Critical illness associated with hypotension refractory to intravenous fluid administration. Rule out adrenal insufficiency.
Adrenal insufficiency is likely.
Continue hydrocortisone 50 mg IV Q 6 hours with or without fludrocortisone 50 àg
enterally once daily.
Continue for 7 days, regardless of clinical
response.
Administer 250 àg cosyntropin intravenously and measure total cortisol
prior to and 60 minutes post- administration to calculate (Δ cortisol = baseline − 60 min value)
Adrenal insufficiency is
unlikely.
Corticosteroid therapy unlikely to help unless indicated for ARDS treatment
(see Chapter 10), but may consider continuing hydrocortisone if clinical
response (decrease in vasopressor support) within
48 hours. Consider performing metyrapone test if
diagnosis still in doubt.***
Random cortisol <10–15 àg/dL OR cortisol ≤9 àg/dL Some experts use random
cortisol <25 àg/dL without cosyntropin testing.
Random cortisol ≥10–15 àg/dL AND cortisol >9 àg/dL Some experts use random
cortisol ≥25 àg/dL without cosyntropin testing.
Administer hydrocortisone 50 mg IV while
awaiting results of cortisol testing**
Consider administering one dose of dexamethasone* 2 mg IV
ARDS, acute respiratory distress syndrome;cortisol, change in cortisol level inμg/dL.
∗Dexamethasone does not interfere with the cosyntropin test and 2 mg IV is equivalent to approximately 50 mg of hydrocortisone.
∗∗Administer only if dexamethasone was not given initially.
∗∗∗Metyrapone blocks the conversion of 11-deoxycortisol to cortisol by CYP11B1 and causes a rapid fall in cortisol.
To test, give metyrapone orally at 30 mg/kg at midnight and measure 11-deoxycortisol and cortisol 8 hours later.
A normal response is an 8AMserum 11-deoxycortisol level of 7–22μg/dL and serum cortisol<5μg/dL. Serum 11-deoxycortisol<7μg/dL indicates adrenal insufficiency. Patients who undergo metyrapone testing should receive at least 24 hours of corticosteroid replacement.Metyrapone is not currently available for routine clinic use in the United States.
Randomized studies have been performed to determine if corticosteroid replace- ment therapy is beneficial in patients with varying diagnostic definitions of adrenal insufficiency. In a placebo-controlled, randomized, double-blind study by Annane et al.
in 2002, 300 patients with septic shock were randomized to receive hydrocortisone
LWBK976-28 LWW-Kollef-educational September 14, 2011 20:59
2 3 6 E N D O C R I N E D I S O R D E R S
50 mg every 6 hours and fludrocortisone 50μg once daily for 7 days, or placebo, after undergoing a 250μg cosyntropin test. Of the 229 patients with relative adrenal insuf- ficiency (115 placebo and 114 corticosteroid), defined as an increase after cosyntropin
of≤9μg/dL, there was significantly reduced mortality (73 patients vs. 60 patients,
p =0.02) and withdrawal of vasopressor therapy within 28 days (46 patients vs.
65 patients, p= 0.001) in the corticosteroid treated patients, without increasing adverse events. In the 70 patients who had a cosyntropin response≥9μg/dL, cor- ticosteroid replacement therapy had no significant effect on the same outcomes and no trend toward efficacy. These data were supported by two meta-analyses examin- ing glucocorticoids administration in severe sepsis and septic shock showing reduced mortality when glucocorticoids were given at similar dosages. Previous studies show- ing a survival disadvantage used higher-dose corticosteroids (23,975 mg vs. 1209 mg) administered for a shorter duration.
However, the preliminary results of the randomized arm of the CORTICUS study found no benefit to treating patients with septic shock, including the subgroup with a cosyntropin response≤9μg/dL. This study differed from the initial randomized study by Annane in 2002 in that patients in the CORTICUS study were randomized up to 72 hours after the onset of septic shock rather than within 8 hours, patients in the CORTICUS study included those with severe sepsis and septic shock rather than septic shock alone (i.e., could be included in the study without hypotension requiring vasopressors), received 11 days of corticosteroid replacement rather than 7 days, and were not given fludrocortisone. Additionally, the CORTICUS study lacked clinical equipoise resulting in a selection bias whereby patients least likely to benefit from corticosteroids were enrolled in the study. How this study will change clinicians’
approach to treating adrenal insufficiency in septic shock has yet to be seen.
Glucocorticoids, however, may not be appropriate for all critically ill patients as highlighted in a case-control study by Britt et al. in 2006 in 100 patients (six with septic shock) in a burn trauma ICU who received steroids, compared with 100 matched control patients, showing that corticosteroid use was associated with increased infection rates, increased ICU and ventilator duration, and a trend toward increased mortality.
Based on these data, there are no definitive diagnostic criteria for relative corti- costeroid insufficiency in critically ill patients. Some experts believe the supraphysi- ologic dose of cosyntropin used in an attempt to increase cortisol production from an already stimulated adrenal gland may be less important than the baseline cortisol alone. However, numerous studies have found benefit from corticosteroid replace- ment in patients with relative adrenal insufficiency defined by cosyntropin stimulated cortisol≤9μg/dL, and there is suggestion, at least in retrospective data, that base- line cortisol levels>10 to 15μg/dL may be less important than adrenal reserves as measured with 250μg cosyntropin stimulated serum cortisol levels at 1 hour.
