Skip to main content

The Hypothalamic-Pituitary-Adrenal Axis in Critical Illness

Critical illness, anesthesia, surgery, trauma, burns, hemorrhage, infection, pain, cold, fever, and emotional disorders are stressors that activate the hypothalamic-pituitary-adrenal (HPA) axis.1,2 The HPA axis is an essential component of the general adaptation to stress and plays a crucial role in cardiovascular, metabolic, and immunologic homeostasis.

As noted in Figure 1, when faced with a stressor, the hypothalamus releases corticotropin-releasing hormone (CRH), which elicits a series of events. The initial event involves the stimulation of the sympathetic nervous system to release norepinephrine and epinephrine, which causes vascular smooth muscle vasoconstriction and enhances inotropic response. Secondly, the posterior pituitary is stimulated to release anti-diuretic hormone (ADH) to promote intravascular fluid retention. Lastly, stimulation of the anterior pituitary leads to the release of corticotropin (ACTH), which stimulates the adrenal cortex to release the mineralcorticoid (aldosterone) to promote intravascular fluid retention and the glucocorticoid (cortisol) to increase availability of substrates for metabolism and modulate the immune/inflammatory response. These responses are beneficial as they augment cardiovascular function, reduce and redirect energy metabolism, and protect against the deleterious biologic effects of the immune response.3
Graphic Figure 1
        

Secretion, Physiologic Actions, and Metabolism of Cortisol

Cortisol is derived from cholesterol in the adrenal cortex. Cortisol secretion has been estimated to be approximately 5.7 mg per square meter of body surface area per day.4 It circulates in the blood in bound and unbound forms. Approximately 90% to 95% of the cortisol in the plasma binds to plasma proteins, primarily cortisol-binding globulin and, to a lesser extent, albumin. This binding protects circulating cortisol from hepatic clearance and gives cortisol a relatively long half-life of 70–90 minutes.5 It is the unbound form, or free cortisol, that is physiologically active.
Serum cortisol levels are reported as micrograms per deciliter (mcg/dL) and nanomoles per liter: 1 mcg/dL is equivalent to 27.6 nanomoles per liter. All concentrations in this review are presented as mcg/dL. Normal serum cortisol concentrations range between 5 to 20 mcg/dL.6 Under conditions of acute stress, cortisol secretion increases. The extent to which serum cortisol levels should rise is under considerable debate. Some advocate that a random serum cortisol level in a severely stressed patient (hypotension, hypoxemia, burn, fever, multiple trauma) should be >25 mcg/dL,2,5,7,8 whereas others advocate it should be >18 mcg/dL 9 or 20 mcg/dL.10 Serum cortisol levels appear to be related to severity and duration of the stressor (Table 1). In the course of critical illness, both extreme hypocortisolemia and hypercortisolemia are associated with increased mortality.19,20 In healthy unstressed individuals, there is a diurnal pattern of cortisol secretion with peak levels between 4 and 8 am and trough levels between 2 and 4 am. However, loss of this diurnal pattern has been noted in critically ill patients with fever, pain, and hypotension.1
Graphic Table 1
Cortisol exerts its effects by passing through cell membranes and binding to receptors in the cytosol of nucleated cells. The steroid-receptor complex migrates to the nucleus and influences gene transcription.
Cortisol is vitally important for cardiovascular reactivity, carbohydrate, protein and fat metabolism, and for its anti-inflammatory effects (Table 2). Cortisol shifts substrate metabolism so that energy is selectively and immediately available to vital organs. Cortisol increases hepatic gluconeogenesis, inhibits glucose uptake in adipose tissue, and increases amino acid release from adipose and body protein stores. Cortisol has a supportive role in maintaining vascular tone and augmenting inotropic actions of catecholamines. It does so by increasing transcription and expression of receptors for catecholamines and angiotensin II. Glucocorticoids are necessary for adequate coupling of adrenergic receptors. High doses of exogenous corticosteriods may result in recoupling of desensitized adrenergic receptors and thus restore responsiveness in blood pressure.21 Cortisol decreases the production of nitric oxide, a potent vasodilator. Cortisol has a profound effect on the inflammatory process as it inhibits the functions of almost every cell involved in inflammation. This inhibition is mediated by altering the transcription of cytokine genes (Interleukins [IL]-1 and -6) and by inhibiting the production of proinflammatory substances (leukotrienes and prostaglandins).22 The net result of this response is to mute the inflammatory cascade and protect against an overresponse.3
Graphic Table 2
Cortisol is metabolized in the liver with subsequent renal excretion of the metabolites. Therefore, liver disease markedly impairs the metabolism of cortisol and renal disease reduces the excretion of the metabolites.
          

