Information
- Diagnosis of Adrenal Insufficiency in Critical Illness: The diagnosis of cortisol deficiency in the setting of critical illness is challenging. Under normal conditions, cortisol is secreted diurnally; however, during critical illness, this diurnal variation is lost. Cortisol secretion is naturally increased during times of physiologic stress, which can result in unmasking of preexisting subclinical adrenal insufficiency. Critical illness may also cause a functional deficiency in cortisol production or responsiveness. Improvement of critical illness in response to glucocorticoids does not necessarily imply diminished adrenal function. It is important to make a conceptual distinction between treatment of adrenal insufficiency and pharmacologic treatment with glucocorticoids, which may improve clinical outcomes independent of adrenal functional status.
- Total Versus Free Cortisol: Cortisol binds to cortisol-binding globulin (CBG) in the blood. Widely available cortisol assays measure total cortisol, although unbound “free” cortisol is physiologically active. Critical illness is often associated with reductions in the level of CBG, as is cirrhosis; thus, total cortisol may be reduced without reduction of free cortisol, leading to overdiagnosis of adrenal insufficiency. As a result of the wide variation in total cortisol levels in septic shock, total cortisol is not a useful test for adrenal insufficiency. Free cortisol assays are not widely available.
- Adrenocorticotrophic Hormone Stimulation Test: The main test for cortisol deficiency is the ACTH stimulation test. A blood sample for cortisol and ACTH is obtained, and synthetic ACTH (cosyntropin) is administered IV or IM. A blood sample for a second cortisol measurement is obtained 30 to 60 minutes after cosyntropin administration. The dose of cosyntropin administered in the stimulation test is a matter of debate.
- High-dose adrenocorticotrophic hormone stimulation test: This test uses an IV dose of 250 µg cosyntropin. One prospective study suggested that a high baseline serum cortisol (>34 µg/dL) level coupled with a diminished cortisol increase (<9 µg/dL) to a high-dose stimulation test is predictive of increased mortality.
- Low-dose adrenocorticotrophic hormone stimulation test: This test uses 1 µg of cosyntropin for stimulation. Advocates of the low-dose stimulation test have argued that the 250 µg cosyntropin stimulation test stimulates cortisol release even in patients who are adrenally insufficient. The low-dose ACTH stimulation test identifies more patients with adrenal insufficiency. In a study comparing the low-dose versus high-dose cosyntropin test, it was found that a cortisol response to the high-dose, but not low-dose, cosyntropin test was predictive of increased mortality.
- Diagnostic criteria for adrenal insufficiency: Because glucocorticoid levels are normally elevated in response to stress, a very low baseline plasma cortisol (<3 µg/dL) in the setting of critical illness is diagnostic of adrenal insufficiency. Moderate levels of baseline cortisol (<10 µg/dL) suggest adrenal insufficiency but may be misleading in the setting of a low CBG where free cortisol may be appropriately elevated. A baseline (nonstimulated) cortisol of more than 18 µg/dL effectively rules out adrenal insufficiency in most patients. For intermediate cortisol levels, a stimulated rise of less than 9 µg/dL has been proposed to identify patients who would benefit from glucocorticoid therapy. A peak level of more than 18 µg/dL effectively excludes primary adrenal insufficiency, although CBG may be elevated in response to oral estrogen and liver inflammation limiting the clinical significance of total cortisol levels.
- Limitations: The utility of the ACTH stimulation test and the appropriate stimulation dose remains unresolved. The ACTH stimulation test does not determine whether sufficient cortisol is being produced by the hypothalamic-pituitary-adrenal axis; it measures only the ability of the adrenal cortex to respond to exogenous ACTH. As a result, it should not be used to diagnose secondary adrenal insufficiency. It may also be altered in the case of etomidate administration due to pharmacologic suppression. In the case of recent-onset central adrenal insufficiency, the adrenal cortex may not yet have atrophied and the stimulated cortisol response may be normal.
