Steven R. Allen
Heidi L. Frankel
Presentation
A 67-year-old female was involved in a motor vehicle crash 7 days ago. Her injuries included a closed head injury consisting of a subarachnoid hemorrhage and bilateral tibia/fibula fractures. She was able to be extubated at postinjury day 3; however, she required reintubation 24 hours previously (using rapid sequence methodology employing etomidate and succinylcholine) for increasing respiratory difficulty. Over the past 24 hours, she has become hemodynamically unstable with progressive tachycardia and hypotension despite adequate fluid.
Differential Diagnosis
A broad differential diagnosis is pertinent to identify and treat all potential etiologies leading to her hemodynamic instability. Due to the patient’s intubation and underlying pulmonary disease, ventilator-associated pneumonia leading to sepsis and septic shock must be highly considered. Other pulmonary causes, such as a pulmonary embolism (PE), may also contribute to this clinical picture. Additionally, with her age and multiple comorbidities, one must consider an acute cardiac event, such as a myocardial infarction (MI). Her orthopedic injuries also predispose her to fat emboli. Adrenal insufficiency (AI) must be considered as she was intubated with etomidate, which has been found to cause AI, although less likely after a bolus dose.
Presentation Continued
Her past medical history is significant for chronic obstructive pulmonary disease, rheumatoid arthritis for which she takes 15 mg of prednisone daily, hypertension, and diabetes. Her vital signs demonstrate a temperature of 38.6°C, a heart rate of 130, and blood pressure of 85/40. Her blood pressure was responsive but dependent on continuous Levophed administration. Her respiratory rate is 25 on the ventilator with a FiO2 of 0.6, PEEP of 7.5, and pressure support of 10 to 12. Her current oxygen saturation is 92%. She has received three 1-L boluses of crystalloid to affect her hemodynamic status with no improvement. Her current central venous pressure is 10 and her urine output is only 10 mL per hour. On physical exam, she is an obese woman who appears her stated age. Neurologically, she is arousable but does not consistently follow commands. Her cardiac exam demonstrates a regular rhythm but tachycardic.
On further workup, blood cultures were consistently negative. Her white blood count was within normal limits of 11,000 and hemoglobin was stable at 9.7. The electrolyte panel demonstrated a sodium of 131 and potassium of 5.3. The CT scan of the chest was negative for a PE and her echocardiogram demonstrated an ejection fraction of 65% with no obvious wall motion abnormalities and cardiac enzymes were not elevated.
Diagnosis and Treatment
In the face of hemodynamic instability unresponsive to fluid resuscitation and dependent on vasopressors and having ruled out other etiologies including sepsis, PE, and acute MI, AI is the one likely diagnosis that remains. AI seems relevant in light of her chronic steroid use and severe stress from injury. Other signs that suggest the diagnosis of AI include persistent, unexplained fever, weakness, and the inability to wean the ventilator support as well as hyponatremia and hyperkalemia. Other tests that may point toward the diagnosis of AI include mild eosinophilia with mean eosinophil counts of 3.5% versus 0.9% in those with normal adrenal function. These laboratory abnormalities are, however, more likely in the face of chronic AI.
Adrenal function may be assessed by several tests although none are considered to be extremely reliable. The random cortisol level may be helpful. Cortisol is normally secreted in a diurnal cycle. However, this diurnal variation is often lost in the critically ill patient. Cortisol may be checked at any time in the critically ill patient for this reason. Within the literature, many values have been proposed as the appropriate minimum value (range, 10 to 34 μg/dL); however, many would agree that a random cortisol over 18 is a normal response to stress.
The adrenocorticotropic hormone (ACTH) stimulation test may also be diagnostic and has been used over the past four decades. This test is conducted by administering 250 μg of ACTH (cosyntropin) either intravenously or intramuscularly. Cortisol levels are measured before and then 30 and 60 minutes after administration of the cosyntropin. This test may be performed at any time of the day. Some have proposed a low-dose version of the ACTH stimulation test where only 1 µg of cosyntropin is administered intravenously. Due to the low dose, it is thought that it is more sensitive for partial AI. This has not been proven and the preferred test is the standard dose of 250 μg.
The “delta 9” may be helpful in making the diagnosis of AI. In a multicenter, randomized trial, Annane demonstrated that those patients who showed a change in baseline cortisol levels by 9 μg/dL at 30 or 60 minutes during the ACTH stimulation test had lower mortality rates if they received corticosteroids. This test is thought to demonstrate adrenal reserve in the face of critical illness or sepsis but does not assess the integrity of the hypothalamic–pituitary–adrenal axis. It is also argued that those patients that are maximally stressed may be effectively secreting the maximum amount of cortisol. Therefore, while it may be sufficient, the delta value may not be very high and may be <9 μg/dL. The utility of the delta 9 may be limited for this reason. The recent multicenter CORTICUS trial did not report outcome differences between responders and nonresponders to a stimulation test, hence does not advocate for performance of this test.
Free cortisol may also be helpful in identifying AI. More than 90% of cortisol is bound to proteins including cortisol-binding globulin and albumin. Experts would agree that the active portion of cortisol is that which is free, or not bound to proteins. Delayed test results make this test impractical in critically ill patients. More work must be done to assess the true utility of this test and to develop a more clinically relevant test.
Other tests exist but should not be utilized in the critically ill population and include the insulin tolerance test and metyrapone test. The insulin tolerance test may lead to profound hypoglycemia, while the metyrapone test may exacerbate an adrenal crisis.