Cortisol Response to Critical Illness

During critical illness, the normal cortisol response is altered. There are 2 clinical conditions associated with adrenal dysfunction that may present during critical illness: prolonged elevation of serum cortisol over the course of critical illness and corticosteroid insufficiency in which there is hyposecretion of cortisol. Both conditions are detrimental. The concept of a balance between too much and too little corticosteroid function during critical illness has prompted a quest for the “eucorticoid state.”5
          
Cortisol Secretion During Acute Versus Chronic Critical Illness
Secretion of cortisol appears to have a biphasic pattern during the course of critical illness. The neuroendocrine characteristics of prolonged critical illness (defined as dependent on ICU support for at least 10 days) appear to be quite different from those observed in the first few hours or days of critical illness.3 In the acute phase, both ACTH and cortisol levels are elevated and remain elevated 24 hours a day. This results in depression of the normal circadian rhythm of higher ACTH and cortisol levels in the morning as compared to the evening.5 During the chronic phase of prolonged critical illness, circulating levels of cortisol remain elevated, but ACTH levels are low. Why ACTH levels are low in the presence of high serum cortisol levels in prolonged critical illness is unclear. There may be a loss of the negative feedback mechanism or it may be that cortisol is released from sources outside the adrenal cortex. Endothelin and atrial natiuretic peptide have been implicated as sources of cortisol.12 During inflammation, levels of free cortisol increase as neutrophils liberate cortisol from corticosteroid binding globulin.23 Inflammatory cytokines increase tissue cortisol levels by changing peripheral cortisol metabolism and increasing the affinity of glucocorticoid receptors for cortisol.24,25 There is also evidence that cortisol clearance from the circulation is impaired in many critically ill patients, which results in decreased tissue uptake and metabolism of cortisol.2
The pathophysiologic effects of hypercortisolemia in prolonged critical illness are not fully understood. There appears to be a deranged relationship among catecholamines, adrenergic receptors, and corticosteriods that leads to adrenergic hyporesponsiveness.10 Adrenergic receptors downregulate, which results in vascular smooth muscle hyporesponsiveness and myocardial depression.26,27 Hypercortisolemia is associated with impaired wound healing, neuropathy and myopathy,3 hypertension, hyperglycemia, and other electrolyte abnormalities.28 Prolonged elevation of cortisol may represent a significant risk factor for the development of acute adrenal insufficiency. This is thought to result from a marked stimulation of the HPA axis, which eventually fails to respond to the combination of underlying disease, critical illness, and homeostatic adaptation.10
          
Adrenal Insufficiency
Many diseases and their treatments interfere with the normal corticosteroid response to critical illness and induce adrenal insufficiency. Acute adrenal insufficiency occurs in patients who cannot increase cortisol production during acute stress. The incidence of adrenal insufficiency in critically ill patients approximates 30%, with an incidence as high as 50% to 60% in patients with septic shock.29 Adrenal insufficiency is related to primary or secondary causes.
          
PRIMARY ADRENAL INSUFFICIENCY
Primary adre-nal insufficiency, known as Addison's disease, is a failure of the adrenal gland and results in a deficiency of cortisol and aldosterone (Figure 2A). Causes of primary adrenal insufficiency are summarized in Table 3. Primary adrenal insufficiency can become a life-threatening disorder with any stressful situation since cortisol secretion cannot be increased. Primary adrenal insufficiency is most commonly caused by autoimmune adrenalitis (slow destruction of the adrenal cortex by cytotoxic lymphocytes) and is sometimes accompanied by autoimmune thyroid disease (Grave's or Hashimoto's) and other autoimmune endocrine disorders (autoimmune polyglandular syndromes).30 Human immunodeficiency virus (HIV) infection and other infections in immunocompromised patients (tuberculosis, cytomegalovirus, fungal infections) are currently the most important causes of primary adrenal insufficiency.2 The adrenal gland is the endocrine gland most commonly involved in patients with acquired immune deficiency syndrome (AIDS).31 In addition, a number of drugs used with HIV infection (Ketoconazale, Rifampin, Megestrol acetate) inhibit enzymes involved in cortisol synthesis, enhance hepatic metabolism of cortisol, and suppress the HPA axis.9 Sepsis and its underlying coagulopathy can cause adrenal hemorrhage and induce adrenal insufficiency.
Graphic Figure 2 Graphic Table 3
          
