorders (eg, adrenoleukodystrophy, adrenomyeloneuropathy), and HIV-related disease (1-3).
Symptoms of primary adrenal insufficiency include weakness, abdominal pain, nausea, weight loss, hypotension or shock, lack of libido, loss of body hair in women, hyperpigmentation, and psychiatric changes. Laboratory findings may include hyponatremia, hyperkalemia, mild acidosis, hypoglycemia, and hypercalcemia.
In secondary adrenal insufficiency, release of corticotropin from the pituitary is impaired. In tertiary adrenal insufficiency, release of corticotropin-releasing hormone from the hypothalamus is insufficient. Symptoms of secondary and tertiary adrenal insufficiency may be identical to those of primary disease, except that hyperpigmentation does not occur in the secondary or tertiary form because of insufficient production of corticotropin and other products of pro-opiomelanocortin metabolism (ie, melanocyte-stimulating hormones). Aldosterone, the potent adrenal mineralocorticoid hormone, is controlled mainly by potassium and the renin-angiotensin system and is therefore only minimally affected by a lack of corticotropin.
Screening tests
The best screening test for evaluating adrenal glucocorticoid function involves rapid stimulation with cosyntropin (Cortrosyn), a synthetic analogue of corticotropin. After baseline plasma cortisol and aldosterone levels are measured, 250 micrograms of cosyntropin is given as an intravenous bolus or an intramuscular injection. Plasma cortisol and aldosterone levels are measured again 30 to 60 minutes later. The blood samples for aldosterone can be held until the results of the cortisol response are known. If the cortisol response is insufficient, the aldosterone levels can help localize the deficiency.
A baseline cortisol level of greater than 18 micrograms/dL is consistent with normal adrenal function (2). However, opinions differ as to how to interpret test results (1-3). I prefer to use a cortisol level that increases twofold over baseline and rises above 20 micrograms/dL as an indication of normal adrenal function. This standard considers both baseline and reserve adrenal function. The baseline corticotropin level can help differentiate primary from secondary or tertiary hypoadrenalism but must be interpreted cautiously because corticotropin release from the pituitary varies throughout the day. A newer and possibly more sensitive test uses 1 microgram of cosyntropin to evaluate the cortisol response, which should be the same as with the cosyntropin 250 micrograms test (4). The aldosterone response in a cosyntropin test is blunted or absent in patients with primary adrenal insufficiency. In secondary or tertiary adrenal insufficiency, aldosterone response is normal (an increase of two times baseline) because the renin-angiotensin axis is not affected by decreased corticotropin.
A corticotropin-releasing hormone stimulation test can be used to differentiate secondary (pituitary) from tertiary (hypothalamic) disease, although this test is seldom necessary. In the past, the central axis was assessed with a metyrapone (Metopirone) test. Metyrapone inhibits cortisol synthesis, which results in release of corticotropin-releasing hormone and corticotropin. However, in patients with suspected adrenal insufficiency, this test can be dangerous because it may precipitate an addisonian crisis. The rapid cosyntropin stimulation test has generally replaced the metyrapone test.
Another test for differentiating primary from secondary or tertiary adrenal insufficiency is a prolonged cosyntropin-stimulation (Rose) test. After baseline plasma cortisol and 24-hour urinary 17-OHCS levels are established, 250 micrograms of cosyntropin is infused continuously over 48 hours. Plasma cortisol and 24-hour urinary 17-OHCS are then remeasured on the second day of the infusion, and a final plasma cortisol level is determined as the infusion is completed (5). In primary adrenal insufficiency, no change is seen in cortisol or 17-OHCS concentrations. In secondary or tertiary adrenal insufficiency, an incremental increase occurs over the course of the infusion. This implies that the cortex has undergone atrophy because of insufficient corticotropin stimulation. However, with longer stimulation, the cortex is capable of functioning.
Once a decision has been made as to whether adrenal insufficiency is primary, secondary, or tertiary, the appropriate imaging study should be used to rule out other treatable causes. Computed tomography (CT) scanning is preferred for study of the adrenal glands for primary disease, and magnetic resonance imaging (MRI) is best for studies of the hypothalamus and pituitary glands for secondary and tertiary disease.
