Published: August 2010
Each adrenal consists of two functionally distinct endocrine glands: the cortex, derived from mesenchymal cells, and the medulla, derived from neuroectodermal cells. The adrenal cortex consists of three concentric zones: The outer glomerulosa secretes the mineralocorticoid aldosterone, the intermediate fasciculata secretes cortisol, and the inner reticularis secretes androgens. The endocrine cells of the adrenal medulla are the chromaffin cells, which are part of the sympathetic nervous system and produce the catecholamine epinephrine.
Glucocorticoid insufficiency can be primary, resulting from the direct insult to the adrenal cortex, or secondary, from adrenocorticotropic hormone (ACTH) or corticotropin-releasing hormone (CRH) hyposecretion as a result of pituitary or hypothalamic dysfunctions.
Primary adrenal insufficiency affects glucocorticoid and mineralocorticoid secretion and may be fatal if untreated. Autoimmune destruction of the adrenal glands (Addison’s disease) is the most common cause of primary adrenal insufficiency in United States (70%-90%), whereas tuberculosis is now the second most common cause worldwide. Some other causes of primary adrenal insufficiency include bilateral adrenal hemorrhage, drugs (e.g., mitotane, etomidate, ketoconazole) and other infectious diseases (e.g., HIV infection, disseminated histoplasmosis, paracoccidiomycosis). HIV infection may result in both primary and secondary adrenal insufficiencies (AI). Addison’s disease may coexist with other autoimmune conditions, such as type 1 diabetes, hypothyroidism, or hypoparathyroidism.
Secondary and tertiary adrenal insufficiencies occur commonly after the discontinuation of glucocorticoids. Less frequently, ACTH deficiency may be caused by pituitary macroadenomas, pituitary surgery or radiation, and parasellar diseases. Megestrol acetate, an appetite stimulator used in some patients with advanced cancer or cachexia related to AIDS may be associated with secondary AI. Tertiary adrenal insufficiency results from the inadequate secretion of CRH. Secondary and tertiary adrenal insufficiencies only affect cortisol secretion, because ACTH has only a minor role in regulation of aldosterone secretion.
All patients with primary adrenal insufficiency complain of fatigue, anorexia, and weight loss. Other clinical and laboratory manifestations of primary adrenal insufficiency are presented in Table 1. Skin hyperpigmentation, initially on the extensor surfaces, palmar creases, and buccal mucosa, results from increased levels of ACTH and other pro-opiomelanocortin–related peptides, including melanocyte- stimulating hormone. Secondary adrenal insufficiency manifests more insidiously with lack of skin hyperpigmentation, salt craving, metabolic acidosis, and hyperkalemia, because mineralocorticoid secretion is intact. Fatigue, hyponatremia, and hypoglycemia are some of the clinical manifestations in secondary adrenal insufficiency.
|Muscle and joint pain||10|
|Hypotension (systolic blood pressure <110 mm Hg)||92|
Evaluating a patient with suspected adrenal insufficiency is a three-step process: establishing the diagnosis, differentiating between primary and secondary adrenal insufficiencies, and looking for the cause of adrenal insufficiency.
Establishing the Diagnosis. Because of circadian secretion of cortisol and overlap among patients with adrenal insufficiency and those with normal adrenal function, determining the random serum cortisol level is only of value during stress (see later, “Adrenal Insufficiency in the Critically Ill Patient”). An algorithm for the evaluation of adrenal function is shown in Figure 1. An early morning (8 am) plasma cortisol level lower than 3 μg/dL confirms adrenal insufficiency, whereas a value higher than 15 μg/dL makes the diagnosis highly unlikely. Cortisol levels in the range of 3 to 15 μg/dL may be seen in patients with primary, secondary, or tertiary adrenal insufficiency.
These patients should be further evaluated by the cosyntropin (Cortrosyn) stimulation test (CST), which can be performed at any time during the day. The standard-dose CST uses an intravenous or intramuscular injection of 250 μg cosyntropin with plasma cortisol before and 30 and 60 minutes after the injection. A normal response is a plasma cortisol concentration higher than 18 μg/dL at 30 minutes. Most individuals with normal adrenal function achieve much higher cortisol levels at 60 minutes after cosyntropin injection. For this reason, the 18 μg/dL cutoff value should be applied only for the 30-minute cortisol level. The standard-dose CST is excellent for excluding primary adrenal insufficiency. However, patients with mild partial or recent-onset pituitary ACTH or hypothalamic CRH deficiency (e.g., within 2-4 weeks after pituitary surgery) may have a normal response to 250 μg of cosyntropin in the CST, because the adrenal glands have not undergone significant atrophy and will still respond to very high concentrations of ACTH stimulation. The sensitivity of CST to pick up mild adrenal insufficiency improves when using the low-dose CST (1 μg ACTH1-2 given intravenously); however, this may result in a higher false-positive rate. The lack of a commercially available 1-μg cosyntropin dose may be a potential for error. In most clinical situations, the 30-minute cortisol value during a standard-dose CST has a diagnostic accuracy close to that of a low-dose CST.1
The insulin tolerance test (ITT) and metyrapone test are generally used for the evaluation of patients suspected to have secondary adrenal insufficiency. ITT is considered the gold standard test for the evaluation of hypothalamic-pituitary-adrenal (HPA) axis. The ITT is contraindicated in older patients (>65 years) and those with acute illness, seizure disorders, or cardiovascular-cerebrovascular disease. The result of the metyrapone test has been validated compared with the ITT. Metyrapone blocks the final step in cortisol biosynthesis, resulting in a reduction in cortisol secretion, which in turn stimulates ACTH secretion. This leads to higher levels of 11-deoxycortisol, which is the precursor to cortisol.
The CRH stimulation test requires intact pituitary and adrenal glands for a response. The test has not been well validated for the evaluation of adrenal function and its current role in the evaluation of adrenal function is limited to research protocols.
Differentiation Between Primary and Secondary Adrenal Insufficiencies. This is done through the measurement of the basal plasma ACTH level. An elevated ACTH level is consistent with primary adrenal insufficiency. A low or normal-range ACTH level, with a low cortisol level, confirms the diagnosis of secondary or tertiary adrenal insufficiency.
Determining the Cause of Adrenal Insufficiency. When the biochemical workup is consistent with primary adrenal insufficiency, computed tomography (CT) scanning of the adrenal glands may help with the differential diagnosis. Enlarged adrenal glands or calcifications suggest an infectious, hemorrhagic, or metastatic cause. In rare circumstances, CT-guided percutaneous fine-needle aspiration of enlarged adrenal glands may help establish the diagnosis. Patients with tuberculous adrenal insufficiency usually have evidence of active systemic disease. When the biochemical work up suggests secondary or tertiary adrenal insufficiency, magnetic resonance imaging (MRI) of the pituitary gland is indicated if glucocorticoid therapy as the cause of the secondary adrenal insufficiency has been ruled out.
Patients with Addison’s disease require lifelong replacement with glucocorticoids and mineralocorticoids. The minimal dosage to treat symptoms should be used, starting with hydrocortisone, 12.5 to 15 mg in the morning, and 2.5 to 5 mg at noon to mimic the physiologic pattern. Some patients may need another dose of 2.5 to 5 mg hydrocortisone at about 6 pm if fatigue continues later in the day. Patients require fludrocortisone, 0.05 to 0.2 mg, for mineralocorticoid replacement. The dose is adjusted based on clinical status, including the presence or absence of orthostatic hypotension, hypertension, and electrolyte imbalance. Patients may need to double the dose of fludrocortisone or increase salt intake during the summer, when the weather is hot. Patients with secondary or tertiary adrenal insufficiency do not need mineralocorticoid replacement.