SECONDARY ADRENAL INSUFFICIENCY
Secondary adrenal insufficiency results from failure of the hypothalamic or pituitary stimulation of the adrenal gland (Figure 2B). Secondary causes of adrenal insufficiency include long-term glucocorticoid use, and tumors or injury to the hypothalamic-pituitary region (Table 3).
Long-term administration of glucocorticoids induces adrenal atrophy and suppresses CRH production. The individual effect is highly variable and depends on the dose and duration of treatment, but should be anticipated in any patient who has been receiving more than 30 mg hydrocortisone per day (7.5 mg prednisone) for more than 3 weeks.9 Additionally, the use of inhaled corticosteroids in asthmatics has been associated with varying degrees of adrenal suppression.32,33
Head injury, pituitary infarction, and central nervous system depressants can impair CRH release from the hypothalamus.34 Certain drugs, frequently administered to critically ill patients, impair cortisol synthesis and metabolism. Phenobarbital and phenytoin enhance hepatic metabolism of cortisol.9,19 Etomidate, administered as a single bolus injection to facilitate endotracheal intubation, is widely used in the ICU because of the virtual absence of cardiovascular depression in hemodynamically unstable patients 35 and inhibits enzymes involved in cortisol synthesis. 36 However, recent studies report that a single bolus of etominate in critically ill patients has resulted in adrenal insufficiency that persists for 24 hours after administration.37,38
Inflammatory cytokines, released during sepsis, appear to promote corticosteroid resistance so that normal adrenal responses are suppressed.9 The systemic inflammatory response syndrome results in the release of cytokines (tumor necrosis factor), which in vitro suppresses CRH.39
Patients with chronic prolonged critical illness may acquire adrenal insufficiency. These patients may have normal adrenal function upon admission, but later acquire adrenal insufficiency. “Functional adrenal insufficiency” is a term used to describe subnormal adrenal corticosteroid production during acute illness.5 The incidence of functional adrenal insufficiency approaches 30% of all ICU patients.40 Age appears to be an additional risk factor, as the incidence among patients >55 years old is almost doubled.41 Inability to mount an adequate adrenal corticosteroid response increases mortality during critical illness.10 Therefore, early identification of functional adrenal insufficiency and treatment with exogenous corticosteroids is beneficial.9,10
          

