Addison’s disease is primary adrenocortical insufficiency from bilateral adrenal cortex destruction. Tuberculosis used to be the most frequent etiology but now is second to autoimmune disease atrophy. Long-term steroid therapy causes adrenal cortex atrophy from disuse, and if steroids are abruptly withdrawn, symptoms of adrenal failure may develop rapidly. This is now the most common cause of addisonian-type crisis. Less common etiologies of Addison’s disease are infection, idiopathic hemorrhage, and replacement by metastatic carcinoma. The most frequent metastatic tumor is from the lung, and it is interesting that there often can be nearly complete replacement without any symptoms.

The salt-wasting forms of congenital adrenal hyperplasia—due to congenital deficiency of certain enzymes necessary for adrenal cortex hormone synthesis—might also be included as a variant of Addison’s disease.

Weakness and fatigability are early manifestations of Addison’s disease, often preceded by infection or stress. Other signs and symptoms of the classic syndrome are hypotension of varying degree, weight loss, a small heart, and sometimes skin pigmentation. Anorexia, nausea, and vomiting occur frequently in adrenal crisis. The most common symptoms of adrenal crisis are hypotension and nausea.

General laboratory tests

Serum sodium is decreased in 50%-88% of patients with primary Addison’s disease, and serum potassium is mildly elevated in 50%-64% of cases (due to concurrent aldosterone deficiency). One investigator reported hypercalcemia in 6% of patients. There occasionally may be a mild hypoglycemia, although hypoglycemia is more common in secondary adrenal insufficiency. Serum thyroxine is low normal or mildly decreased and TSH is upper normal or mildly increased. Plasma aldosterone is usually decreased and plasma renin is elevated. There often is a normochromic-normocytic mild anemia and relative lymphocytosis with a decreased neutrophil count. Total eosinophil count is usually (although not always) close to normal.

In primary adrenal insufficiency, a morning serum cortisol value is typically decreased and the plasma ACTH value is increased. Arginine vasopressin (AVP or ADH) is usually elevated.

Diagnostic tests in Addison’s disease

Screening tests. A single random serum cortisol specimen has been used as a screening procedure, since theoretically the value should be very low in Addison’s disease and normal in other conditions. Unfortunately, there are sufficient interfering factors so that its usefulness is very limited. Because serum cortisol normally has a circadian rhythm with its peak about 6-8 A.M., the specimen must be drawn about 6-8 A.M. in order not to misinterpret a lower value drawn later in the day. Stress increases plasma cortisol levels, although the increase is proportionately much less in Addison’s disease than in normal persons. The classic patient with Addison’s disease in crisis has an early morning cortisol level of less than 5 µg 100 ml (138 nmol/L), and a level of 5-10 µg/100 ml (138-276 nmol/L) is suspicious for Addison’s disease, especially if the patient is under stress. Patients with milder degrees of illness or borderline deficiency of cortisol may have a morning cortisol value of more than 10 µg/100 ml. It is often difficult to determine whether mild elevation of more than 10 µg/100 ml is due to stress or is a normal level. An early morning cortisol level of more than 20 µg/100 ml (550 mmol/L) is substantial evidence against Addison’s disease. Many endocrinologists do not consider a single random cortisol level to be reliable in screening for Addison’s disease. In spite of this it is usually worthwhile to obtain a serum specimen early for cortisol assay for diagnostic purposes (if it excludes Addison’s disease) or as a baseline (if it does not). A plasma sample should be obtained at the same time (EDTH anticoagulant) and frozen in case ACTH assay is needed later. As noted previously, serum sodium (and also chloride) is often low in classic cases, and if so would be suggestive of Addison’s disease if it were associated with a normal or elevated urine sodium and chloride level. However, as noted previously, serum sodium can be within population reference range in 12%-50% of patients.