During minor illness (e.g., flu or fever >38° C [100.4° F]) the hydrocortisone dose should be doubled for 2 or 3 days. The inability to ingest hydrocortisone tablets warrants parenteral administration. Most patients can be educated to self administer hydrocortisone, 100 mg IM, and reduce the risk of an emergency room visit. Hydrocortisone, 75 mg/day, provides adequate glucocorticoid coverage for outpatient surgery. Parenteral hydrocortisone, 150 to 200 mg/day (in three or four divided doses), is needed for major surgery, with a rapid taper to normal replacement during the recovery. Patients taking more than 100 mg hydrocortisone/day do not need any additional mineralocorticoid replacement. All patients should wear some form of identification indicating their adrenal insufficiency status.
Acute adrenal insufficiency (adrenal crisis) is a life-threatening emergency, which usually manifests with nausea, vomiting, abdominal pain, and shock. Patients may be previously undiagnosed or have chronic primary adrenal insufficiency, with no or inadequate glucocorticoid replacement. Abdominal tenderness and fever are common findings, and adrenal crisis may manifest as an acute abdomen. In these cases, surgical exploration without glucocorticoid coverage can be lethal. The major hormonal factor precipitating adrenal crisis is mineralocorticoid deficiency. Therefore, adrenal crisis rarely occurs with secondary adrenal insufficiency.
Treatment of adrenal crisis should not be delayed. Diagnostic workup in a patient with no history of AI should include a plasma sample for cortisol and ACTH level determination, immediately followed by an IV bolus of hydrocortisone, 100 mg, and adequate fluid replacement (normal saline). Hydrocortisone should be continued, 50 mg every 8 hours, while awaiting laboratory results.
The overall incidence of AI in critically ill patients is less than 10%, but an incidence as high as 50% in a patient with septic shock has been reported. The concept of total adrenal insufficiency has gradually been replaced by relative adrenal insufficiency, which may be fatal in critically ill patients. Hypotension in patients with adrenal insufficiency may mimic hypovolemic or septic shock and should be considered in the differential diagnosis. Serial follow-up of adrenal function in critically patients with clinical features suggestive of AI is recommended.
Intensive care unit (ICU) patients with hemodynamic instability, despite fluid resuscitation (especially in the presence of shock), should be tested for AI. A random serum cortisol level determination and the standard CST are the two commonly used tests for evaluating adrenal function in these patients. However, the cortisol level that reflects an adequate response is uncertain. In critically ill patients with near-normal albumin levels (>2.5 g/dL), a random plasma cortisol level lower than 15 μg/dL, and a maximum cortisol level lower than 20 μg/dL during the CST strongly suggest adrenal insufficiency. In patients with equivocal biochemical results, a trial of 2 or 3 days of stress dosage glucocorticoids is appropriate, as long as it will be discontinued in the absence of any significant hemodynamic improvement. The value of the delta cortisol during the CST (stimulated minus basal cortisol levels) of 9 μg/dL or lower as an indicator of partial AI during septic shock is a matter of debate and does not have strong literature support.
More than 90% of circulating cortisol is bound to cortisol- binding globulin (CBG) and albumin. Free cortisol is biologically active. During severe illness caused by a marked decrease in the CBG level, albumin has a more significant influence on the total concentration of cortisol. Critically ill patients with significant hypoalbuminemia (albumin level >2.5 g/dL) have subnormal serum total cortisol concentrations, but their baseline or cosyntropin-stimulated serum free cortisol level is in the high-normal range or elevated.3 In patients with hypoalbuminemia, total serum cortisol is a poor indicator of glucocorticoid activity. Studies are needed to establish normative ranges for free cortisol for different levels of stress, including septic shock.3
Hydrocortisone, 50 mg IV every 6 to 8 hours is an adequate replacement dose for critically ill patients with suspected adrenal insufficiency. Treatment with this dose should be continued for 2 or 3 days. After hemodynamic improvement, a gradual taper of hydrocortisone, depending on the patient’s condition, should be instituted. A subset of patients with septic shock may benefit from physiologic stress dose glucocorticoids. Such therapy has not yet been proven to be effective and safe in all patients with septic shock. Lifelong glucocorticoid replacement therapy should not be sanctioned on the basis of an equivocal biochemical test result in an acutely ill patient. Adrenal function should be re-evaluated after recovery from acute illness.
Diseases related to adrenal hyperfunction are relatively rare, but they have significant mortality and morbidity if untreated. In the following sections, Cushing’s syndrome, primary hyperaldosteronism, pheochromocytoma, and androgen-producing adrenal tumors are reviewed.
Cushing’s syndrome (CS) is composed of symptoms and signs associated with prolonged exposure to inappropriately high levels of plasma glucocorticoids. Exogenous glucocorticoid intake is the most common cause of CS. The endogenous causes are divided into ACTH-dependent and ACTH-independent CS, as shown in Box 1.
|Box 1 Etiology of Endogenous Cushing’s Syndrome|
|Cushing's syndrome (67%)|
|Ectopic ACTH secretion (12%)|
|Ectopic CRH secretion (<1%)|
|Adrenal adenoma (10%)|
|Adrenal carcinoma (8%)|
|Micro-and macronodular adrenal hyperplasias (1%)|
ACTH, adrenocorticotropic hormone; CRH, corticotropin-releasing hormone.
Harvey Cushing initially described the clinical features of CS in the early 20th century, including centripetal obesity, moon face, hirsutism, and plethora; however, such a classic clinical picture is not always present and a high index of suspicion is usually required. Box 2 describes some of the clinical features suggestive of CS. Weight gain is almost always present, except in ectopic ACTH secretion (EAS) caused by malignancy. The striae in CS are red-purple and usually wider than 1 cm. The skin is thin and minimal trauma results in easy bruising. Patients have a plethoric appearance and acne may be present. Proximal myopathy involving the lower limb and shoulder girdle may be present. Moon face and supraclavicular and dorsocervical fat pads (buffalo hump) are nonspecific and accompany obesity from other causes.
|Box 2 Clinical Features Suggestive of Cushing’s Syndrome|
|Wide purplish striae (>1 cm)|
|Changes in serial photographs|
Women have menstrual irregularity and hirsutism. Men and women exhibit loss of libido. Agitated depression and lethargy are among the most common psychiatric abnormalities seen in CS patients, but paranoia and overt psychosis may occur. Irritability is an early symptom and insomnia is common. CS has profound effects on bone, causing poor linear growth in children and osteoporotic vertebral collapse and pathologic fractures in adults.
Patients with ectopic ACTH syndrome caused by small cell lung carcinoma lack many of the typical clinical features. The rapid course and high levels of ACTH and cortisol often result in hyperpigmentation, myopathy, peripheral edema, glucose intolerance, and hypokalemic alkalosis. Female patients with adrenal carcinomas often present with signs and symptoms of virilization (hirsutism and acne), breast atrophy, deepening of the voice, temporal hair recession, and clitoromegaly caused by hypersecretion of androgens, along with the cortisol.