Diagnosis of Adrenal Insufficiency

Clinical features common to both primary and secondary adrenal insufficiency include: weakness, fatigue, depression, anorexia, weight loss, nausea, vomiting, diarrhea, and a craving for salt.9,19,30 Findings on physical examination include decreased body hair, fever, hypotension, and tachycardia. Laboratory findings include hyponatremia, hypoglycemia, and normocytic normochromic anemia, and eosinophilia.
There are several features that distinguish primary and secondary adrenal insufficiency (Table 4). One of the most distinguishing features is that mineralcorticoid deficiency is present only in primary adrenal insufficiency. Mineralcorticoid deficiency causes urinary losses of water, sodium, chloride, and potassium. The result is hyponatremia, loss of extracellular fluid volume, hyperkalemia, and hypotension. Hyperpigmentation, present in primary adrenal insufficiency, is caused by elevation of ACTH. The amino acid sequence of ACTH is similar to that of the melanocyte-stimulating hormone.
Graphic Table 4
When one considers these manifestations in the face of critical illness, it is obvious that adrenal insufficiency is difficult to detect in critically ill patients. Many patients have altered levels of consciousness and are unaware of the physical symptoms. Fever and hypotension are common clinical findings in patients with hypovolemia or sepsis. Electrolyte abnormalities in the critically ill are not only common, but also masked by continuous and frequent exogenous electrolyte replacement. Hypoglycemia and eosinophilia are uncommon in critically ill patients and should alert clinicians to the possibility of adrenal insufficiency.2,5,42 The eosinphil may be the only accessible marker of adequate cortisol concentrations, although the sensitivity and specificity of eosinophil counts for the diagnosis of adrenal insufficiency have not been determined.5 There is substantial agreement that a high index of suspicion of adrenal insufficiency should be considered in the presence of unexplained catecholamine-resistant hypotension, despite adequate fluid resuscitation and ongoing evidence of inflammation (without an obvious source) that does not respond to empirical treatment.1,2,5,9,30
There is considerable debate about the diagnostic criteria for adrenal insufficiency based on serum cortisol levels. Laboratory evaluation of serum cortisol in critically ill patients is difficult for a variety of reasons. Assays for serum cortisol measure the total cortisol concentration—bound and unbound. Because more than 90% of circulating cortisol in human serum is bound to protein, alterations in the binding proteins could affect the measured concentrations of serum cortisol. Furthermore, there is no absolute serum cortisol level that distinguishes an adequate from an insufficient adrenal response. Some advocate that cortisol levels in critically ill patients should be >20 mcg/dL,2,8 whereas others argue that cortisol levels should be >18 mcg/dL,9 particularly in patients who have hypoalbuminemia.43 Although the diagnostic cortisol level is under debate, it is clear that serum cortisol levels <25 mcg/dL in critically ill patients with significant trauma, burns, and inflammation is distinctly unusual.5
As previously discussed, expected cortisol levels vary in a biphasic pattern over the course of critical illness. However, this pattern can change when an acute event (infection, surgery) occurs in the chronic phase of critical illness. In this situation, data are difficult to interpret in the presence of a mixed acute/chronic pattern of cortisol activity.3
Traditionally, HPA reactivity has been assessed using the corticotropin stimulation test. To perform this test, 250 mcg IV of synthetic corticotropin (cosyntropin) is administered. Serum cortisol levels were obtained before and 30 minutes and 60 minutes following infusion. In response to this stimulation, the normal adrenal gland should produce plasma cortical concentrations >20 mcg/dL.6,44 Diagnostic findings associated with adrenal insufficiency include stimulated serum cortisol levels <18 mcg/dL, or an increase in the cortisol concentration <9 mcg/dL, regardless of the baseline cortisol value.8,45 However, the use of the corticotropin stimulation test to diagnose adrenal insufficiency in critically ill patients remains controversial. The test, developed to diagnose primary adrenal insufficiency in the outpatient setting, may lack the sensitivity for the diagnosis of adrenal insufficiency in critically ill patients.45 If the HPA is already maximally stimulated (Figure 1) and patients already have high baseline cortisol levels, then subsequent increments in cortisol after stimulation may be small. Furthermore, the corticotropin test has limitations in the presence of hypoadrenalism secondary to recent injury or surgery in the hypothalamic or pituitary regions.
Several studies suggest that such a high dose of corticotropin creates false negatives in patients with secondary adrenal insufficiency 46,47 because the 250 mcg dose of corticotropin is supraphysiologic (>100-fold higher than normal maximal stress ACTH levels).2,19 A number of studies have demonstrated that a lower dose of corticotropin (1 mcg) is a more sensitive tool in the diagnosis of secondary adrenal insufficiency,33,41,42,46–48 although this is not widely accepted.9,49 The normal response to a 1 mcg dose of corticotropin is a rise in cortisol level 18 mcg or greater at baseline 30 to 60 minutes after infusion.44
Taking all these factors into account, Marik and Zaloga 2,7 proposed an algorithm to evaluate adrenal insufficiency in critically ill patients. The diagnostic work-up begins with a random cortisol level. It is not necessary to obtain the cortisol levels at a specific time of the day since critically ill patients loose the diurnal variation.11 A random cortisol level >25 mcg/dL represents an adequate response to the stress of critical illness and adrenal insufficiency is not present. If the random cortisol level is <20 mcg/dL, the patient is normotensive but has unexplained fever, eosinophilia, or mental status changes, then a trial of replacement doses of corticosteriods should be initiated. If the random cortisol level is <25 mcg/dL and the patient is hypotensive, then low-dose and a high-dose corticotropin stimulation tests should be performed to distinguish causes of adrenal insufficiency (Figure 3).
Graphic Figure 3
Patients who fail to respond to both low-dose and high-dose stimulation tests most likely have primary adrenal insufficiency. Those who respond to both high-dose and low-dose stimulation tests lack secretion of CRH or corticotropin and represent patients with secondary adrenal insufficiency. Patients who fail to respond to low dose, but respond normally to high-dose stimulation tests have corticotropin resistance. These patients do not respond to a physiologic dose of cortisol, but rather require a supraphysiologic dose of corticotropin to produce a normal response. ACTH resistance may result from the production of circulating antagonists or from receptor–postreceptor resistance.7
Acute adrenal insufficiency occurs in patients who are unable to increase their production and secretion of cortisol during acute stress.2 Acute adrenal insufficiency can develop in patients with primary adrenal insufficiency or secondary adrenal insufficiency.
Because the HPA axis response to the stress can change over the course of critical illness (Figure 1), patients who have initial normal laboratory test results can develop adrenal insufficiency later in the illness. The development of new clinical features suggestive of adrenal insufficiency (hypotension, eosinophilia) should prompt serial follow-up of adrenal function in long-term critically ill patients.
          