Hypercortisolism
Cushing's syndrome is defined as any chronic increase in glucocorticoid activity. By far the most common cause is prolonged use of glucocorticoid agents for treatment of chronic inflammatory diseases (eg, rheumatoid arthritis) or after organ transplantation. Other possible causes include pituitary adenomas (Cushing's disease), adrenal adenomas or carcinomas, and ectopic production of corticotropin-releasing hormone or corticotropin. Use of corticosteroids or chronic use of inhaled corticosteroids or corticosteroid creams must also be considered.
As with adrenal insufficiency, Cushing's syndrome may be elusive, and a high level of suspicion is needed. Symptoms can include worsening obesity, new-onset hypertension, skin changes (eg, easy bruising, striae), poor wound healing, facial plethora, hirsutism, acne, muscle weakness and wasting, peripheral edema, and neuropsychiatric changes (eg, depression, mania).
Cushing's syndrome can be classified as corticotropin-dependent or corticotropin-independent (table 1). Corticotropin-dependent Cushing's syndrome is further classified as caused by pituitary adenoma (Cushing's disease) or by ectopic production of corticotropin or corticotropin-releasing hormone. Corticotropin may be produced ectopically by malignancy (ie, lung carcinoma, renal-cell carcinoma), as is the case in about 80% of patients, or by other tumors, such as carcinoid tumors in the lung, pancreas, or thymus (1,6).
Table 1. Causes of Cushing's syndrome
Corticotropin-dependent
Pituitary adenoma (Cushing's disease)
Ectopic corticotropin production from carcinoma (eg, lung, breast, renal-cell)
Ectopic corticotropin-releasing hormone production from carcinoid tumors (eg, chest, thymus)
--------------------------------------------------------------------------------
Corticotropin-independent
Adrenal adenoma
Adrenal carcinoma
Micronodular dysplasia
Macronodular hyperplasia
Iatrogenic causes
--------------------------------------------------------------------------------
Corticotropin-independent Cushing's syndrome is caused by autonomous production of cortisol. Possible sources include adrenal adenomas, adrenal carcinomas, primary pigmented nodular adrenal dysplasia (micronodular dysplasia), macronodular hyperplasia, and other, relatively rare, disorders.
Screening tests
The first step in evaluating Cushing's syndrome is to determine whether hypercortisolism is present. The most convenient screening procedure is the overnight dexamethasone suppression test, in which 1 mg of dexamethasone is given orally at bedtime (usually between 10 and 11 pm). A fasting blood sample is drawn when the patient arises the next morning (preferably by 8 am), and cortisol is measured. The fasting plasma cortisol level should be less than 5 micrograms/dL, although some authorities believe that a level of less than 3 micrograms/dL is more sensitive and specific (7). A level greater than 5 micrograms/dL warrants further testing.
A 24-hour urine collection for urinary free cortisol is also an excellent screening test. High-performance liquid chromatography enhances conventional protein-binding assays and also increases sensitivity and specificity (7). False-negative results may occur because of inadequate collection, daily fluctuations in cortisol levels, or abnormalities caused by other medications the patient may be using. Therefore, at least three samples should be obtained. If all three samples show normal cortisol levels, Cushing's syndrome can be ruled out. If any of the three values is abnormal, further testing is warranted (6,8).
If either the overnight dexamethasone suppression test or any of the 24-hour urine evaluations are abnormal, false-positive results and pseudo-Cushing's syndrome need to be considered. Traditionally, a low-dose (2 mg/day) dexamethasone test has been used. High-dose (8 mg/day) dexamethasone suppression is used to distinguish Cushing's disease (pituitary) from other causes of Cushing's syndrome (table 2).
Table 2. Low-dose followed by high-dose dexamethasone suppression test
Day 1
Obtain baseline plasma cortisol and corticotropin values
Begin baseline 24-hr urine collection for free cortisol and 17-OHCS
Day 2 (low-dose dexamethasone suppression)
Complete baseline 24-hr urine collection
Start dexamethasone, 0.5 mg orally every 6 hr
Day 3
Continue dexamethasone, 0.5 mg orally every 6 hr
Begin second 24-hr urine collection for free cortisol and 17-OHCS
Day 4 (high-dose dexamethasone suppression)
Measure plasma cortisol
Complete second 24-hr urine collection
Begin dexamethasone, 2 mg orally every 6 hr
Day 5
Continue with dexamethasone, 2 mg orally every 6 hr
Begin third 24-hr urine collection for free cortisol and 17-OHCS
Day 6
Complete third 24-hr urine collection
Measure plasma cortisol and corticotropin
--------------------------------------------------------------------------------
17-OHCS, 17-hydroxycorticosteroids.