Pseudo-Cushing’s syndrome refers to features of CS combined with some features of hypercortisolism that resolve after resolution of the underlying cause, such as psychiatric disorders (e.g., major depression, anxiety disorder, obsessive-compulsive disorder, anorexia nervosa), morbid obesity, poorly controlled diabetes, and chronic alcoholism.
CS as a result of exogenous administration of glucocorticoids must be excluded. The diagnosis of Cushing’s syndrome then involves two steps: demonstration of inappropriate cortisol secretion and localization of its cause.
Screening for CS may be done by a 24-hour urinary free cortisol (UFC) determination, 1-mg dexamethasone suppression test (DST), or midnight salivary cortisol test (Fig. 2). Currently, a 24-hour UFC test (with simultaneous urinary creatinine level measurement) is the most widely used initial screening test but a midnight salivary cortisol determination is becoming a good alternative. An elevated salivary cortisol level in most cases should be confirmed by a 24-hour UFC test before referral to surgery. A normal sleeping pattern is necessary for an accurate salivary cortisol level measurement for the evaluation of CS. Values of UFC above three or four times upper normal for the assay are usually diagnostic for CS, especially when repeated and confirmed. In patients with a glomerular filtration rate lower than 20 mL/min, urinary cortisol excretion is significantly decreased and may thus be normal, despite excessive cortisol production. On the other hand, normal individuals with excess fluid intake (>4 L/day) may have a false elevation of the UFC.
The 1-mg DST is performed by the administration of dexamethasone, 1 mg at 11 pm, followed by measurement of the plasma cortisol level at 8 am. A cortisol level lower than 1.8 μg/dL is a normal response. It is an excellent test to rule out CS, with less than a 2% false-negative test result; however, because of a high false-positive rate (≤40%), a positive test result needs to be further confirmed by other tests, such as a 24-hour UFC assay.
During the low-dose DST, dexamethasone, 0.5 mg, is taken orally every 6 hours, for a total of eight doses, starting at 6 am. A cutoff cortisol value of 1.8 μg/dL in the morning (at 7 or 8 am), after the last dose of dexamethasone at midnight, has been reported to have an excellent sensitivity and specificity of about 98% in the evaluation of patients suspected to have CS. Measurement of the 24-hour UFC during the second day of the low-dose DST is of lesser value, and usually does not add any further to the diagnostic accuracy of the baseline 24-hour UFC test.
The combined DST-CRH stimulation test may be used for differentiating between CS and pseudo-Cushing’s syndrome. The test consists of the oral administration of 0.5 mg dexamethasone every 6 hours for eight doses, starting at noon, followed by the IV administration of 100 μg CRH at 8 am (2 hours after the last dose of dexamethasone). A plasma cortisol level higher than 1.4 μg/dL 15 minutes after the CRH injection strongly suggests CS.
Certain drugs, such as phenytoin, phenobarbitals, and rifampin, increase the clearance rate of dexamethasone, resulting in false- positive results during a DST. In these cases, it may be necessary to measure the plasma dexamethasone level. It is always important to remember the effect of increased levels of CBG on serum cortisol levels during pregnancy or in women taking estrogens. These states do not affect the 24-hour UFC or midnight salivary cortisol levels.
Because of the challenging nature of CS, reaching the proper diagnosis requires a stepwise evaluation, knowledge of the limitations of each test, and avoidance of shortcuts (see Fig. 2). Discrepant or inconclusive test results require re-evaluation of the entire clinical picture and a biochemical workup, because CS is almost always a progressive disease. Another clinical scenario that may be associated with normal screening tests is cyclic CS, which needs to be ruled out by periodic evaluation. Fortunately, true cyclic CS is a rare condition.
Once Cushing’s syndrome is biochemically confirmed, the plasma ACTH level should be measured, preferably in the morning. Normal values typically range between 5 and 50 pg/mL. A suppressed or low ACTH level (<10 pg/mL) is consistent with ACTH-independent CS and should be followed by adrenal CT scanning. ACTH values of 10 to 20 pg/mL may be seen in patients with both adrenal and pituitary causes for CS; these patients should undergo a CRH stimulation test. A flat response of ACTH to CRH during the test suggests an adrenal cause, but a more than 50% increase in the ACTH level during the test is consistent with Cushing’s disease. ACTH levels higher than 20 pg/mL suggest ACTH-dependent CS. About 90% of patients with ACTH-dependent CS have a pituitary cause and the rest are ectopic in origin. Pituitary MRI with gadolinium enhancement should be performed. In patients with ACTH-dependent CS, the presence of a pituitary adenoma larger than 6 mm strongly suggests a pituitary origin, but 50% of CS patients do not have any abnormality on MRI.4 ACTH levels tend to be higher in ectopic CS compared with Cushing’s disease, but there is significant overlap.
A combination of the CRH stimulation test and an overnight high-dose DST are used to differentiate ectopic CS from Cushing’s disease. A more than 50% increase in the ACTH level after the CRH test and more than an 80% reduction in the morning cortisol level (8-9 am) after taking 8 mg dexamethasone at 11 pm during a high-dose DST is consistent with a pituitary source, and in the presence of a pituitary adenoma, almost establishes the definitive diagnosis of CS. If the CRH stimulation test and DST results are not concordant and MRI does not show a pituitary adenoma, then inferior petrosal sinus sampling to distinguish ectopic from Cushing’s disease is indicated. Localizing tumors that produce ectopic ACTH is accomplished by chest and abdominal CT studies, followed by neck CT if no source is found. An octreotide scan may be of some value in patients with ectopic CS and negative imaging studies.
Surgical (trans-sphenoidal) removal of the ACTH-secreting pituitary tumor is the treatment of choice. Experienced neurosurgeons usually achieve 70% to 80% long-term remission rates following surgery. An undetectable cortisol level postoperatively, when the patient is off glucocorticoids, is considered to be an excellent indication of long-term cure.
Cushing’s syndrome caused by an adrenal adenoma is usually cured by laparoscopic unilateral adrenalectomy. Adrenal carcinoma is typically an aggressive tumor with a poor prognosis; surgical resection at an early stage, along with lifelong mitotane therapy started soon after surgery, offers the only chance for cure or long-term remission.
Surgical removal of an ectopic ACTH-producing tumor, if possible, results in cure. When the tumor is not resectable, bilateral adrenalectomy is performed to correct hypercortisolemia. Medical therapy with ketoconazole, metyrapone, aminoglutethimide, or mitotane may be considered for patients with a limited life expectancy or for alleviation of hypercortisolemic symptoms before surgery.
During an international workshop in Italy, in October 2002, a relatively comprehensive consensus statement about the diagnosis and complications of Cushing’s syndrome was published.4 The workshop recommended a low threshold for screening patients for CS, including patients with metabolic syndrome, especially if young and resistant to conventional treatment. The consensus emphasized that if the diagnosis of Cushing’s is suspected clinically but initial screening tests are normal, the patient should be re-evaluated at a later date and invasive procedures postponed. A 24-hour UFC assay, overnight 1-mg DST, and midnight salivary cortisol determination were described as first-line screening tests for CS. The workshop briefly discussed the promising results from the midnight salivary cortisol test, which we believe should be part of the diagnostic workup for most patients suspected to have CS, particularly if early disease is suspected.
