Definition

Steroids are the mainstay of therapy for a variety of disorders requiring immune suppression. These include dermatologic, pulmonary, and autoimmune disease as well as prevention of rejection in organ transplantation. It is also the primary replacement therapy for those with inadequate cortisol production due to a pituitary or adrenal etiology. In the perioperative setting the incidence of acute adrenal insufficiency is estimated to be between 0.01% and 0.1%. The most likely trigger of an Addisonian crisis in this circumstance is insufficient cortisol levels to meet the increased glucocorticoid demands of trauma, infection, and surgical stress. Following surgery, peak plasma cortisol concentrations are achieved 4 to 5 hours postoperatively and may remain elevated for 48 to 72 hours, especially after major procedures. In a normal stress response, minor surgery induces less than 50 mg of cortisol production during the first 24 hours whereas major surgery produces 75 to 100 mg of cortisol in the same period. With maximal stress (e.g., septic shock or major trauma), the adrenals may produce as much as 300 to 500 mg of cortisol per day.

Adrenal insufficiency is classified as primary, secondary, or tertiary, based on the anatomic level of impairment within the hypothalamic-pituitary-adrenal (HPA) axis. With primary adrenal insufficiency, the abnormality is in the adrenal gland. More than 90% of the adrenal gland must be destroyed before symptoms of glucocorticoid and mineralocorticoid deficiency are evident. The most common cause of primary adrenal insufficiency in the United States is autoimmune in nature. In secondary adrenal insufficiency, the abnormality is at the level of the pituitary gland. Such patients show symptoms of glucocorticoid deficiency but usually have intact mineralocorticoid function given the additional influence of angiotensin II on aldosterone secretion. Tertiary adrenal insufficiency is the most prevalent type in the perioperative setting and is caused by suppression of the HPA axis by chronic exogenous steroid use. This is the result of increased plasma cortisol levels secondary to glucocorticoid administration and its negative feedback effect on the hypothalamus. The reduction in corticotropin-releasing hormone reduces pituitary adrenocorticotropic hormone (ACTH) secretion and hence production of endogenous glucocorticoids (and to a lesser degree mineralocorticoids) by the adrenals. The return of functionality to the HPA axis is variable both clinically and biochemically.

In addition to the potential risk of impairment of the HPA axis, chronic steroid ingestion can create a host of other clinical problems in the perioperative period. Friable skin can present a problem with difficult intravenous access and skin breakdown. These patients are also at increased risk for infection, making aseptic technique particularly important. Last, they may have occult gastrointestinal bleeding or ulcer disease that is unrecognized.

Recognition

The clinical presentation of an Addisonian crisis varies from mild, nonspecific constitutional symptoms to the presence of profound shock unresponsive to vasopressor therapy. In the perioperative period, clinicians should maintain a high index of suspicion when caring for patients chronically taking or recently discontinuing steroid therapy. Mild symptoms of adrenal insufficiency include nausea, vomiting, anorexia, fatigue, and abdominal pain. Mineralocorticoid deficiency infrequently accompanies an Addisonian crisis, yet when it occurs it can present with hyponatremia, hyperkalemia, and metabolic acidosis. With severe adrenal insufficiency, arterial hypotension with postural accentuation is common. Such hypotension may be refractory to fluid and vasopressor therapy given glucocorticoids’ profound effect on vasomotor sensitivity to catecholamines. This clinical presentation is highly unlikely in patients who have received their usual daily dose of steroids, especially if they are not experiencing high levels of physiologic stress.

Risk Assessment

The following are important risk factors for the development of adrenal insufficiency in the perioperative period:

• •
  

  Daily dose of steroids (>20 mg/day of prednisone or equivalent)

  
  
• •
  

  Duration of steroid therapy (>3 weeks)

  
  
• •
  

  Suppression of the HPA axis based on serum testing (cortisol level, ACTH stimulation test)

  
  
• •
  

  High degree of physiologic stress (surgery, trauma, burns, infection)

  
  
• •
  

  Primary or secondary adrenal insufficiency requiring steroid replacement therapy

It has been shown that the total daily dose of steroids determines the responsiveness of the HPA axis to stress. LaRochelle and colleagues demonstrated that when the total daily dose of prednisone was 5 mg or less, there was a normal response to the ACTH stimulation test. With doses greater than 5 mg/day, responses to the ACTH test can vary widely. Others have suggested that doses equivalent to 20 mg/day of prednisone may also leave the HPA axis unsuppressed if administration is limited to less than 3 weeks and discontinued for greater than 14 days. For those receiving higher doses of steroid therapy (>20 mg prednisone) and/or for a longer duration (>3 weeks), the HPA axis should be considered impaired for up to 6 months after therapy has been discontinued. When in doubt, a two-tiered approach is recommended. If time permits, morning cortisol can be tested with the patient withdrawn from steroids for 24 hours. If the cortisol level is less than 5 μg/dL, the axis is impaired and supplemental steroid dosing is needed based on the anticipated degree of surgical stress. A morning cortisol of greater than 10 μg/dL connotes an intact HPA axis, and no additional dosing is needed except the patient’s chronic dosing. A result between 5 and 10 μg/dL is indeterminate and requires that an ACTH simulation test be performed, if feasible, to determine whether the HPA axis has returned to normal.

In the case of surgical patients with concomitant major trauma, burns, or sepsis, random serum cortisol levels can be used to determine whether the physiologic response to major surgery and critical illness is adequate. A review by Marik and Zaloga supports judicious use of supplemental steroids in these critically ill patients using a 25 μg/dL serum cortisol level as a cutoff for treatment. When the random serum cortisol level is found to be low, particularly in patients with sepsis, stress doses of steroids are often administered in the intensive care unit setting.

Implications

Addisonian crisis secondary to inadequate steroid supplementation is rare, but vigilance by the anesthesiologist for subtle signs of adrenal insufficiency is important to avoid complications. The cumulative daily dose, the duration of chronic steroid therapy, and the nature of surgery are important factors for predicting the integrity and responsiveness of the HPA axis. A preoperative Cushingoid appearance is indicative of excessive steroid use and should be an alert for probable HPA axis suppression. Recent exposure to etomidate use as an induction agent also puts the patient at higher risk for HPA axis suppression and should be considered when postoperative hypotension is unresponsive to conventional treatments. The anesthesiologist should consider avoiding etomidate administration in those patients at higher risk for HPA axis abnormalities because of its known ability to inhibit steroid synthesis. Topical and inhaled steroids generally present little risk of adrenal insufficiency, but higher doses over prolonged periods of time require special attention (e.g., >2 g/day topically of superpotent agents), as do three or more intraarticular or epidural steroid injections within 3 months of surgery.

Surgical patients with accompanying septic shock, major trauma, or burns represent a special group. Work by Annane studying septic shock patients has shown improved clinical outcomes when treating with supraphysiologic doses of steroids. Given the potential risk of high-dose steroids (hyperglycemia, hypertension, fluid retention, increased infection risk), dosing should be based on measured random cortisol levels. Finally, patients with primary or secondary adrenal insufficiency are also at high risk for perioperative exacerbation due to their inability to increase cortisol secretion in times of physiologic stress. Therefore it is prudent that they receive supplemental dosing in addition to their chronic therapy.