- Therapy: Therapy should be directed at addressing the cause of the adrenal insufficiency and steroid replacement. Ideally, cortisol and ACTH levels should be drawn before any empiric therapy is begun. Empiric therapy with dexamethasone (1 mg q6h) does not interfere with the ACTH stimulation test. Once the post-cosyntropin cortisol sample has been drawn, hydrocortisone, which provides complete glucocorticoid and mineralocorticoid replacement, should be used at a total daily dose of 300 mg in divided doses every 6 or 8 hours, or given via continuous infusion at 10 mg/h. Although dexamethasone lacks mineralocorticoid activity, it can nevertheless be used in an adrenal crisis because glucocorticoid deficiency is the primary driver of hemodynamic instability. It should be noted that the appropriate “stress” dose of glucocorticoids is a matter of controversy. Commonly used dosages may provide glucocorticoid activity in excess of that produced by the normal adrenal gland during critical illness. There is no reliable test to determine the appropriateness of the replacement dose, so the dosing must be adjusted empirically. Doses of hydrocortisone should not be reduced below a replacement dose for a patient who is hospitalized with known adrenal insufficiency, 50 to 60 mg orally (PO) divided in two doses, until follow-up testing demonstrates adequate adrenal function. The equivalent dose of prednisone is 10 to 15 mg daily, whereas the equivalent for dexamethasone is 1.5 to 2.5 mg daily. Doses of hydrocortisone less than 50 mg daily do not provide sufficient mineralocorticoid activity for patients with primary adrenal insufficiency. Note that prednisone and dexamethasone have minimal mineralocorticoid activity. Patients requiring mineralocorticoid replacement should be treated with fludrocortisone at doses of 0.1 to 0.2 mg po daily. Long-term risks of glucocorticoids include hyperglycemia, weight gain, and hypertension. See Table 29.6 for the relative glucocorticoid and mineralocorticoid activities of various steroids.
- Identifying the Cause of Adrenal Insufficiency: An adrenocorticotropin (ACTH) level drawn before initiation of empiric glucocorticoid therapy can help localize the cause of adrenal insufficiency. An elevated ACTH suggests primary adrenal insufficiency and a low (or inappropriately normal) ACTH is consistent with central adrenal insufficiency. If the test suggests primary cortisol insufficiency (elevated ACTH levels), the evaluation should include imaging of the adrenal glands to evaluate for neoplastic, inflammatory, or infiltrative processes. Adrenal protocol computed tomography (CT) scans include a noncontrast series, portal venous phase contrast-enhanced series, and delayed contrast-enhanced series. In the setting of central cortisol deficiency, imaging of the pituitary and hypothalamus is indicated. Pituitary protocol magnetic resonance imaging (MRI) is the most appropriate test, although CT can rule out large tumors or gross hemorrhage. See Table 29.5 for the differential diagnosis of adrenal insufficiency.
- Pituitary Apoplexy: Pituitary apoplexy, also known as Sheehan syndrome, is a clinical syndrome caused by hemorrhage or infarct within a preexisting pituitary mass lesion. It is a rare cause of adrenal insufficiency but it deserves special mention in the ICU because it is one of the true endocrine emergencies. Pituitary apoplexy can lead to abrupt and severe adrenal insufficiency in the setting of severe physiologic stress, a combination that may prove fatal if not treated promptly. In addition, mass effect on surrounding structures, including the optic nerve and cranial nerves III and VI, can lead to permanent visual deficits or blindness. Therefore, pituitary apoplexy should always be considered in the differential diagnosis of adrenal insufficiency. Unfortunately, the signs and symptoms of apoplexy (headache, visual disturbance) may be masked in the critically ill. Imaging is key to making the diagnosis. Noncontrast head CT is not reliable for detection of pituitary hemorrhage or infarction but is sensitive for detecting pituitary masses greater than 1 cm, the substrate for most cases of apoplexy. MRI can be used to identify hemorrhage or infarction if a mass is found. Management of pituitary apoplexy involves high-dose glucocorticoids and rapid surgical decompression.