Evaluation for the plasma circadian rhythm, midnight plasma cortisol level, 2-day low-dose dexamethasone suppression test (LDDST), and combined LDDST-CRH test were described by the workshop as second-line screening tests. In our experiences and in spite of being labor intensive, the combined LDDST-CRH test is of significant value in differentiating CS from pseudo-Cushing’s syndrome. The workshop recommended the use of bilateral inferior petrosal sinus sampling (BIPSS) in patients whose clinical, biochemical, or radiologic study results are discordant or equivocal. We usually carry out BIPSS in most of our patients with ACTH- dependent CS who have no clear adenoma on pituitary MRI. After pituitary tumors are treated, lifelong medical follow-up is necessary to detect early recurrence, monitor hormone replacement, and treat any complications related to the tumor.
Conn first described primary hyperaldosteronism in 1955 in a patient with an adrenal adenoma. Hyperaldosteronism may be more common than once believed. Some investigators have proposed a prevalence as high as 10% in hypertensive patients.5 Women in their fourth to sixth decade of life are affected more often than men. A solitary aldosterone-producing adenoma (65%) and bilateral idiopathic hyperplasia (30%) are the most common subtypes of primary aldosteronism. The adenomas are usually benign and smaller than 2 cm in diameter. Idiopathic adrenal hyperplasia may be accompanied by adrenocortical nodules and is associated with lower aldosterone levels and less severe hypertension, compared with adenomas.
Two forms of familial hyperaldosteronism (FH) have been described: FH type I and FH type II. FH type I, or glucocorticoid-remediable hyperaldosteronism (GRH), is an autosomal dominant disease characterized by a chimeric gene between the 11β-hydroxylase and aldosterone synthase, with varying degrees of hyperaldosteronism, which responds to exogenous glucocorticoids. FH type II is an autosomal dominant disorder of both the aldosterone-producing adenoma (APA) and idiopathic hyperaldosteronism (IHA).
The clinical picture varies from asymptomatic to symptoms related to hypertension, hypokalemia, or both. Patients may have headaches, polyuria, nocturia, polydipsia, parasthesias, weakness, and muscle cramps. There are no specific physical findings. The degree of hypertension is usually moderate to severe, and may be refractory to conventional antihypertensive agents. Malignant hypertension and leg edema are rare. The left ventricular hypertrophy is disproportionate to the level of blood pressure and improves after treatment of hyperaldosteronism, even if hypertension persists.
Routine laboratory tests may show slightly high serum sodium levels (143-147 mEq/L), hyperglycemia, hypokalemia, metabolic alkalosis, and hypomagnesemia. Although most patients with hyperaldosteronism are not hypokalemic, a low serum potassium level may be noted, either spontaneously or after thiazide or loop diuretic use. Hypokalemia may be severe and difficult to correct. Its presence reduces the secretion of aldosterone and thus should be corrected before the laboratory evaluation of hyperaldosteronism.
The workup of a patient for primary hyperaldosteronism involves the following steps: screening tests for primary hyperaldosteronism, establishing the autonomy of aldosterone secretion, and determination of the source of hyperaldosteronism (Fig. 3).
The following categories of patients should be tested with priority for primary hyperaldosteronism:
Although hypokalemia in a hypertensive patient is suggestive of hyperaldosteronism, normokalemia does not exclude the diagnosis. Up to 60% of patients with hyperaldosteronism do not have hypokalemia. Adequate sodium intake may be necessary to unmask the hypokalemia. Measurement of the 24-hour urinary potassium level can be useful in assessing the cause of a low potassium level, including surreptitious vomiting or laxative abuse. Inappropriate urinary potassium excretion of more than 30 mEq/24 hours in a patient with hypokalemia suggests primary hyperaldosteronism, especially if plasma renin activity (PRA) is low.
The ratio of the plasma aldosterone concentration (PAC) to PRA (PAC/PRA) is the best screening test for primary hyperaldosteronism. The test can be done while the patient is on antihypertensive medications (except spironolactone and eplerenone), without requiring postural stimulation. Both spironolactone and eplerenone should be discontinued for 6 weeks before biochemical testing and after the potassium level reaches the normal range. A PAC/PRA above 20, with a concomitant PAC above 10 ng/dL, needs to be pursued by confirmatory tests (see Fig. 3). Low or suppressed PRA during therapy with angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) suggests hyperaldosteronism.
Establishing Autonomous Aldosterone Secretion. An elevated PAC/PRA by itself is not diagnostic for hyperaldosteronism and must be confirmed by a 24-hour urinary aldosterone level determined during 4 days of oral salt loading by adding one flat teaspoon of salt to the meals every day. Diuretics, ACE inhibitors, and ARBs should be discontinued for 2 weeks before the tests and potassium needs to be replaced to the normal range. Hypertension may be controlled by calcium channel blockers, beta blockers, or α1-adrenergic receptor blockers, which do not significantly affect the biochemical workup for hyperaldosteronism. Potassium supplementation is stopped in most patients without any underlying heart disease at the beginning of the salt-loading test. During day 4 of salt loading, a 24-hour urine sample must be collected by the patient for aldosterone, sodium, potassium, and creatinine level testing. A urinary sodium level higher than 200 mEq/24 hours confirms an adequate salt load. An aldosterone level higher than 12 μg/day during salt loading is almost always diagnostic of hyperaldosteronism. Hypokalemia with an inappropriately high urine potassium loss supports the diagnosis. Another supportive measure is correction of hypokalemia with the addition of aldactone at the end of biochemical workup.
IV administration of 2 L of isotonic saline over 4 hours in the recumbent patient is a less favored way to establish the diagnosis. A plasma aldosterone level higher than 10 ng/dL at the end of the infusion supports the diagnosis of primary hyperaldosteronism. The test may entail risk in older patients, those with uncontrolled hypertension, or those with decompensated heart disease.
Establishing the Source of Aldosterone Excess. Because of differences in therapy, distinguishing an aldosterone-producing adenoma APA and idiopathic hyperaldosteronism is important (Table 2). Patients with an APA generally have more severe hypertension, more frequent hypokalemia, higher plasma (>25 ng/dL) and urinary (>30 μg/24 hr) aldosterone levels, and are younger (<50 years). The APA is typically hypodense and smaller than 2 cm on a CT scan of the adrenals. The presence of a solitary adrenal tumor with a normal contralateral adrenal gland is usually consistent with an APA. Determination of plasma aldosterone levels before and 2 hours after an upright posture (with ambulation) and a plasma 18-hydroxycorticosterone level can further help distinguish APA from IHA. Patients with a solitary adrenal tumor, a plasma 18-hydroxycorticosterone level higher than 100 ng/dL, and no significant increase (less than 30%) or a paradoxical decrease in the aldosterone level during a posture test are presumed to have an APA and should be referred for surgery. Patients with an inconclusive adrenal CT scan or discordant results are referred for adrenal venous sampling.