Thus, although our understanding of relative adrenal insufficiency in the critically ill continues to evolve, nearly all evidence agrees that adrenal insufficiency is likely when the baseline cortisol is<10 to 15μg/dL in critically ill patients, and that cosyntropin stimulation testing may have additional diagnostic value in identifying which patients are likely to respond to corticosteroid replacement, at least in the setting of volume unresponsive hypotension in patients with septic shock. We recommend that critically ill patients with shock unresponsive to IV fluids requiring vasopressor support be evaluated for adrenal insufficiency and treated as illustrated in Algorithm 28.1. Patients may be given dexamethasone 2 mg IV while waiting for cosyntropin testing results, as
Endocrine Disorders rAdrenal Insufficiency in Critical Illness 2 3 7
dexamethasone does not interfere with the cosyntropin test. Otherwise, patients should be started on hydrocortisone immediately following cosyntropin testing while waiting for the results. Once treatment is initiated, the cause of adrenal insufficiency can be explored further. Many patients with adrenal insufficiency related to critical illness can be expected to regain normal function of the HPA axis with recovery from their illness. Some experts, however, still use the random cortisol alone, feeling that adrenal insufficiency is unlikely when the random cortisol is≥25μg/dL and that cosyntropin testing has no additional diagnostic value. Certainly more studies are needed.
S U G G E S T E D R E A D I N G S
Annane A, Bellissant E, Bollaert PE, et al. Corticosteroids for severe sepsis and septic shock:
a systematic review and meta-analysis.BMJ.2004;329:480–489.
Meta-analysis of 16 randomized and quasi randomized trials involving 2063 patients with severe sepsis and septic shock showing that long courses (≥5 days) with low dose corticosteroids (≤300 mg hydrocortisone or equivalent) reduced 28 day and hospital mortality.
Annane D, Maxime V, Ibrahim F, et al. Diagnosis of adrenal insufficiency in severe sepsis and septic shock.Am J Respir Crit Care Med.2006;174:1319–1326.
An update on the diagnosis of adrenal insufficiency using cosyntropin stimulation.
Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low dose of hydrocorti- sone and fludrocortisone on mortality in patients with septic shock.JAMA.2002;288:
862–871.
A placebo controlled, randomized, double-blind study showing a 7-day treatment with low doses of hydrocortisone and fludrocortisone in patients with septic shock and relative adrenal insufficiency, significantly reduced the risk of death without an increase in adverse events.
Annane D, Sebille V, Troche G, et al. A 3-level prognostic classification in septic shock based on cortisol levels and cortisol response to corticotropin.JAMA.2000;283:1038–1045.
A prospective inception cohort study evaluating the prognostic value of cortisol and the cosyntropin stimulation test in patients with septic shock.
Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients.N Engl J Med.
2003;348:727–734.
A focused review on the pathogenesis and treatment of adrenal insufficiency in the critically ill.
Lipiner-Friedman D, Sprung CL, Laterre PF. Adrenal function in sepsis: The retrospective Cor- ticus cohort study.Crit Care Med.2007;35:1012–1018.
Retrospective cohort study from 20 European intensive care units examining the relationship between baseline and cosyntropin stimulated cortisol levels and mortality in patients with severe sepsis and septic shock showing that delta cortisol and not basal cortisol levels were associated with clinical outcomes.
Marik PE. Critical illness-related corticosteroid insufficiency.Chest.2009;135:181–193.
Concise review of the causes, pathophysiology, diagnosis and treatment of adrenal insufficiency in critically-ill patients.
Marik PE, Pastores SM, Annane D, 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.
Evidence-based guidelines for the diagnosis and medical management of adrenal insufficiency in critically ill patients. This guideline recommends that adrenal insufficiency in critical illness is best diagnosed by a delta cortisol (after 250 mcg cosyntropin) of<9 mcg/dL or a random total cortisol of<10 mcg/dL. Treatment with hydrocortisone (200–300 mg/
day) is recommended for patients with septic shock who have responded poorly to fluid resuscitation and vasopressor agents.
LWBK976-28 LWW-Kollef-educational September 14, 2011 20:59
2 3 8 E N D O C R I N E D I S O R D E R S
Sprung CL, Annane D, Keh D, et al. Hydrocortisone therapy for patients with septic shock.
N Engl J Med.2008;358:111–124.
The prospective CORTICUS study which demonstrated that hydrocortisone did not improve survival or reversal of shock in patients with septic shock, either overall or in patients who did not have a response to corticotropin, although hydrocortisone hastened reversal of shock in patients in whom shock was reversed.
