Adrenal Function in Critically Ill Patients, Evaluation and Management

Key Points

• Adrenal failure can be absolute/intrinsic, which is due to pathology within the components of hypothalamus-pituitary-adrenal axis (referred to as adrenal crisis)
• Critical illness–related corticosteroid insufficiency (previously mislabeled relative adrenal failure) refers to the concept of acquired adrenal insufficiency due to dysregulation of the hypothalamus-pituitary-adrenal axis in prolonged critical illness
  • Diagnostic methodology and management remain difficult and controversial
• Prevalence of adrenal insufficiency in ICU patients is difficult to discern owing to lack of agreed-upon reliable diagnostic measures
• Diagnosis requires consideration of history, examination, and ICU course in the context of laboratory studies including plasma ACTH (adrenocorticotropic hormone) levels, serum free cortisol levels, and possibly incremental serum free cortisol responses to ACTH stimulation test or incremental ACTH responses to a corticotropin-releasing hormone stimulation test
• This author recommends a hydrocortisone dose of approximately 60 mg/day as a sufficient substitute for a failing adrenal cortex in critically ill patients and a pharmacologic dose of 200 mg/day as a brief intervention for refractory shock
• Upon discharge, patients should have follow-up consultation with endocrinologist and be monitored for long-term sequelae of glucocorticoid therapy

Terminology

• Absolute adrenal failure, also termed intrinsic adrenal failure or adrenal crisis, can be due to pathology in 1 of the 3 levels of the hypothalamus-pituitary-adrenal axis
  • Primary adrenal failure refers to disease within the adrenal cortex
  • Secondary adrenal failure refers to disease within the anterior pituitary gland
  • Tertiary adrenal failure refers to disease within the hypothalamus
• Relative adrenal failure, an outdated term for a concept that was diagnosed using the result of a short ACTH (adrenocorticotropic hormone) stimulation test, refers to a maximally activated adrenal cortex that would not result in sufficiently high systemic cortisol availability to allow survival;¹ currently considered a misnomer
• Critical illness–related corticosteroid insufficiency refers to acquired adrenal insufficiency due to dysregulation of the hypothalamus-pituitary-adrenal axis in patients in the ICU²