Adrenal venous sampling is technically difficult because the right adrenal vein is small. Adrenal venous sampling is performed during continuous ACTH infusion and relies on demonstration of a gradient for plasma aldosterone in unilateral disease. The cortisol level should be measured to confirm proper catheter placement.
|HTN, potassium concentration [K+]||HTN more severe, higher likelihood of hypokalemia||HTN less severe, less likelihood of hypokalemia|
|Upright posture test||Decrease or <30% increase of serum aldosterone level||Increase by >30% of serum aldosterone level|
|18-Hydroxycorticosterone||>100 ng/mL||<100 ng/mL|
|Computed tomography scan||>1-cm adrenal tumor, with normal contralateral adrenal||No adrenal tumor, bilateral thickening of adrenals|
|Bilateral adrenal venous sampling||Lateralization||No lateralization|
APA, aldosterone-producing adenoma; HTN, hypertension; IHA, idiopathic hyperaldosteronism.
*Clinical, biochemical, and radiologic findings
The treatment goals are to reduce the morbidity and mortality associated with hypertension, hypokalemia, and cardiovascular damage by normalization of blood pressure and aldosterone levels. Unilateral adrenalectomy, usually by a laparoscopic approach, results in normalization of hypokalemia and improvement in hypertension in all patients. About 50% of patients continue to have a milder hypertension. The blood pressure response to spironolactone before surgery often predicts the blood pressure response to surgery in those with an APA.
Medical treatment is reserved for patients with IHA or those with APAs who are poor surgical candidates. Spironolactone, 50 to 200 mg/day, is the treatment of choice. Side effects include painful gynecomastia, nausea, headaches, impotence, and irregular menstruation. Serum potassium and magnesium levels should be monitored to avoid hyperkalemia and hypomagnesemia. Eplerenone is a steroid-based antimineralocorticoid that blocks the aldosterone receptor selectively and has a better side effect profile. However, spironolactone is the favored initial treatment, because it is more potent and costs less. Alternatively, potassium-sparing diuretics such as amiloride or triamterene may be used, but they lack the mineralocorticoid receptor antagonist benefits.
For the first few weeks after surgery, patients should increase their salt intake to compensate for hypoaldosteronism that may occur because of chronic suppression of the renin-angiotensin system. The use of a mineralocorticoid is usually not necessary.
Familial hyperaldosteronism type I, or glucocorticoid-remediable hyperaldosteronism (GRH), is an autosomal dominant disease in which aldosterone is synthesized in the zona fasciculata of the adrenal gland under the control of ACTH.6 Affected individuals are usually younger than 40 years, exhibit hypertension resistant to standard therapy, and have a family history of primary hyperaldosteronism or a cerebrovascular accident at a young age. Most patients have a normal serum potassium level.
Similar to other causes of primary hyperaldosteronism, the PAC/PRA ratio is higher than 20. A 2-day low-dose DST (0.5 mg dexamethasone orally, every 6 hours) will suppress aldosterone to levels lower than 4 ng/dL. The test is not specific. A very high urinary 18-hydroxycortisol level (>3000 nmol/24 hr) or genetic testing to detect a chimeric gene establishes the diagnosis. Genetic testing for glucocorticoid-remediable aldosteronism (GRA) is available for no charge through the International Registry for Glucocorticoid-Remediable Aldosteronism (available at www.brighamandwomens.org/gra). Affected individuals should have neurovascular screening for cerebral aneurysms. Treatment with glucocorticoids is effective to normalize blood pressure. Eplerenone, spironolactone, amiloride, and triamterene have also been used successfully.
There are no consensus statements for the evaluation and therapy of patients with hyperaldosteronism. Although some experts have advocated screening all hypertensive patients for hyperaldosteronism, we recommend screening only high-risk patients. The PAC/PRA ratio is widely accepted as the best initial screening test, with some variations in cutoff values because of differences in patient preparation (e.g., diet, medications, position) or assays used.
There is some disagreement about the role of adrenal venous sampling in patients with biochemical proof of hyperaldosteronism. Adrenal venous sampling is selectively used in some centers, whereas some other centers perform the test in almost all patients.7 In our opinion, if there is a clear adenoma on CT scan on one of the adrenals and the results of the posture test also suggest APA, the patient can be assumed to have the disease and should be referred for surgery. Surgery for APA is more cost-effective than long-term medical therapy.8
Pheochromocytomas (PHEOs) are rare chromaffin cell tumors that may occur at any age. Although their true prevalence is unknown, they occur in about 0.3% of hypertensive patients. If untreated, the disease can have severe consequences, such as myocardial infarction, heart failure, cerebrovascular accident, and death.
Most PHEOs are benign, sporadic, unilateral, and located within the adrenal gland. Extra-adrenal pheochromocytomas (paragangliomas) occur in about 15% of cases in the superior and inferior para-aortic areas, including the Zuckerkandl organ (75%), bladder (10%), thorax (10%), and head, neck, and pelvis (5%). Paragangliomas tend to occur in younger patients <20 years) and are uncommon in those older than 60 years. They are multifocal in about 15% to 30% of cases. Bilateral adrenal pheochromocytomas (5%-10% of cases) are usually seen as part of familial syndromes. Malignant pheochromocytomas (10% of adrenal cases) have a higher prevalence in ectopic PHEOs and lower prevalence in familial PHEOs. No clinical, imaging, or laboratory criteria absolutely predict malignancy; tumors larger than 5 cm have a greater potential to metastasize. The diagnosis of malignant pheochromocytoma relies on the presence of metastases and not on the histology of the tumor.
About 10% to 15% of PHEOs are hereditary in nature. Familial predisposition to pheochromocytoma is seen in patients with multiple endocrine neoplasia (MEN) types IIA and IIB, von Hippel-Lindau disease, neurofibromatosis type 1, and familial paragangliomas. Genetic screening in patients with apparently sporadic pheochromocytoma is recommended for the following categories of patients: age younger than 20 years, bilateral disease, multiple paragangliomas, or family history of pheochromocytoma or paraganglioma.
Patients with PHEO may have paroxysmal hypertension (48%) or persistent hypertension (29%), or be normotensive (13%).9 Wide fluctuations in blood pressure and resistance to antihypertensive medications are typical of those with pheochromocytoma. Norepinephrine-secreting tumors are associated with sustained hypertension and norepinephrine and epinephrine-secreting tumors are associated with labile hypertension. Large cystic PHEOs may not be associated with hypertension because most of the catecholamines are metabolized within the tumor before being released into circulation. Those with a familial form of PHEO are also more likely to be normotensive.
The triad of headaches, palpitations, and diaphoresis suggests the diagnosis of PHEO, but absence of these symptoms does not exclude the disease. Attacks are usually precipitated by emotional stress, exercise, anesthesia, abdominal pressure, or ingestion of tyramine- containing foods. Both pallor and flushing may be seen in patients. Other symptoms include orthostatic hypotension, weight loss, dyspnea, polyuria, polydipsia, visual blurring, focal neurologic symptoms, and change in mental status.
Testing for PHEO is indicated for any patient with suggestive clinical manifestations. Priority has to be assigned to monitoring patients with the following:9
No single test achieves 100% diagnostic accuracy. Plasma-free metanephrines (total free normetanephrine and metanephrine) have 99% and 97% sensitivity, respectively, for sporadic and familial PHEOs, but there is approximately a 10% to 15% false-positive rate for sporadic PHEO. Clinicians should be familiar with their assay’s diagnostic characteristics. In general, a plasma-free metanephrine level more than three or four times the upper normal has 100% specificity for PHEO. Renal failure is only associated with mild increases in plasma-free metanephrine levels. If the plasma-free metanephrine concentration is normal, no other diagnostic test is necessary except in those with small tumors found during workup for familial disease, patients with a history of PHEO or, rarely, patients with a dopamine-producing paraganglioma.
