Background

Sepsis syndrome represents the body’s host response to an infection. The causative agent and the host’s activated inflammatory cascade overwhelm the body’s defenses and regulatory systems, leading to disruption in homeostasis. Tachycardia, tachypnea, fever, and immune system activation are common manifestations. If the body is unable to overcome this insult, cellular injury, tissue damage, shock, multiorgan failure, or death may ensue.

In 1992, the American College of Chest Physicians/Society of Critical Care Medicine issued a consensus statement to establish uniform criteria defining the sepsis syndromes.¹ For the first time, this allowed a common nomenclature for disease classification and systematic comparisons across studies of septic patients. Systemic inflammatory response syndrome (SIRS) is defined as two or more of the following: tachycardia, tachypnea, hyperthermia or hypothermia, high or low white blood cell count, or bandemia; sepsis is the combination of infection plus SIRS; severe sepsis is sepsis plus organ dysfunction; and septic shock is severe sepsis plus hypotension, defined as a systolic blood pressure below 90 mm Hg, not responsive to a fluid challenge (Box 138-1). The importance of this nomenclature is to provide clinicians and researchers with a common classification. Efforts to validate this classification scheme in the emergency department (ED) population have demonstrated that the term sepsis, when it is characterized by fulfilling of the SIRS criteria alone, is perhaps overly sensitive and nonspecific and does not convey an increased mortality risk, although it should prompt further evaluation.² However, organ dysfunction and shock were shown to be significant predictors of adverse outcome. Newer efforts have proposed the PIRO approach, which may help better understand and prognosticate the severity of illness.^3,4 Assessment of predisposing conditions, infection source, response of the host, and organ dysfunction is proposed to help improve classification. The clinician should approach the septic patient with goals to determine who they are (e.g., underlying comorbidities), what infection they have, how they are responding, and where/if organ dysfunction is present.

Bacteremia is often present, but positive cultures are not obligatory in the diagnosis of sepsis. In several prospective studies, only 17 to 27% of patients with sepsis, 25 to 53% of patients with severe sepsis, and 69% of patients with septic shock actually had positive blood cultures.5–7 Culture-negative and culture-positive septic populations have similar outcomes in patients with similar severity of illness.⁸ Pneumonia, abdominal abscess with viscus perforation, and pyelonephritis are common primary causes of sepsis. Gram-positive organisms account for 25 to 50% of infections, gram-negative organisms for 30 to 60%, and fungi for 2 to 10%. The distribution varies with the study and, more important, with host factors such as the status of the host immune system, age of the patient, recent hospitalizations, and presence of indwelling vascular catheters.

The health status of the host is a crucial risk factor in the development and progression of sepsis.⁹ Elders and those with multiple comorbidities are overwhelmed more easily by systemic infection. Chemotherapy-induced neutropenia, acquired immunodeficiency syndrome, and steroid dependency increase susceptibility to sepsis. Increased use of indwelling devices such as intravascular catheters, prosthetic devices, and endotracheal tubes contributes to the risk of systemic infection and sepsis.

Epidemiology

Sepsis is now the tenth most common cause of death in the United States.¹⁰ It is estimated that 571,000 cases of severe sepsis present to U.S. EDs each year.¹¹ Mortality rates from sepsis are estimated between 20 and 50%.^12–14 Sepsis accounts for 4 of 1000 ED visits in the United States.¹⁴ The incidence of sepsis as a reason for hospitalization is rising disproportionately among elders compared with the young (Fig. 138-1). The incidence of sepsis in patients younger than 65 years is less than 5 in 100,000, whereas it is 26 in 100,000 in those 65 years or older.¹² The cost of caring for septic patients is estimated to be $17 billion per year in the United States. Hospital discharge data have shown that the incidence of sepsis is increasing as identification improves and the population ages. Estimates have suggested that the incidence will rise 1.5% per annum or more. The number of ED visits for sepsis has risen proportionally with the rise in ED volume during the past 15 years.¹⁴ Recent research has confirmed the long-held belief that respiratory and genitourinary infections are the most common causes of sepsis.¹¹