Background

• The study of adrenal dysfunction and its management in critically ill patients has continued evolving because of differing interpretations of research findings and subsequent controversies over optimal management approaches
• Increased systemic cortisol availability is essential for survival of critical illnesses through the key roles of cortisol in the following:^3,4
  • Controlling inflammation
  • Providing essential metabolic substrates
  • Fluid retention
  • Activation of the cardiovascular system to increase blood pressure and cardiac output
• Insufficient cortisol for the body’s physiologic stress response can be due to 1 of the following:
  • Absolute adrenal failure, which is an intrinsic disease within the hypothalamus-pituitary-adrenal axis
  • Critical illness–related corticosteroid insufficiency, which is dysfunction within the hypothalamus-pituitary-adrenal axis acquired during the course of critical illness
• Absolute failure of the hypothalamus-pituitary-adrenal axis can be the primary cause of critical illness
  • Patients suffering from critical illness evoked by such absolute adrenal failure represent only a minority of the ICU population
• In contrast, some patients are thought to have partially impaired hypothalamus-pituitary-adrenal axis function secondary to their critical illness
  • This was initially termed relative adrenal failure to differentiate it from absolute failure¹
  • Formerly thought to be due to a maximally activated adrenal cortex, which does not result in sufficiently high systemic cortisol availability to allow survival
    • This was based on results of an association study that showed that an incremental change in total cortisol of 9 mcg/dL or less (irrespective of baseline plasma cortisol) in response to a 250 mcg bolus of synthetic ACTH (cosyntropin) was most predictive of mortality¹
    • Such a result of the ACTH stimulation test was further interpreted as necessitating treatment with stress doses of hydrocortisone to increase odds of survival^2,5
  • Further randomized controlled trials have revealed contradictory findings,^5,6,7 in part because of confounding by the use of etomidate,⁵ a drug that suppresses cortisol synthesis
  • Studies in critically ill patients (including those with septic shock) have now shown that plasma ACTH levels are typically low or low-normal irrespective of the level of plasma cortisol^8,9,10
    • These low plasma ACTH levels do not support the concept of a maximal centrally activated adrenal cortex
  • Therefore, the term relative adrenal insufficiency is a misnomer and the concept evolved and has been renamed critical illness–related corticosteroid insufficiency²
    • Thought to be due to failure of 1 or more sites in the hypothalamus-pituitary-adrenal axis, which could involve a damaged or dysregulated hypothalamus, anterior pituitary gland, or adrenal cortex
• Research has subsequently shown, however, that use of the term relative adrenal failure and the current practice guidelines for diagnosis and treatment of critical illness–related corticosteroid insufficiency² are flawed, on the following basis:
  • Increased systemic cortisol availability in ICU patients has been shown to be mostly driven by low plasma binding and reduced cortisol breakdown rather than by increased adrenocorticotropic hormone–driven cortisol production^{8,10,11,12,13}
  • The low plasma ACTH levels are not due to organ damage but can be explained by negative feedback inhibition exerted at the central level of the hypothalamus-pituitary-adrenal axis by peripherally driven high free cortisol availability^8,11,12,14
  • Low total incremental cortisol responses to the ACTH stimulation test are caused by increased cortisol distribution volume^8,10 because of low plasma binding of cortisol
    • This finding invalidates use of the ACTH stimulation test for diagnostic purposes in the ICU context^2,8,11,14
  • Hence, the subset of patients with septic shock who present with a low total incremental cortisol response to the ACTH stimulation test and who respond to steroids with improved hemodynamic parameters should not be considered as suffering from critical illness–related corticosteroid insufficiency
    • The effect should be interpreted as a pharmacologic response to steroid treatment
• In addition, recent studies suggest that specifically prolonged critically ill patients who require intensive care for weeks or longer, as opposed to acute critically ill patients suffering from septic shock, may be at risk of developing acquired absolute central adrenal insufficiency because of sustained lack of trophic ACTH signaling^10,11,12,14
  • This condition represents critical illness–related corticosteroid insufficiency, which is not caused by hypoxic or inflammatory damage of the hypothalamus-pituitary-adrenal axis; rather, it is the consequence of sustained central activation of the glucocorticoid receptor by cortisol or other glucocorticoid receptor ligands

Epidemiology

• Prevalence of any type of adrenal insufficiency in critically ill patients treated in ICUs is unknown because of limitations of plasma cortisol quantification^8,13 and inappropriateness of the ACTH stimulation test¹⁰ in this population

Etiology and Risk Factors

• Absolute adrenal failure, either primary or central, as the intrinsic cause of critical illness requiring intensive care is mostly of autoimmune, infectious, or infiltrative nature
  • May also be evoked by hemorrhage or surgery or by drugs affecting cortisol synthesis or breakdown^3,4,15

• Table 1. Etiologies of intrinsic absolute adrenal failure.

  ACTH, adrenocorticotropic hormone; CMV, cytomegalovirus; DIC, disseminated intravascular coagulation; HPA hypothalamus-pituitary-adrenal; TB, tuberculosis.

  Source	Pathophysiology	Findings	Comments
  Intrinsic absolute adrenal failure			Absolute failure of component(s) of the HPA axis can be the primary cause of critical illness but is uncommon in ICU populations
  Primary	Disease within adrenal cortex	↓ Cortisol, ↑ ACTH	Infectious: disseminated TB, HIV, CMV, histoplasmosis, cryptococcus Adrenal hemorrhage: DIC, supratherapeutic anticoagulation, trauma, Waterhouse-Friderichsen syndrome Autoimmune: due to antiadrenal autoimmunity in polyendocrine failure syndrome Metastatic disease: including lymphoma, lung cancer, breast cancer, melanoma Adrenoleukodystrophies Congenital adrenal hypoplasia Bilateral adrenalectomy Drug related: rifampin, azoles, etomidate, phenytoin, mitotane
  Secondary	Disease within the anterior pituitary	↓ Cortisol, ↓ ACTH	Infectious (eg, HIV, TB) Pituitary tumors Pituitary surgery trauma Infarction Hemorrhage Infiltrative disorders Metastatic disease Drug related (eg, megestrol acetate, opiates, chronic glucocorticoid therapy)
  Tertiary	Disease within hypothalamus	↓ Cortisol, ↓ ACTH	Mass lesion (eg, craniopharyngioma) Radiation Infiltrative lesion (eg, sarcoidosis) Trauma Infection (eg, viral encephalitis) Drug related (eg chronic glucocorticoid therapy)