Treatment of Acute Adrenal Insufficiency

Patients who develop acute adrenal insufficiency are unable to increase their production and secretion of cortisol in times of acute stress and therefore require exogenous cortisol. Treatment regimes for acute adrenal insufficiency in critically ill patients are divided into 2 categories: (1) those with chronic inflammatory or autoimmune diseases (rheumatoid arthritis, asthma, ulcerative colitis) who are (or recently have been) treated with steroids and require additional “stress doses” and (2) those with acute adrenal insufficiency.
          
Stress Dosing
Patients with secondary adrenal insufficiency, maintained on exogenous glucocorticoids, do not have the ability to increase cortisol secretion in response to the stress of surgery or critical illness. The current standard of practice is to increase their usual dose of cortisol in the context of anticipated or actual stress. However, there is considerable controversy about the recommended dose of steroids for stress coverage. Recent studies demonstrate that patients on glucocorticoids undergoing major surgery do not require more steroids than their regular dose to maintain cardiovascular function.50–52 If the operation is complicated or the critical illness is prolonged, higher doses may be required.1 A reasonable approach to this issue has been proposed:19,53 hydrocortisone 25 mg for minor surgical stress (hernioplasty), 50 to 75 mg for moderate surgical stress (abdominal hysterectomy), and 100 to 150 mg for major surgical stress (cardiopulmonary bypass) for a period of 1 to 3 days.
           
Acute Adrenal Insufficiency
According to a treatment regime proposed by Marik and Zaloga 2 for patients with hypotension who are suspected of having (or at risk for) acute adrenal insufficiency, empiric treatment of hydrocortisone (100 mg every 8 hours) should be initiated, pending results of random cortisol testing. Others suggest a smaller dose of hydrocortisone (50 mg every 6 hours).9 If the patient clinically improves with hydrocortisone but the random cortisol level is >25 mcg/dL, the hydrocortisone should be continued for a few days (unless there is a contraindication). If the random cortisol level is <25 mcg/dL, the hydrocortisone should be continued and as the patient's clinical status improves, the dose should be tapered. This treatment regime applies to patients with acute adrenal insufficiency due to primary adrenal insufficiency, secondary adrenal insufficiency, and ACTH resistance.
Hydrocortisone is the corticosteroid of choice for patients with acute adrenal insufficiency. Both prednisone and cortisone should be avoided in hypotensive patients because they require hydroxylation to achieve the active compound (converting cortisone to cortisol and prednisone to prednisolone) and can only be administered orally.19
Patients with primary insufficiency also require mineralcorticoid replacement because the adrenal cortex does not secrete aldosterone. If the patient is on hydrocortisone, mineralcorticoid replacement is not necessary because hydrocortisone has combined glucocorticoid and mineralcorticoid activity. However, if the patient is on oral steroids (prednisone), mineralcorticoid replacement is necessary using fludrocortisone acetate (Florinef), a potent synthetic mineralcoricoid. No specific monitoring for treatment is necessary, other than occasional plasma potassium.6
Intravenous administration of hydrocortisone is continued for 24 to 48 hours. If at this time, the patient is hemodynamically stable, oral glucocorticoid therapy may be started (50 mg every 8 hours) for 48 hours.44 Following maintenance oral therapy, a hydrocortisone taper is initiated until the dose is 30 to 50 mg/day in divided doses. Mineralcorticoid supplementation may be required, particularly in the presence of hyperkalemia.
It is often unclear whether acute adrenal insufficiency is functional and transient during critical illness or whether it is due to established structural disease of the HPA axis.9 Long-term corticosteroid replacement should not be made on the basis of physical findings and laboratory tests conducted during critical illness. That decision should be made when diagnostic tests are repeated after resolution of the critical illness.
           