Patients with indeterminate levels of plasma-free metanephrines should have their urinary metanephrine and serum catecholamine levels measured (Fig. 4). Urinary total metanephrine levels higher than 1.8 mg/day are diagnostic for PHEO. Total plasma catecholamine (norepinephrine plus epinephrine) concentration higher than 2000 pg/mL is diagnostic of PHEO, values between 1000 and 2000 pg/mL are highly suggestive of the disease, and values lower than 1000 pg/mL in a patient with severe signs and symptoms at the time of sampling almost always rules out the disease. Urinary vanillylmandelic acid (VMA) measurement has a high false-negative rate (41%) and should not be used for screening purposes, but a positive result has approximately 86% and 99% specificity for sporadic and hereditary PHEOs, respectively.
In patients with nondiagnostic ranges of plasma and urine catecholamine and metanephrine levels, and clinical features suggestive of PHEO, repeating the measurements at the time of symptoms or proceeding with dynamic testing may be used to support or exclude the diagnosis.10 Clonidine normally suppresses plasma catecholamines, a response that is lost in PHEO. During the clonidine suppression test, clonidine, 0.3 mg, is administered orally and plasma catecholamine and metanephrine levels are measured before and 3 hours after clonidine is administered. Normalization of the plasma metanephrine level or a decrease in the plasma catecholamine levels by at least 50%, into the normal range for the assay, is considered a normal response.
The list of medications and interfering substances varies according to what is measured-plasma or urine catecholamines or their metabolites-and to the specific assay used. Thus, it is usually more practical to try to avoid all of them (Box 3). Patients should abstain from caffeinated beverages and alcohol for 24 hours and medications listed in Box 3 for 3 to 5 days before biochemical evaluation. Selective alpha1 blockers (e.g., doxazosin), calcium channel blockers, and beta blockers such as metoprolol or atenolol can be used for blood pressure control during biochemical evaluation and do not cause any significant false-positive results. Because of the potential for inducing a hypertensive crisis, beta blockers should never be used without simultaneously using alpha blockers.
|Box 3 Medications and Stimulants to Avoid Before Measurement of Plasma and Urinary Catecholamines and Metanephrines|
|Monoamine oxidase inhibitors|
|Sympathomimetics-ephedrine, pseudoephedrine, amphetamines, albuterol|
|Stimulants-caffeine, nicotine, theophylline|
|Miscellanous-levodopa, carbidopa, alcohol, cocaine|
Caffeinated beverages and alcohol should preferably be avoided for 24 hours. Medications should be stopped for 3 to 5 days, except labetalol, which needs to be discontinued for 2 weeks (causes analytic interference with some assays).
PHEO is diagnosed in about 6.5% of incidentally discovered adrenal tumors. PHEOs are usually larger than 3 cm in diameter and tend to be cystic, with areas of necrosis with increased size. In a patient suspected to have PHEO, CT scanning of the abdomen and pelvis should be performed first, followed by CT of the chest and neck if no tumor is found (Fig. 5). An adrenal tumor with a noncontrast Hounsfield unit (HU) lower than 10 is extremely unlikely to be PHEO. In a review of more than 30 proven PHEO cases at the Cleveland Clinic, all patients had a noncontrast CT scan HU of more than 20.11
Chemical shift MRI has a sensitivity of 93% to 100% and specificity of approximately 50% in detecting pheochromocytomas. The low lipid content of PHEO tumors results in a signal intensity decrease from the in-phase to opposed-phase T1-weighted image. PHEO tumors typically exhibit signal isointensity with the liver, kidneys, and muscle on T1-weighted images and a characteristically bright, high signal intensity on T2-weighted images (Fig. 6). We usually prefer CT scanning over MRI as the initial imaging study with a biochemical workup diagnostic for PHEO because of cost and widespread availability.
Functional imaging using iobenguane sulfate131I (formerly called meta-iodobenzylguanidine [MIBG]) should be used when multiple or metastatic tumors are suspected, such as in younger patients (<20 years), those with familial pheochromocytoma, tumors larger than 5 cm, or an extra-adrenal tumor, or when CT and MRI fail to localize the tumor despite biochemical evidence supporting the diagnosis of pheochromocytoma (see Fig. 5). The specificity of iobenguane sulfate131I is very high, 95% to 100%, and the sensitivity is about 85%. Nasal decongestants, some antihypertensives, antidepressants, antipsychotics, and cocaine affect iobenguane sulfate131I uptake and have to be withheld for 1 to 3 days before the study. Labetolol in particular has been shown to decrease iobenguane sulfate131I uptake. Uptake of iodine by the thyroid gland should be blocked with potassium iodide (SSKI; five drops three times a day, starting on the day of iobenguane sulfate131I and 3 days afterward). If the iobenguane sulfate131I study is negative, fluorodeoxyglucose positron emission tomography (PET) scanning or oxidronate (Octreoscan) may be used with some success for visualization (see Fig. 5).
Surgical resection is the treatment of choice for pheochromocytoma. Adequate medical preparation is essential and usually achieved in 10 to 14 days. α1-Adrenergic receptor blockers (e.g., doxazosin) are first-line therapy, with increasing dosage as tolerated. Beta blockers (e.g., metoprolol, atenolol) are added if tachycardia develops while the patient is adequately hydrated. Beta blockers should be added only after alpha blockade has been instituted to prevent unopposed alpha receptor activation, which may result in hypertensive crisis. Calcium channel blockers (CCBs) can be used for medical preparation or may be added if there is persistent or labile hypertension. The nonspecific α1-adrenergic blocker phenoxybenzamine has a greater side effect profile and may result in prolonged hypotension after surgery.
Surgery for PHEO has shifted from the open conventional procedure to the laparoscopic approach over the past decade. Patients should have electrocardiography and cardiac echocardiography before surgery, with further cardiac evaluation if any abnormality is detected. It is preferable that patients be admitted to the hospital 1 day before surgery for close observation. In those with orthostatic hypotension, an isotonic saline infusion at a rate of 100 to 200 mL/hr should be started. Close blood pressure monitoring is necessary during surgery. The anesthesia team should be familiar with the care of patients with PHEO during surgery. After surgery, blood pressure and blood glucose levels need to be closely monitored, antihypertensive agents withheld, and normal saline infusion continued. Fluid overload should also be avoided. Antihypertensive agents may be resumed if the patient continues to be hypertensive. Lifelong follow-up is necessary for all patients, with closer follow-up of those with familial, large, extra-adrenal, or bilateral tumors.
Acute hypertensive crises can be treated with IV nitroprusside, nitroglycerin, the short-acting α-adrenergic blocker phentolamine (competitive α1-adrenergic and weak α2-adrenergic receptor antagonist), magnesium sulfate, or nicardipine, a calcium channel blocker. Metastatic lesions should also be resected if possible. In patients with aggressive tumors, combination chemotherapy (cyclophosphamide, vincristine, dacarbazine) or iobenguane sulfate131I-tagged radiotherapy may be considered, but results have not been promising.