--------------------------------------------------------------------------------
For the low-dose dexamethasone suppression test, baseline plasma cortisol and corticotropin levels are measured and a 24-hour urine collection is made on day 1 to establish free cortisol and 17-OHCS levels. Beginning on day 2, after the urine collection is completed, 0.5 mg of dexamethasone is given orally every 6 hours for 8 doses (2 days). On day 3, a second 24-hour urine sample is collected, and free cortisol and 17-OHCS values are measured again. The high-dose dexamethasone suppression test starts on day 4. After the second 24-hour urine collection is completed, the dexamethasone dose is increased to 2 mg orally every 6 hours for 8 doses (2 days). On day 5, dexamethasone is continued and a third 24-hour urine collection is begun. Finally, on day 6, plasma cortisol and corticotropin are measured again. The third 24-hour urine sample can be held, pending the results of the second 24-hour sample, and urine cortisol and 17-OHCS levels can be determined if results of the second 24-hour urine study were abnormal.
In patients who have normal cortisol metabolism and those with pseudo-Cushing's syndrome, the second 24-hour urine collection (low-dose dexamethasone suppression) shows urinary free cortisol levels of less than 4 mg per 24 hours, and the plasma cortisol level is less than 3 micrograms/dL. The high-dose dexamethasone test is then used to distinguish Cushing's disease from other causes of Cushing's syndrome. In Cushing's disease, the third 24-hour urine collection shows a 90% decrease in cortisol from baseline and a 64% decrease in 17-OHCS (8-10). Lesser degrees of suppression indicate nonpituitary-dependent Cushing's syndrome (eg, adrenal adenoma or carcinoma, ectopic corticotropin production).
Administration of low- and high-dose dexamethasone to suppress cortisol production is cumbersome and may be difficult to complete properly. A high-dose overnight dexamethasone suppression test is simpler and may prove equally effective. With this test, baseline plasma cortisol and corticotropin levels are obtained from the fasting patient (by 8 am), 8 mg of dexamethasone is given orally at bedtime, and cortisol levels are remeasured the next morning. In patients with Cushing's disease, the follow-up cortisol values usually decrease by 50% from the baseline (8). The high-dose overnight test has been favorably compared with the standard high-dose dexamethasone suppression test (9). Some tumors (ie, carcinoids) cause some degree of suppression on dexamethasone testing.
Another test using synthetic corticotropin-releasing hormone may be able to differentiate Cushing's disease from other causes of Cushing's syndrome. In Cushing's disease, there is a paradoxical increase in the level of corticotropin after administration of corticotropin-releasing hormone. However, considerable overlap is seen among patients with normal levels and those with Cushing's syndrome. Therefore, this test should not be used routinely. Corticotropin levels should be measured by immunoradiometric assay, which has greater specificity and sensitivity, although this assay cannot detect an unusual type of corticotropin (ie, "big" corticotropin), which may also have biologic activity (8).
Scanning techniques
Appropriate radiologic and nuclear medicine studies should be used as directed by the biochemical studies. MRI of the pituitary, with and without gadolinium, is superior to CT scanning. Nonetheless, between 40% and 50% of pituitary tumors are missed by MRI in patients with Cushing's disease (6,8). CT scanning is preferred for viewing the adrenal glands and chest. However, adrenal CT scans must be interpreted cautiously, since 2% to 15% of patients have nonfunctioning adenomas (incidentalomas) (6,11). Iodo-seleno-cholesterol scans are used to evaluate synthetic function within the adrenal gland. Octreotide scans are used to localize ectopic corticotropin-producing tumors (carcinoids), many of which have somatostatin receptors (6).