Rapid ACTH screening (“Cortrosyn”). Most investigators now prefer a rapid ACTH stimulation test rather than the single cortisol assay, since the rapid test can serve as a screening test unless the patient is extremely ill and in some patients may provide the same information as a confirmatory test. After a baseline serum cortisol specimen is obtained, 25 units of ACTH or 0.25 mg of corsyntropin (Cortrosyn or Synacthen, a synthetic ACTH preparation) is administered. There is variation in technique among the descriptions in the literature. Most measure plasma cortisol response after administration of corsyntropin but a few assay urinary 17-OHCS. Some inject corsyntropin intramuscularly and others intravenously. Intravenous (IV) administration is preferred but not required under ordinary circumstances. If the patient is severely ill or is hypotensive, IV is recommended to avoid problems in corsyntropin absorption. Some obtain a serum cortisol specimen 30 minutes after giving corsyntropin, whereas others do so at 60 or 120 minutes. Some obtain samples at two intervals instead of one. The majority appear to obtain one sample at 60 minutes. Some also obtain a sample at 30 minutes; this helps confirm the 60-minute value and avoids technical problems. However, the 30-minute specimen is not considered to be as reliable as the 60-minute specimen, especially if intramuscular (IM) injection was used. Theoretically, patients with primary adrenal insufficiency should demonstrate little response, whereas patients with pituitary insufficiency or normal persons should have stimulated cortisol levels that exceed 20 µg/100 ml (550 mmol/L). Some endocrinologists require an increment of at least 7 µg above baseline in addition to a peak value of 20 µg or more, especially when baseline cortisol is over 10µg/100 ml (225 mmol/L). However, increments less than or greater than 7µg are not as reproducible (on repeat corsyntropin tests) as the 20-µg peak cutoff value. Some patients with pituitary insufficiency demonstrate normal response to corsyntropin and some have a subnormal response. Because corsyntropin test results are not uniform in patients with pituitary insufficiency, it has been suggested that aldosterone should be measured as well as cortisol. Aldosterone levels should increase in pituitary hypofunction but should not rise significantly in primary adrenal failure. The metyrapone test is also useful to diagnose pituitary insufficiency.

Some patients may have equivocal rapid test results, and others may have been treated with substantial doses of steroids for considerable periods of time before definitive tests for etiology of Addison’s disease are attempted. Under long-term steroid suppression, a normal adrenal cortex may be unable to respond immediately to stimulation. A definitive diagnosis of Addison’s disease is possible using prolonged ACTH stimulation. The classic procedure is the 8-hour infusion test. If biologic rather than synthetic ACTH is used, many recommend giving 0.5 mg of dexamethasone orally before starting the test to prevent allergic reactions. A 24-hour urine specimen is taken the day before the test. Twenty-five units of ACTH in 500 ml of saline is given intravenously during an 8-hour period while another 24-hour urine specimen is obtained. In normal persons, there will be at least a twofold to fourfold increase in cortisol or 17-OHCS levels. In Addison’s disease, there is practically no response. If pituitary deficiency is suspected, the test should be repeated the next day, in which case there will be a gradual, although relatively small, response. If exogenous steroids have been given over long periods, especially in large doses, the test period the classic approach is to repeat the 8-hour ACTH infusion procedure daily for 5-7 days. Patients with primary Addison’s disease should display little daily increment in cortisol values; those with secondary Addison’s disease should eventually produce a stepwise increase in cortisol values. Some have used a continuous ACTH infusion for 48 hours or depot IM synthetic ACTH preparations once daily instead of IV infusions or standard IM injections twice daily. If the patient has symptoms of adrenal insufficiency, both the rapid ACTH test and the prolonged ACTH test can be performed while the patient is taking 0.5-1.0 mg of dexamethasone per day, as long as therapy has not extended more than 5-6 days before starting the tests. Dexamethasone can be used because at these low doses it will not be a significant part of either serum or urine cortisol assays. Long periods of glucocorticoid therapy will interfere with the pituitary-adrenal feedback response necessary for rapid cortisol response to ACTH and will require longer periods of ACTH stimulation in the prolonged ACTH stimulation test.

Thorn Test. If steroid measurements are not available, the Thorn test could be substituted, although it is not as accurate. First, a count of total circulating eosinophils is done. Then the patient is given 25 units of ACTH, either intravenously in the same way as in the ACTH test just described or intramuscularly in the form of long-acting ACTH gel. Eight hours after ACTH administration is started, another total circulating eosinophil count is made. Normally, cortisone causes depression of eosinophil production. Therefore a normal cortisol response to ACTH stimulation would be a drop of total circulating eosinophils to less than 50% of baseline values. A drop of less than 50% is considered suspicious for adrenal insufficiency. False positive responses (less than a 50% drop) may occur in any condition that itself produces eosinophilia (e.g., acute episodes of allergy).

Adrenocorticotropic hormone (ACTH) assay. Plasma ACTH measurement has been used to help confirm the diagnosis of Addison’s disease and to differentiate primary from secondary adrenal failure. In primary adrenal failure, the ACTH value should be high and cortisol levels should be low. In hypothalamic or pituitary insufficiency, both ACTH and cortisol values theoretically should be low. Unfortunately, a considerable number of patients have cortisol or ACTH values within reference range. A specimen for plasma ACTH determination can be drawn at the same time as the specimen for baseline cortisol before stimulation tests and can be frozen for availability if needed.

Antiadrenal antibodies. In primary Addison’s disease, antiadrenal antibodies have been detected in 60%-70% of patients. This test would have to be performed in large reference laboratories or certain university medical centers. Currently, this test is not being used as a primary diagnostic procedure.