There are no consensus statements for the evaluation and therapy of PHEO patients. Most experts’ opinions are based on personal experience in centers with large referral bases. Although there are still disagreements about the best initial screening test for PHEO, serum metanephrine or 24-hour urinary metanephrine level determination appears to be the most favored initial test of choice.
The adrenal glands are an important source of androgens, especially in children and women. The primary adrenal androgens, dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS sulfate), are under ACTH control and have little intrinsic androgenic activity. However, they are converted to androstenedione and then to testosterone (and estrogen) in both the adrenal gland and peripheral tissues. DHEA and DHEA sulfate are responsible for adrenarche (pubic hair development). Peak levels occur in the third decade of life and decline progressively to 25% of peak levels around the age of 80 years.
Features of adrenal androgen excess differ with age and gender. In female neonates, androgen excess causes female pseudohermaphroditism (ambiguous genitalia). Male infants exhibit penile enlargement.
In prepubertal children, boys and girls, androgen excess manifests as increased rate of growth in height and skeletal maturation, leading to premature epiphyseal fusion and short adult height. In addition, boys exhibit penile enlargement, hair growth in androgen-dependent areas, deepening of the voice, and other secondary sexual characteristics (isosexual precocious puberty). Girls have hirsutism, acne, and clitoromegaly (heterosexual precocious puberty).
Androgen excess with onset at puberty causes premature skeletal maturation and short adult height in boys. In girls, it manifests as primary or secondary amenorrhea, different degree of virilization, and increased skeletal maturation, resulting in short adult height.
Manifestations of androgen excess with onset at adult age are limited in men. A decrease in size of the testicles, testosterone secretion, and spermatogenesis caused by inhibition of gonadotropin secretion may occur. In women, hirsutism, acne, menstrual irregularities, male pattern baldness, infertility, decreased breast tissue, increased muscle mass, android body habitus, and clitoromegaly may occur, depending on the degree of hyperandrogenism.
Of the various causes of adrenal hyperandrogenism, nonclassic (adult-onset) congenital adrenal hyperplasia (CAH), and androgen-producing adrenal tumors are discussed in this section. Exogenous androgen intake for body building or to increase erythropoiesis may cause acne, hirsutism, and oligomenorrhea or amenorrhea in women and small testes, gynecomastia, and impaired spermatogenesis in men. Hyperprolactinemia may cause an increase in serum DHEAS and androstenedione levels by direct stimulation of the adrenal gland.
Nonclassic congenital adrenal hyperplasia (NCAH) is an autosomal recessive disorder, with most patients exhibiting a defect in 21-hydoxlyase enzyme activity, resulting in decreased cortisol production. The associated increased ACTH secretion stimulates adrenal steroid production upstream of the defective enzyme, resulting in increased androgen secretion. Hyperandrogenic symptoms are usually diagnosed in late puberty and early adulthood. NCAH affects from 1% to 10% of hyperandrogenic women, depending on their ethnic background. It is rare among African Americans. The main differential diagnosis for NCAH is polycystic ovary syndrome (PCOS), which is about 40 to 50 times more common (Table 3).12 Distinguishing the two only on clinical grounds is difficult, because both can be associated with varying degrees of hyperandrogenism and ovulatory dysfunction. (See elsewhere in this text, “Polycystic Ovary Syndrome,” for further discussion.) DHEAS and testosterone levels are not reliable in differentiating the two. The diagnosis of NCAH is established based on a basal or post–250-μg cosyntropin-stimulated level of 17-hydoxyprogesterone higher than 15 ng/mL (45 nmol/L). Morning basal 17-hydoxyprogesterone levels during the follicular phase of the menstrual cycle (shortly after spontaneous or induced vaginal bleeding) lower than 2 ng/mL essentially rules out NCAH. The diagnosis can usually be confirmed by genotyping.
|Prevalence in reproductive age women||4%-6%||0.1%-0.05%|
|Prevalence in hyperandrogenic patients||50%-80%||1%-10%|
|Racial, ethnic distribution||No predilection||Predominantly white or Ashkenazic Jews|
|Inheritance mechanism||Complex trait||Autosomal recessive|
|Specific hormonal diagnosis||None||ACTH-stimulated 17-HP >15 ng/mL|
ACTH, adrenocorticotropic hormone; 17-HP, 17-hydroxyprogesterone; NCAH, nonclassic congenital adrenal hyperplasia; PCOS, polycystic ovary syndrome.
Glucocorticoids, oral contraceptives (OCPs), antiandrogens such as aldactone, or a combination of these medications are used to treat women with NCAH. OCPs and antiandrogens are usually adequate for therapy when fertility is not desired. Glucocorticoid therapy is often necessary when the patient plans to get pregnant. Prednisone, 2.5 mg orally twice daily, is our preferred initial glucocorticoid therapy, with further adjustment depending on androgen levels. Both CAH and NCAH are associated with an increased incidence of adrenal adenomas and testicular adrenal rest tumors.13 This warrants CT scanning of the adrenal glands and testicular ultrasonography (men) in patients with biochemical confirmation. The consensus statement from the Lawson Wilkins Pediatric Endocrine Society and the European Society for Pediatric Endocrinology about CAH almost exclusively discusses the diagnosis and management of the classic form of the disease. The National Institutes of Health sponsored a clinical staff conference about CAH and NCAH in 1999.13
Primary adrenocortical carcinoma may be associated with excess androgen secretion. Production of steroid intermediates and sometimes cortisol may also occur. It is a rare disease, with an incidence of 1 per 600,000 to 1,600,000 and a prevalence of 4 to 12 per 1,000,000. Female patients may exhibit virilization with very high levels of DHEA sulfate (500 μg/dL or higher), testosterone, and urinary 17-ketosteroids. Primary adrenocortical carcinomas are highly malignant, with a poor prognosis. Tumors are usually larger than 6 cm, invade the capsule, metastasize early, and typically recur after surgery. Surgical resection at an early stage, along with lifelong mitotane therapy starting soon after surgery, offers the only chance for cure or long-term remission. Androgen-secreting adrenal adenomas are rare, typically smaller than 4 cm, and are associated mostly with high levels of DHEAS and testosterone. Surgical resection is the therapy of choice.
An adrenal incidentaloma is an adrenal mass larger than 1 cm in diameter discovered during radiologic examination done for a reason not related to the adrenal gland. The definition of incidentaloma excludes patients undergoing imaging procedures as part of staging and workup for cancer. Incidental adrenal masses are found in 0.5% to 5% of patients undergoing CT of the abdomen and in up to 9% of postmortem autopsies in unselected populations.
Management decisions are based on the need to address whether the tumor is functional (i.e., producing hormones) or malignant.
Evaluation for Hormonal Hypersecretion. Up to 35% of adrenal incidentalomas may be functional. All patients should undergo hormonal evaluation for Cushing’s syndrome and pheochromocytoma, and those with hypertension should also be evaluated for hyperaldosteronism. Isolated excess androgen secretion by adrenal adenomas or carcinomas is rare. Thus, only women with an adrenal mass and physical findings suggestive of hyperandrogenism should have their testosterone and DHEAS levels measured.