Conclusion

The stress associated with critical illness and treatment results in activation of the HPA axis and subsequent release of cortisol, which is essential for cardiovascular, metabolic, and immunologic homeostasis. The extent to which serum cortisol levels should increase in under considerable debate. During critical illness, the normal cortisol response is altered and 1 of 2 conditions can be present: hypercortisolemia or adrenal insufficiency. Both are associated with significant morbidity and mortality. The eucorticoid state supports a balance between proinflammatory and anti-inflammatory activities that allows the invaluable effects of inflammation (host defenses, wound healing) to proceed while keeping the harmful effects (decreased host defenses and wound healing, development of multiple system organ dysfunction) in check. Despite our understanding that circulating cortisol is beneficial, there is no consensus about what the most advantageous serum cortisol level is during critical illness. Most advocate that a random serum cortisol level in a critically ill patient should be >25 mcg/dL. Levels less than this may indicate adrenal insufficiency. Acute adrenal insufficiency occurs in patients who cannot increase cortisol production during critical illness. Primary adrenal insufficiency is caused by a failure of the adrenal gland and can be life threatening. Secondary adrenal insufficiency results from failure of hypothalamic or pituitary stimulation of the adrenal gland. Signs and symptoms of adrenal failure are ubiquitous in critically ill patients and therefore make diagnosis difficult. However, a high index of suspicion should be considered in critically ill patients with eosinophilia who have unexplained catecholamine-resistant hypotension, despite fluid resuscitation, and have ongoing evidence of inflammation that does not respond to empirical treatment. HPA reactivity can be assessed using random serum cortisol levels, and low-dose or high-dose corticotropin stimulation tests. The use of these tests in critically ill patients is controversial. Patients with adrenal insufficiency require exogenous corticosteroid replacement, preferably with hydrocortisone. After achieving hemodynamic stability, the corticosteriods can be tapered and an oral agent can be administered. Acute adrenal insufficiency may be a transient condition during critical illness. Long-term corticosteriod replacement should be reevaluated after resolution of the critical illness.

Comments

Popular posts from this blog

Classification Systems for Acute Kidney Injury

Classification Systems for Acute Kidney Injury Background:- Acute kidney injury (AKI), formerly called acute renal failure (ARF), is commonly defined as an abrupt decline in renal function, clinically manifesting as a reversible acute increase in nitrogen waste products—measured by blood urea nitrogen (BUN) and serum creatinine levels—over the course of hours to weeks. The vague nature of this definition has historically made it difficult to compare between scholarly works and to generalize findings on epidemiologic studies of AKI to patient populations. Several classification systems have been developed to streamline research and clinical practice with respect to AKI.   Acute Kidney Injury Network:- In September 2004, the Acute Kidney Injury Network (AKIN) was formed. AKIN advised that the term acute kidney injury (AKI) be used to represent the full spectrum of renal injury, from mild to severe, with the latter having i...

Thinking About Plan B? - Nurses in Business

Thinking About Plan B? - Nurses in Business It is an exciting time to be a nurse! There are so many options for nurses including owning a business. Although most nurses do not see themselves as entrepreneurs, nurses are educated in a scientific method of problem solving and resolution that equips them for a successful transition into an entrepreneurial role. It is an exciting time to be a nurse! There are so many options for today’s nurse and one area that is growing is nurses interested in becoming business owners. What is creating this need for  Plan   B usiness? Well one thing is that the number of nurses employed by hospitals is decreasing from what it has been for decades. Today, about 58% of nurses are employed by hospitals compared to 68% in 1980. Many nurses, over 30%, report being burned out and highly dissatisfied with their jobs. The dissatisfaction relates to feelings of frustration, being overwhelmed with new technology (equipment and EMRs) and overworked du...

Gastrointestinal Infections in the ICU

Gastrointestinal Infections in the ICU    Intra-abdominal infections are a major cause of morbidity, mortality and antibiotic expenditure in the ICU [21]. Accurate and timely diagnosis can have a major impact on clinical outcome, antimicrobial selection, healthcare cost and need for surgical intervention. Spontaneous bacterial peritonitis in the ICU is commonly seen in decompensated cirrhotic patients, likely due to the translocation of overgrowing enteric bacteria (usually gram negative organisms, although MRSA has been commonly described in ICU patients) across an anatomically intact gastrointestinal tract. Gastrointestinal wall perforation or ulceration can result in polymicrobial seeding into neighboring areas, resulting in signs of acute abdomen. Localized pain suggests the infection is walled-off in the area directly associated with the area of seeding, whereas diffuse pain suggests generalized peritonitis. Intra-abdominal abscesses, bowel perforation, cholecystitis, a...