Pathophysiology

Sepsis results from the complex interaction of detection molecules, signaling molecules, and numerous inflammatory and coagulation mediators in response to infection. More than 30% of patients with sepsis do not have a definite microbiologic diagnosis.¹⁵ Although our understanding of the pathophysiologic process of sepsis has evolved in the past two decades, it remains incomplete.¹⁶ The initial host response is to mobilize inflammatory cells, particularly neutrophils and macrophages, to the site of infection. These inflammatory cells then release circulating molecules, including cytokines, which trigger a cascade of other inflammatory mediators that result in a coordinated host response. Synthesis of the components of the cascade is increased at many steps along the pathway. If these mediators are not appropriately regulated, sepsis will occur. In the setting of ongoing toxin release, a persistent inflammatory response occurs with ongoing mediator activation, cellular hypoxia, tissue injury, shock, multiorgan failure, and potentially death.

Mediators of Sepsis

Host response and pathogen characteristics are both important in the pathogenesis of sepsis. More than 100 discrete markers have been identified and attributed to the sepsis cascade, but the true culprits have not been clearly identified.1,17 A pathogen is sensed by pattern recognition receptors, most notably toll-like receptors, located on the surface of the white blood cell.¹⁶ The resulting host-pathogen interaction activates the inflammatory and coagulation cascades.^18–21 The subsequent inflammatory signaling occurs through cytokines, chemokines, and other soluble mediators, including increased circulating levels of the interleukins IL-1, IL-6, and IL-8 and tumor necrosis factor alpha (TNF-α). Activation of the clotting cascade results in increased D-dimer levels and decreased circulating levels of protein C.²²

In benign conditions, a self-limited response helps clear the pathogen. If the innate immune response is inadequate, mediators create a procoagulant state. Coagulation and fibrinolytic components are proinflammatory, precipitating a worsening cycle of procoagulant and proinflammatory mediators. This cascade ultimately creates end-organ damage and disseminated intravascular coagulation (DIC). If it is not effectively reversed, the process leads to cellular hypoxia, organ dysfunction, shock, and death.

The primary mediators are cytokines that are primarily proinflammatory, anti-inflammatory, or growth promoting. The molecular mechanisms by which they are regulated are not well understood. The initial cytokine, TNF-α, is found in serum approximately 90 minutes after the administration of endotoxin to healthy volunteers. IL-6 and IL-8 reach peak levels at approximately 120 minutes. The main proinflammatory cytokines are IL-1, TNF, and IL-8. The primary anti-inflammatory cytokines are IL-10, IL-6, transforming growth factor beta, soluble receptors to TNF, and IL-1 receptor antagonist (IL-1RA). If the resultant inflammatory response is adequate, the infection is controlled and cleared. If the response is deficient or excessive, however, a persistent and worsening cascade is produced, ultimately leading to shock, organ failure, and potentially death.

Instability in vascular tone is becoming increasingly important in understanding of the pathophysiologic mechanism of sepsis. Vasopressin, also known as antidiuretic hormone, is a naturally occurring hormone that is essential for cardiovascular stability. It is produced as a prohormone in the hypothalamus. The hormone is stored in the pituitary gland and released in response to stressors such as pain, hypoxia, hypovolemia, and hyperosmolality.²³ In severe sepsis, there is a brief rise in circulating vasopressin levels followed by a prolonged and severe suppression. This pattern of secretion is different from other forms of shock, in which vasopressin levels remain elevated. Vasopressin has numerous physiologic effects, including vasoconstriction of the systemic vasculature, osmoregulation, and maintenance of normovolemia.

Nitric oxide (NO) is a gas that has an important role in septic shock, regulating vascular tone by an indirect effect on smooth muscle cells. NO also contributes to platelet adhesion, insulin secretion, neurotransmission, tissue injury, and inflammation and cytotoxicity. Its half-life is short (6-10 seconds), and it easily diffuses into cells. Although its mechanisms of action are not well understood, it seems to be a key mediator of sepsis. Animal data show that nitric oxide synthase, the enzyme that produces NO, is upregulated in cases of sepsis.²⁴ Enhanced NO production is thought to contribute to the profound vasodilation found in patients in septic shock.

In the setting of ongoing inflammatory activation, the mediators of sepsis continue to be produced and the cascade is perpetuated. Unless it is appropriately and rapidly controlled, the ultimate effect is a sequence of events starting with cellular dysfunction and ultimately leading to tissue damage, organ dysfunction, and death.