Subclinical Cushing’s syndrome (SCS) is a relatively recent and poorly described disorder, occurring in about 5% to 24% of patients with adrenal incidentalomas. The prevalence varies greatly, depending on the diagnostic criteria and the screening methods used. Patients do not have the typical signs and symptoms of Cushing’s syndrome, but generally have increased frequency of hypertension, glucose intolerance, diabetes, and possibly osteopenia. We support the 2002 NIH consensus panel recommendation of a 1-mg overnight DST as the initial biochemical evaluation of choice. Some experts have recommended two or more of the following abnormal tests of the HPA axis for the diagnosis of SCS: lack of suppression during 1-mg overnight DST, increased urinary free cortisol levels, loss of diurnal cortisol rhythm, low or suppressed ACTH level, and impaired ACTH response to CRH. Development of adrenal insufficiency after surgical removal of the adrenal tumor is the best way to confirm the diagnosis.
Evaluation for Malignancy. Adrenal tumor size has been used to differentiate between benign and malignant adrenal masses. Risk of malignancy increases with adrenal tumor size. Cutoff values ranging from 4 to 6 cm have been proposed by different clinicians for surgical resection of adrenal masses. In spite of a relationship between the risk of malignancy and adrenal tumor size, there is a significant overlap, and even a cutoff value of 2 cm could not achieve a 100% specificity to rule out a malignant adrenal mass.11
A noncontrast CT attenuation coefficient (in HU) is the most promising imaging tool to differentiate benign adrenal masses from malignant tumors. Intracytoplasmic fat is often abundant in adrenal adenomas, but rare in adrenal metastases, pheochromocytoma, and adrenocortical carcinomas. Adrenal masses with a noncontrast HU of less than 10 are always benign.11 The noncontrast HU is superior to adrenal size in differentiating benign from malignant adrenal tumors.
Some studies have reported the value of delayed enhancement washout percentage in attenuation value during CT scanning of the adrenal glands to differentiate lipid-poor adrenal adenomas (noncontrast HU of >10) from nonadenomas. An absolute enhancement washout percentage of less than 60% at 15 minutes postcontrast had 95% to 100% specificity in identifying nonadenomas.14 This needs to be further validated by studies of large numbers of patients. MRI is another noninvasive method used to differentiate benign and malignant adrenal masses. Its sensitivity and specificity for diagnosis of a benign adrenal mass seem to be close to the noncontrast CT attenuation value.
Our approach to incidentally discovered adrenal masses is shown in Figure 7. We usually do not routinely obtain any follow-up imaging studies in nonfunctional adrenal masses with a noncontrast HU of less than 10. Such patients undergo annual follow-up evaluations for any excess hormone hypersecretion for 5 years and then intermittently afterward. We recommend that all adrenal tumors larger than 6 cm be surgically removed; however, myelolipoma is an exception, with a characteristically low attenuation value on noncontrast CT scan. Adrenal masses smaller than 6 cm with a noncontrast HU of more than 10 need to have their absolute percentage washout calculated at 15 minutes and undergo follow-up imaging studies in 6 to 12 months (see Fig. 7). There is no good evidence supporting continued radiologic surveillance if the follow-up study at 6 to 12 months shows no change in adrenal tumor size.
Fine-needle aspiration of an adrenal mass may be done to rule out metastasis once pheochromocytoma has been ruled out. Adrenal biopsy may not differentiate adrenocortical carcinomas from adenomas. Surgical resection of the adrenal mass is usually considered for patients with functional or malignant adrenal masses. Medical therapy may be acceptable in the case of primary hyperaldosteronism secondary to adrenocortical adenoma or hyperplasia.
The National Institute of Health (NIH) held a state-of-the-science conference in February 2002, during which the management of adrenal incidentalomas was discussed by a 12-member panel.15 The panel recommended measuring serum aldosterone levels and plasma renin activity only in patients with hypertension and an overnight 1-mg DST and plasma or urine metanephrine levels for all patients with an adrenal incidentaloma.
The panel recommended a noncontrast CT HU threshold of 10 as a reliable radiologic feature to differentiate benign from malignant adrenal tumors. The panel recommended surgical resection of all adrenal masses larger than 6 cm and a repeat CT scan in 6 to 12 months for those with adrenal incidentalomas that are not surgically removed. The panel concluded that there are no data supporting continued radiologic evaluation if an adrenal mass is stable in size. We usually do not obtain a routine follow-up imaging study in adrenal masses with a noncontrast CT HU of lower than 10, because the risk of developing a primary adrenocortical carcinoma from an adenomatous or hyperplastic adrenal mass is extremely rare.
Pharmacologic doses of synthetic glucocorticoids are used in a wide variety of diseases for their anti-inflammatory and immunosuppressive effects.
Two main clinical problems arise with the chronic use of pharmacologic doses of glucocorticoids, iatrogenically induced Cushing’s syndrome and adrenal insufficiency on abrupt cessation of therapy. Cushing’s syndrome may become clinically apparent within 1 month after onset of treatment. The duration of glucocorticoid therapy, the highest dose, and the total cumulative dose have long been considered important predictors of suppression of the HPA axis. Secondary adrenal insufficiency caused by exogenous steroid intake should be anticipated in any patient who takes more than 30 mg of hydrocortisone (or 7.5 mg prednisone) per day for more than 3 weeks.16 Excluding depot glucocorticoid injections, therapy for less than 2 weeks rarely results in clinically significant suppression of HPA axis.
Alternate-day therapy and avoidance of nighttime doses are associated with less suppression of the HPA axis. The time to recovery of the HPA axis after discontinuation of glucocorticoid therapy is variable and can be as short as 2 to 5 days or as long as 9 to 12 months.
Inhaled glucocorticoid use for a long duration may result in suppression of the HPA axis. All topical glucocorticoids are absorbed to some extent, depending on the area of the body (intertriginous area absorption higher than forehead and scalp absorption; least absorption is in the forearm). Inflammation of the skin and application of occlusive dressings increase systemic absorption. Case reports have been published, with patients demonstrating a grossly cushingoid appearance after long-term use of steroid-containing topical creams or eye drops.
There are two important issues that need particular attention when trying to wean a patient from chronic steroid therapy. First, the underlying condition for which the steroid therapy was initiated should always be kept in mind, and any tapering of the dose should be done accordingly. For example, too rapid tapering of glucocorticoid therapy in a patient with asthma may result in exacerbation of the underlying condition. Second, if the underlying disorder for which glucocorticoid therapy was initiated has been resolved, then a rapid tapering of the glucocorticoid dose to about 2 to 3 times the physiologic replacement dose is safe in most patients.
Our approach is to change different glucocorticoid preparations to hydrocortisone, 20 mg in the morning and 10 mg at noon, which after 2 to 4 weeks is changed to 20 mg hydrocortisone once daily in the morning. While on hydrocortisone therapy, patients are advised to double their dose for 2 to 3 days if they develop any acute illness. Patients are then followed at 1- to 2-month intervals, measuring the serum cortisol level during a CST after holding the morning hydrocortisone. A cortisol value higher than 18 μg/dL at 30 minutes during a low-dose (1 μg) or standard-dose (250 μg) CST indicates normal recovery of adrenal function. Those patients with levels between 14 and 18 μg/dL may only take hydrocortisone during acute illness, with re-evaluation of their adrenal axis at 2- to 3-month intervals to ensure full recovery. All patients chronically treated with glucocorticoids should wear medical alert tags and be instructed about supplemental coverage during infection or surgery and the need for parenteral glucocorticoid therapy if unable to tolerate oral medications.