• Acquired absolute central adrenal insufficiency may develop in critically ill patients requiring intensive care for several weeks or longer owing to sustained suppressed circulating ACTH with risk of adrenocortical atrophy^10,12,16
  • Condition is known as critical illness–related corticosteroid insufficiency and the following may play a role:^11,17,18
    • Peripherally increased circulating cortisol
    • Prolonged treatment with opioids or other drugs affecting the hypothalamus-pituitary-adrenal axis
    • Presence of endogenous glucocorticoid receptor–binding ligands distinct from cortisol (eg, bile acids)
• Etiology of metabolic derangements that the term relative adrenal failure was based on is most likely secondary to substantially increased cortisol distribution volume, which is due to low cortisol plasma binding found in septic shock

Approach to Diagnosis

• Determine whether adrenal insufficiency is present during initial assessment; if so, determine whether it is due to an absolute/intrinsic cause or secondary to patient’s prolonged critically ill state
  • Signs and symptoms of adrenal failure are nonspecific in the context of intensive care and partially resemble those of septic shock
• Suggestive findings may include impaired consciousness, hypovolemia and hemodynamic instability, abdominal pain, or ileus
  • However, these may be absent or may be caused by something else in the context of critical illness^3,19
• Nonspecific laboratory findings may include:
  • Eosinophilia
  • Hyponatremia
  • Hypercalcemia
  • Anemia
  • Hypoglycemia

Absolute/Intrinsic Adrenal Insufficiency

• Suspect in patients who are critically ill in the following circumstances:
  • Recent use of supraphysiologic glucocorticoids
  • Anticoagulant use and patient history of sudden flank or back pain
  • Hypotension
  • Unexplained hyponatremia, anemia, and/or hypoglycemia
  • Signs and symptoms related to latent or previously undiagnosed primary disease (eg, extreme fatigue, hyperpigmentation, anorexia/weight loss, low blood pressure/syncopal episodes, unexplained nausea/vomiting or diarrhea)
• Laboratory findings diagnostic of absolute/intrinsic adrenal insufficiency
  • Primary adrenal insufficiency: low plasma cortisol, high plasma ACTH
  • Secondary adrenal insufficiency: low plasma cortisol, low plasma ACTH
  • Tertiary adrenal insufficiency: low plasma cortisol, low plasma ACTH

Critical Illness–Related Corticosteroid Insufficiency

• Suspect in patients in the ICU for a prolonged period who present with hemodynamic instability that persists despite adequate fluid resuscitation and vasopressor treatment
• No formal, widely accepted diagnostic criteria exist for critical illness–related corticosteroid insufficiency
• Laboratory diagnosis of critical illness–related corticosteroid insufficiency is complicated because of the many confounders that invalidate simple tests (eg, ACTH stimulation tests, random cortisol value)
  • ACTH stimulation test that assesses total incremental cortisol response to 250 mcg synthetic ACTH (cosyntropin) is not valid for diagnosis of adrenal failure in the context of critical illness because it is confounded by increased cortisol distribution volume^10,11,20
  • Assays used to quantify cortisol in clinical laboratories give results that are highly variable and not comparable^21,22
    • Therefore, it is not possible to identify a specific cut-off level of free cortisol below which systemic cortisol availability would be insufficient
• When reviewed together and considered within the context and duration of the critical illness, results of the following may aid diagnosis:
  • Plasma ACTH and free cortisol levels measured over time in the ICU
  • Documentation of low incremental free cortisol response to 250 mcg synthetic ACTH (cosyntropin)
  • Documentation of low ACTH response to CRH (corticotropin-releasing hormone) stimulation test¹²