Organ System Dysfunction

The organ dysfunction that results from sepsis is central to the pathogenesis of the disease. A 3000-patient ED-based study demonstrated that organ dysfunction with septic shock portends increasingly worse outcomes.² Patients with suspected infection alone had a mortality rate of 2.1%, whereas the presence of SIRS criteria and suspected infection had a mortality rate of only 1.3% (Fig. 138-2A). However, the mortality rate was 9% for those patients with severe sepsis (sepsis plus organ dysfunction) and 28% for those with septic shock. The risk of death from sepsis approximately doubles for each organ that fails. The mortality rate for patients with no organ dysfunctions was 1%; rates for dysfunction of a single organ, two organs, three organs, and four or more organs were 6%, 13%, 26%, and 53%, respectively (Fig. 138-2B).

Novel Concepts

In recent years, there has been increasing evidence that genetics are a risk factor for outcome of sepsis.30,31 The impact of genetics on future treatment modalities for sepsis remains unclear, but the prospect of customized genetic therapy for sepsis is a promising early development.

Symptoms and Signs

The approach to a patient with sepsis relies on identification of the presence of a systemic infection and localization of the source of the initial infection. This allows appropriate and aggressive treatment directed to the source of infection. Often, the source is not readily apparent, but early identification of the septic state allows implementation of broad-spectrum antibiotics that may be potentially lifesaving.

The septic patient manifests signs of systemic infection through tachycardia, tachypnea, hyperthermia or hypothermia, and, if severe, hypotension. A septic patient will often have flushed skin with warm, well-perfused extremities secondary to the early vasodilation and hyperdynamic state. Alternatively, the severely hypoperfused patient with an advanced shock state may appear mottled and cyanotic. Very early in the patient’s presentation, vital sign changes such as tachycardia and tachypnea may be the only early indicators of sepsis. If the patient is in shock, a rapid assessment that excludes other causes, such as hypovolemic or cardiogenic shock, is essential to the proper initial treatment. A complete detailed clinical examination will help the physician determine the cause of the shock state (see Chapter 6). The aforementioned are “classic” signs; however, these findings may not be manifested in a septic patient, and signs and symptoms may be subtle.

Both underlying comorbidities and the etiology of sepsis should be considered. Risk factors such as immunocompromised states (acquired immunodeficiency syndrome, malignant disease, diabetes, splenectomy, and concurrent chemotherapy), older age, debilitation or high-risk environments for iatrogenic infections (such as long-term care facilities), and multiple comorbidities should be considered.

The respiratory system is the most common source of infection in the septic patient. A history of a productive cough, fevers, chills, upper respiratory symptoms, and throat and ear pain should be sought. Both the presence of pneumonia and the findings of tachypnea or hypoxia have been found to be predictors of death in patients with sepsis.³² Physical examination should also include detailed evaluation for focal infection, such as exudative tonsillitis, sinus tenderness, tympanic membrane injection, and crackles or dullness on lung auscultation. Also, pharyngeal thrush should be noticed as a potential marker of an immunocompromised state.

The gastrointestinal system is the second most common source of sepsis. A history of abdominal pain, including its description, location, timing, and modifying factors, should be sought. Further history, including the time of the last bowel movement and the presence of nausea, vomiting, and diarrhea, should be noted. A careful physical examination, looking for signs of peritoneal irritation, abdominal tenderness, and hyperactive or hypoactive bowel sounds, is critical in identifying the source of abdominal sepsis. Particular attention should be paid to physical findings suggestive of common sources of infection or disease: Murphy’s sign indicating cholecystitis, pain at McBurney’s point indicating appendicitis, left lower quadrant pain suggesting diverticulitis, or rectal examination revealing a rectal abscess or prostatitis.

The neurologic system is examined by looking for signs of meningitis, encephalitis, or epidural abscess, including nuchal rigidity, fevers, and change in consciousness. A detailed neurologic examination is important. Lethargy or altered mentation may indicate primary neurologic disease or may be the result of decreased brain perfusion from a shock state.