When Cushing's disease is confirmed, inferior petrosal sinus sampling may help localize the cause. The inferior petrosal sinuses are selectively and simultaneously catheterized, and baseline blood samples for corticotropin are simultaneously obtained from both sinuses as well as peripherally. Corticotropin-releasing hormone (100 micrograms or 1 microgram/kg of body weight) is injected, and blood samples for corticotropin are drawn from the sinuses and peripherally at 2, 3, 5, and 10 minutes. Ratios are established between inferior petrosal sinus levels and peripheral corticotropin levels. A ratio greater than 2.0 is consistent with Cushing's disease. An interpetrosal gradient (eg, right versus left) greater than 1.2 after corticotropin-releasing hormone injection predicts the location of a lesion in 70% to 80% of patients (6,8,9). However, the procedure may be complicated by cavernous sinus thrombosis, infection, hemorrhage, and brainstem ischemia (8,9,12). Because of its complexity and risk for complications, inferior petrosal sinus sampling should be performed only in centers with considerable expertise.
Summary
The rapid cosyntropin stimulation test offers a simple means for detecting adrenal insufficiency. In contrast, assessment of suspected hypercortisolism (Cushing's syndrome) is difficult because cortisol levels fluctuate with intermittent release of corticotropin from the pituitary or from tumors. Also, a number of medications affect cortisol levels, leading to false-positive or false-negative results. The classic low-dose followed by high-dose dexamethasone test is cumbersome, and other, simpler studies, such as the overnight high-dose dexamethasone suppression test, may prove more practical and cost-effective. With both high and low levels of adrenal glucocorticoids, awareness and early recognition of the symptoms are important. An endocrinologist should be consulted when the overnight dexamethasone suppression test or the 24-hour urine cortisol collection is abnormal or if clinical suspicion is high despite normal results on screening tests.
References
Orth DN, Kovacs WJ, DeBold CR. The adrenal cortex. In: Wilson JD, Foster DW, et al, eds. Williams textbook of endocrinology. 8th ed. Philadelphia: Saunders, 1992:489-619
Federman DD. The adrenal. In: Dale DC, Federman DD. Scientific American medicine. New York: Scientific American, 1995:1-18
Davenport J, Kellerman C, Reiss D, et al. Addison's disease. Am Fam Physician 1991;43(4):1338-42
Dickstein G, Shechner C, Nicholson WE, et al. Adrenocorticotropin stimulation test: effects of basal cortisol level, time of day, and suggested new sensitive low dose test. J Clin Endocrinol Metab 1991;72(4):773-8
Rose LI, Williams GH, Jagger PI, et al. The 48-hour adrenocorticotrophin infusion test for adrenocortical insufficiency. Ann Intern Med 1970;73(1):49-54
Hough FS. Diagnosis of Cushing's syndrome. S Afr Med J 1996;86(8 Suppl):1022-5
Lin CL, Wu TJ, Machacek DA, et al. Urinary free cortisol and cortisone determined by high performance liquid chromatography in the diagnosis of Cushing's syndrome. J Clin Endocrinol Metab 1997;82(1):151-5
Findling JW, Doppman JL. Biochemical and radiologic diagnosis of Cushing's syndrome. Endocrinol Metab Clin North Am 1994;23(3):511-37
Kaye TB, Crapo L. The Cushing syndrome: an update on diagnostic tests. Ann Intern Med 1990;112(6):434-44
Flack MR, Oldfield EH, Cutler GB Jr, et al. Urine free cortisol in the high-dose dexamethasone suppression test for the differential diagnosis of the Cushing syndrome. Ann Intern Med 1992;116(3):211-7
Ross NS, Aron DC. Hormonal evaluation of the patient with an incidentally discovered adrenal mass. N Engl J Med 1990;323(20):1401-5
Oliverio PJ, Monsein LH, Wand GS, et al. Bilateral simultaneous cavernous sinus sampling using corticotropin-releasing hormone in the evaluation of Cushing disease. Am J Neuroradiol 1996;17(9):1669-74
--------------------------------------------------------------------------------
Dr Hasinski is assistant professor of medicine, division of endocrinology and metabolism, Allegheny University of the Health Sciences, Hahnemann Division, Philadelphia. Correspondence: Stefan Hasinski, MD, Division of Endocrinology and Metabolism, Allegheny University/Hahnemann Division, 230 N Broad St, Mail Stop 426, Philadelphia, PA 19102.