History

• Assess whether patients received chronic treatment with corticosteroids
  • Unintended withdrawal is the most common cause of central adrenal insufficiency and acute adrenal crisis in ICU patients^17,19
• Check whether patient receives a drug with potential impairment of the hypothalamus-pituitary-adrenal as an adverse effect¹⁹
  • Medications of importance include:¹⁹
    • Chronic or high-dose opioids
    • Ketoconazole
    • Fluconazole
    • Megestrol acetate
    • Aminoglutethimide
    • Trilostane
    • Etomidate
    • Phenobarbital
    • Phenytoin
    • Rifampin
    • Anticoagulants
    • Tyrosine kinase inhibitors
    • Mifepristone
    • Mitotane
• Assess for (recent or remote) history of cerebral trauma or other injury to the brain, which can be relevant as a risk factor for central adrenal insufficiency²³ and may be unmasked in the setting of surgery, trauma, infection, or other critical illness
• Assess for symptoms indicative of preexisting or latent adrenal insufficiency
  • Unexplained lethargy, weight loss, and abdominal complaints
  • In a critically ill patient with signs and symptoms of distributive shock, latent adrenal insufficiency may have triggered shock upon stress evoked by surgery, trauma, infection, or other critical illness¹⁵

Physical Examination

• In the context of critical care, signs and symptoms of adrenal failure are nonspecific and often resemble those of septic shock; these include:
  • Elevated or suppressed cardiac output
  • Tachycardia
  • Hypotension
  • Large pulse pressure
  • Warm extremities with brisk capillary refill
  • Peripheral edema
  • Oliguria

Laboratory Tests

• ACTH stimulation test that assesses total incremental cortisol response to 250 mcg synthetic ACTH (cosyntropin) is not valid for diagnosis of adrenal failure in the context of critical illness, and it is unclear whether a cut-off level of (estimated or measured) free cortisol (below which systemic cortisol availability would be insufficient) can be identified¹⁰
• Given lack of appropriate tests to rule out either absolute/intrinsic adrenal insufficiency or critical illness–related corticosteroid insufficiency, either watchful waiting or initiating substitution therapy can be advocated
• Results of repeated plasma ACTH and free cortisol measurements (possibly combined with documenting incremental free cortisol response to 250 mcg synthetic ACTH [cosyntropin] and/or low ACTH response to CRH stimulation test¹²) can aid diagnosis when quantified together and reviewed within the context and duration of the critical illness

Imaging Studies

• Adrenal Imaging
  • Abdominal imaging can be undertaken to evaluate for major adrenal hemorrhage or infiltration via MRI, CT, or ultrasonography²⁴

Approach to Treatment

• Upon suspicion of adrenal insufficiency, attempt to reduce or stop drugs that are known to suppress adrenal function, which also include opioids in doses often used for analgesia in the ICU (Figure 1)^17,19

Absolute/Intrinsic Adrenal Insufficiency

• Adrenal crisis due to absolute/intrinsic adrenal failure requires immediate treatment
  • Treatment includes IV hydrocortisone as well as IV fluid resuscitation (isotonic saline to restore intravascular volume and replace urinary salt losses; dextrose infusion may also be added to prevent hypoglycemia)
  • Administer hydrocortisone 100 mg IV bolus, followed by 200 mg IV over 24 hours as a continuous infusion or in divided doses every 6 hours²⁵
  • On day 2 reduce hydrocortisone to 100 mg/day²⁵
  • Thereafter, taper to maintenance dose of approximately 60 mg/day while still requiring intensive care and approximately 15 to 20 mg/day upon full recovery^8,25