Presentation Continued
Since the differential diagnosis had been narrowed to include AI without any other obvious cause of the hemodynamic instability, random cortisol level was 10 µg/dL. Additionally, a standard ACTH stimulation test was performed. At 60 minutes after injection, the cortisol level was 17 mg/dL. The change in cortisol levels was less than the proposed 9 mg/dL. It was determined that our patient was indeed suffering from AI.
Treatment of AI
AI may be separated into AI in those who regularly take exogenous corticosteroids for other medical problems such as rheumatoid arthritis or asthma, also known as secondary AI and acute AI. Faced with the effects of AI, one must appropriately supplement the patient with corticosteroids. For those on chronic steroids in the face of stress from critical illness, one should consider increasing the dose of corticosteroids. How long one should supply the elevated dose or whether an elevated dose is even required is not well described in the literature.
Hydrocortisone is the corticosteroid of choice as both prednisone and cortisone require hydroxylation to obtain the active compound of prednisolone and cortisol, respectively. One must consider the replacement of mineralocorticoids as well in AI. This is not necessary if hydrocortisone is used due to the combined activity of both glucocorticoids and mineralocorticoids. However, the major protocol difference between two large trials that had discordant results in patients with septic shock was that the “positive” trial also included use of a mineralocorticoid in addition to administration of hydrocortisone.
The administration of hydrocortisone (150 to 200 mg daily for 5 to 7 days) has been shown to lead to a decreased vasopressor requirement as well as improved organ dysfunction, fewer ventilator days, fewer ICU days, and most importantly lower 28-day mortality. The exact dose is controversial as many studies have demonstrated positive effects with varying dosages ranging from 50 mg every 6 hours to 100 mg every 8 hours with a treatment length of 1 to 5 days.
A 2002 study by Annane et al. demonstrated a 28-day survival benefit in patients with septic shock and AI who received hydrocortisone and fludrocortisone with no difference in adverse events between the study group and the placebo control group. However, one must use caution as a more recent study known as the CORTICUS trial demonstrated no significant difference in mortality between patients who received hydrocortisone versus placebo. Additionally, there was no difference in patients who did not have a response to the ACTH stimulation test compared to those who did respond. While shock was reversed more quickly in those who received hydrocortisone, there was no survival benefit. A recent meta-analysis by Annane confirmed these results.
Case Conclusion
The patient was treated with 50 mg of hydrocortisone every 6 hours for 5 days. Within several hours of her first dose of steroids, her hemodynamic status stabilized and the pressor support was weaned without incident and she was subsequently extubated. She was transitioned from hydrocortisone to oral prednisone as her condition improved and she was able to take medications by mouth. She was later discharged from the intensive care unit in stable condition on her home dose of prednisone.
TAKE HOME POINTS
· AI may present with subtle signs that mimic other clinical etiologies including sepsis, pulmonary embolus, and acute MI.
· One must establish a broad differential diagnosis in order appropriately rule out each of the life-threatening entities.
· Clinical signs may include hypotension, tachycardia, and fever as well as weakness and an inability to wean the patient from the ventilator.
· Laboratory studies that may point toward AI include hyponatremia and hyperkalemia as well as a mild eosinophilia, although these are more common in those with chronic AI.
· The diagnosis of AI may be made by a random cortisol level as well as the results of the standard ACTH stimulation test.
· While the absolute cortisol level may point to the diagnosis, the “delta 9” rule may also help discriminate the actual diagnosis.
· Treatment with hydrocortisone at a dose of 50 to 100 mg every 6 to 8 hours is considered the preferred standard. The length of the treatment course is dependent on the patient’s clinical response.
SUGGESTED READINGS
Annane D, et al. Corticosteroids in the treatment of severe sepsis and septic shock in adults: a systematic review. JAMA. 2009;301:2362–2375.
Annane D, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA. 2002;288:862–871.
Annetta M, et al. Use of corticosteroids in critically ill septic patients: a review of mechanisms of adrenal insufficiency in sepsis and treatment. Curr Drug Targets. 2009;10:887–894.
Cooper MS, Stewart PM. Adrenal insufficiency in critical illness. J Intensive Care Med. 2007;22:348–362.
Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients. N Engl J Med. 2003;348:727–734.
Edwin SB, Walker PL. Controversies surrounding the use of etomidate for rapid sequence intubation in patients with suspected sepsis. Ann Pharmacother. 2010;44: 1307–1313.
Grossman AB. Clinical review#: the diagnosis and management of central hypoadrenalism. J Clin Endocrinol Metab. 2010;95:4855–4863.
Hamrahian A. Adrenal function in critically ill patients: how to test? When to treat? Cleve Clin J Med. 2005;72: 427–432.
Johnson KL, Rn CR. The hypothalamic-pituitary-adrenal axis in critical illness. AACN Clin Issues. 2006;17:39–49.
Marik PE, et al. Recommendations for the diagnosis and management of corticosteroid insufficiency in critically ill adult patients: consensus statements from an international task force by the American College of Critical Care Medicine. Crit Care Med. 2008;36:1937–1949.
Nylen ES, Muller B. Endocrine changes in critical illness. J Intensive Care Med. 2004;19:67–82.
Rivers EP, et al. Adrenal insufficiency in high-risk surgical ICU patients. Chest. 2001;119:889–896.
Sprung CL, et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med. 2008;358:111–124.