29 Diabetic Ketoacidosis and Hyperosmolar Hyperglycemic State
David A. Rometo, Marin H. Kollef, and Garry S. Tobin
Diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state (HHS) are life- threatening hyperglycemic complications of diabetes mellitus (DM), and common reasons for admission to the intensive care unit. The annual incidence of DKA ranges from 4.6 to 8 episodes per 1,000 patients with DM, resulting in more than 500,000 hospital days per year. The annual incidence of HHS is lower than DKA and accounts
for<1% of primary diabetic admissions. The mortality rate is 1% to 5% in DKA and
5% to 20% in HHS, with worse outcomes at extremes of age, and with the presence of coma, hypotension, and severe comorbidities. Precipitating factors include inadequate insulin treatment or noncompliance, new-onset DM, infections (most commonly pneumonia and urinary tract infections), cardiovascular events, cerebrovascular acci- dent, pancreatitis, drugs, and pregnancy. DKA typically occurs in patients with type 1 DM, but does occur in patients with type 2 DM (as high as 1/3 of DKA cases). HHS is typically confined to patients with type 2 DM, often precipitated by loss of access to water in the elderly or chronically ill. DKA and HHS are the result of insulin deficiency in patients with DM. In both disorders, insulin deficiency causes increased hepatic glycolysis, gluconeogenesis, and impaired glucose utilization by peripheral tissues, lead- ing to hyperglycemia. The hyperglycemia in both disorders causes an osmotic diuresis.
Table 29.1 highlights the differences between DKA and HHS.
D I A B E T I C K E T O A C I D O S I S
DKA is characterized by hyperglycemia (blood glucose [BG] typically≥250 mg/dL), anion gap metabolic acidosis (arterial pH≤7.3), and ketosis (positive urine and plasma ketones), along with dehydration and electrolyte abnormalities in varying degrees.
Common ketone assays use nitroprusside, measuring acetoacetate and acetone, but not beta-hydroxybutyrate. Prominent presenting symptoms include nausea/vomiting, abdominal pain, labored breathing (Kussmaul respirations), and polyuria. A mixed acid–base disorder may also be present, such as a concomitant severe contraction metabolic alkalosis elevating the serum bicarbonate level, masking the underlying metabolic acidosis.
In patients with DKA, the absolute lack of insulin causes an increase in coun- terregulatory hormones (cortisol, growth hormone, catecholamines, and glucagon),
2 3 9
LWBK976-29 LWW-Kollef-educational September 21, 2011 13:10
2 4 0 E N D O C R I N E D I S O R D E R S
TABLE 29.1 Initial Laboratory Values in DKA and HHS DKA
Value Mild Moderate Severe HHS
Plasma glucose (mg/dL)
>250 >250 >250 >600
Arterial pH 7.25–7.30 7–7.24 <7 >7.30
Serum bicarbonate (mEq/L)
15–18 10–14 <10 >15
Urine and serum ketones
Positive Positive Positive Trace/small Serum osmolarity
(mOsm/L)
<320 <320 <320 330–380
Anion gap >10 >12 >12 <12
Mental status Alert Alert/drowsy Stupor/coma Stupor/coma
Sodium (mmol/L) 125–135 125–135 125–135 135–145
Potassium (mmol/L) Normal to↑ Normal to↑ Normal to↑ Normal Creatinine (mg/dL) Slight↑ Slight↑ Slight↑ Moderate↑ DKA, diabetic ketoacidosis; HHS, hyperosmolar hyperglycemic state.
which promote lipolysis in adipose tissue and the release of free fatty acids. In the liver, the free fatty acids are converted to ketones. These patients suffer from a metabolic acidosis as a result of these circulating ketoacids. Ketone bodies contribute to the osmotic diuresis, and diuresis causes loss of sodium and potassium. Although initial laboratory values are variable, total body sodium and potassium are depleted.
While prompt management of DKA as described below is essential, finding and treating the precipitating cause should not be forgotten or delayed. Blood, urine, sputum cultures, chest x-ray, EKG, and empiric treatment based on clinical suspicion should be part of the initial management of DKA. One should evaluate the cause of abdominal pain if the symptom does not resolve with correction of dehydration and metabolic acidosis.
Treatment of DKA requires reversal of the hyperglycemia by administering insulin, and replacing the circulating and total body volume and electrolyte deficits (as outlined in Algorithm 29.1). Normal saline boluses for acute volume resuscitation, followed by1/2NS replacement of the remaining fluid deficit, and later 5% dextrose containing fluids when glucose falls below 250 mg/dL is the mainstay of fluid ther- apy. Insulin should be given as an intravenous (IV) bolus, followed by continuous infusion. Hyperglycemia resolves faster than acidosis, and supraphysiologic insulin is needed to overcome hyperglycemia-induced insulin resistance to achieve euglycemia.
After this is achieved, insulin is still required to drive peripheral ketone use, resolving the remaining acidosis. Therefore, BG should be kept between 150 and 200 mg/dL with 5% dextrose until the anion gap is closed. Intermediate or long-acting subcuta- neous (SC) insulin must be given before insulin drip discontinuation. Failure to do this properly can lead to rebound hyperglycemia and DKA. The patient should be able to tolerate PO fluids, and dextrose containing IV fluids should be stopped at that time.
Endocrine Disorders rDiabetic Ketoacidosis and Hyperosmolar Hyperglycemic State 2 4 1