Critical Illness–Related Corticosteroid Insufficiency

• Treatment of the entity of critical illness–related corticosteroid insufficiency is controversial
  • Historically, the dose of hydrocortisone that has been recommended in the literature is 200 mg/day, as this was the dose used in large randomized controlled trials investigating the impact of corticosteroids in patients suffering from acute septic shock^5,6,7,26
    • Studies have shown that this generates cortisol plasma levels that are several-fold higher than those of critically ill patients without suspicion of adrenal failure²⁷
      • This is because of suppressed expression and activity of cortisol-metabolizing enzymes in the liver and kidney in critically ill patients, which causes suppressed cortisol breakdown
      • Therefore, a dose of 200 mg/day or higher of hydrocortisone should be considered a pharmacologic treatment rather than substitution for suspected adrenal insufficiency
  • This author suggests a daily dose of approximately 60 mg/day, which is the equivalent of the daily production quantified in ICU patients with use of tracer technology and which therefore likely suffices as a substitution dose for treatment of any form of suspected adrenal insufficiency in this context^8,9,19
  • When turnaround times for measurements of free cortisol or corticosteroid-binding globulin are too long, treatment can be initiated and stopped again as soon as possible after a negative test result or lack of clinical response
  • When quantification of ACTH and/or cortisol does not allow a conclusive diagnosis, it may be justified to watchfully wait or initiate a brief treatment course with approximately 60 mg hydrocortisone per day while monitoring clinical response
• Patients with any type of adrenal failure respond spectacularly and immediately to hydrocortisone with fast shock reversal and the ability to taper hydrocortisone to maintenance dose quickly
  • Response to pharmacologic doses of hydrocortisone in acute septic shock is usually much more modest and slower to set in
• Per the 2021 Surviving Sepsis Guidelines, fluid- and vasopressor-resistant hemodynamic instability in patients suffering from acute septic shock (without evidence of preexisting hypothalamus-pituitary-adrenal axis pathology and without signs of acute hemorrhage in or infiltration of the adrenal gland) may be considered an indication for brief treatment with 200 mg hydrocortisone per day for shock reversal without testing adrenal function²⁸
  • Such intervention represents a pharmacologic treatment rather than a substitution for a suspected failing adrenal cortex

Drug Therapy

• Hydrocortisone is the corticosteroid formulation of choice for suspected adrenal insufficiency in critically ill patients
• This author recommends using 2- to 3-fold the normal healthy substitution dose of hydrocortisone (approximately 60 mg/day) for suspected adrenal insufficiency in critically ill patients
• Alternatively, a brief course of pharmacologic doses of hydrocortisone (200 mg/day) for shock reversal can be used for acute septic shock that is resistant to fluids and vasopressors
  • This dose theoretically also results in full occupancy of the mineralocorticoid receptors and thus addition of fludrocortisone is not required (although 2 studies have used that combination^5,26)

• Table 3. Medications for use in adrenal insufficiency during critical illness in the ICU.

  ESICM, European Society of Intensive Care Medicine; GT, gastric tube; SCCM, Society of Critical Care Medicine.

  Medication	Dosing	Indication	Comments
  Hydrocortisone	IV 60 mg/day Either as continuous infusion or as bolus of 40 mg AM and 20 mg PM to mimic diurnal rhythm	Substitution dose for suspected adrenal insufficiency in critically ill adults	Recommended by this author
  Hydrocortisone	IV 200 mg/day Dose: 4 x 50 mg bolus injections or as continuous infusion; typically, 5 to 7 days in duration with taper guided by clinical response	Pharmacologic dose for shock reversal in septic shock states that are refractory to fluid and vasopressor support	In agreement with SCCM/ESICM guidelines2,28
  Fludrocortisone	50 mcg via GT once a day	May be combined with 200 mg hydrocortisone per day in acute septic shock	2 trials showed mortality benefit used the combination of hydrocortisone with fludrocortisone5,26

Monitoring

• Critically ill patients who received hydrocortisone treatment in the ICU for suspicion of adrenal insufficiency should be reinvestigated by an endocrinologist after recovery to allow accurate and nonconfounded diagnosis

Complications

• Short-term complications related to hydrocortisone therapy, in doses higher than substitution dose, may include:
  • Hyperglycemia
  • Immune suppression and secondary infections
  • Critical illness myopathy and polyneuropathy
  • Peptic ulcer disease
  • Pancreatitis
  • Delirium
• Long-term adverse effects as a result of treatment of critically ill patients with corticosteroids, in particular when doses are used that exceed the equivalent of 60 mg of hydrocortisone per day for longer than a few days, may include:
  • Persisting weakness
  • Neurocognitive impairment

Prognosis

• Data are scarce. In 1 study, reactivation of the hypothalamus-pituitary-adrenal axis, assessed 1 week after ICU discharge, was documented for most surviving long-term ICU patients¹⁰

Referral

• Follow-up by endocrinologist after hospital discharge for all patients treated with hydrocortisone while in ICU for suspicion of adrenal insufficiency