The genitourinary history includes queries about the presence of flank pain, dysuria, polyuria, discharge, Foley catheter placement, and genitourinary instrumentation. A sexual history should assess for the risk of sexually transmitted diseases. Genitalia should be evaluated for ulcers, discharge, and penile or vulvar lesions, looking specifically for the woody induration of Fournier’s gangrene. A rectal examination should be performed, looking for a tender, boggy prostate, consistent with prostatitis. A red and friable cervix, cervical discharge, or cervical motion tenderness is consistent with a sexually transmitted disease. Adnexal tenderness in a toxic-appearing woman potentially represents a tubo-ovarian abscess. Also, a pelvic examination in women should also include an inspection to ensure that there is no retained tampon.

Musculoskeletal history includes the presence of any localizing symptoms to a particular joint. Redness, swelling, and warmth over a joint, especially if there is a decreased range of motion in that joint, may be signs of septic arthritis and may mandate arthrocentesis. A patient should be completely exposed and the skin examined for evidence of cellulitis, abscess, wound infection, or traumatic injury. Deep injuries, foreign bodies, and fasciitis may be difficult to identify clinically. The physician should look for crepitus, bullae, or skin edema extending beyond areas of erythema representing the presence of an aggressive, gas-forming organism. Back pain and fever are concerning for an epidural abscess. Local lymphadenopathy, swelling, and streaking should also be noted as signs of an advancing infection. Petechiae and purpura may represent a Neisseria meningitidis infection or DIC. Generalized erythroderma and rash may represent an exotoxin from pathogens such as Staphylococcus aureus and Streptococcus pyogenes.

A history of fevers or chills in the setting of injection drug abuse, an artificial heart valve, or a mitral valve prolapse should increase the suspicion for endocarditis. The clinician should suspect endocarditis in the presence of a murmur or other stigmata of endocarditis (e.g., splinter hemorrhages, Roth’s spots, Janeway’s lesions).

Clinicians must identify the severity of illness in patients with infection and initiate aggressive resuscitation for those patients with the potential of becoming critically ill. Although a patient may meet SIRS criteria, this alone has little predictive value in determining the severity of illness and mortality.² The Mortality in Emergency Department Sepsis (MEDS) score is a method to risk stratify ED patients with sepsis.³³ The MEDS prediction rule assigns point values to specific clinical characteristics (Table 138-1). The total score can be used to assess risk of death. Thus the greater the number of risk factors, the more likely a patient is to die during the hospitalization.

Hematology

The white blood cell count is a marker of inflammation and activation of the inflammatory cascade. Leukocytosis is associated with infection and is incorporated in the consensus definition of sepsis; however, it is often insensitive and nonspecific, limiting its absolute utility in the ED. The febrile neutropenic patient has been shown to be at increased risk for severe infection. Thus a neutrophil count of less than 500 cells/mm³ should prompt admission, isolation, and empirical intravenous antibiotics in most chemotherapy patients. A bandemia (≥10% bands on a peripheral smear) represents the release of immature cells from the bone marrow and is considered to be a sign of infection and inflammation. Like the white blood cell count, it is an imperfect indicator of infection. The absence of leukocytosis or bandemia does not preclude the possibility of severe sepsis. The hemoglobin and hematocrit should be determined to ensure adequate oxygen delivery in shock. Platelets are an acute-phase reactant and may be elevated in the presence of infection. Conversely, a low platelet count is a significant predictor of bacteremia in patients with shock. Thrombocytopenia, elevated prothrombin time, elevated activated partial thromboplastin time, decreased fibrinogen, and increased fibrin split products are associated with DIC and severe sepsis syndrome.

Chemistry

Electrolyte abnormalities should be identified and corrected. Low bicarbonate level suggests acidosis and inadequate perfusion. An elevated anion gap acidosis in the setting of sepsis syndrome commonly represents lactic acidosis or diabetic ketoacidosis, but other causes need to be ruled out. Elevated serum creatinine concentration or decreased glomerular filtration rate signals renal dysfunction or failure, which, if it is due primarily to sepsis, indicates organ failure and a worse prognosis. Calcium, magnesium, and phosphorus levels should be checked.

An elevated lactate level is associated with inadequate perfusion, shock, and poorer prognosis. One ED-based study showed a progression in mortality rate with increasing venous lactate level: a lactate level of 0 to 2.5 mg/dL was associated with a 5% mortality rate; a lactate level of 2.5 to 4.0 mg/dL, 9% mortality; and a lactate level greater than 4 mg/dL, 28% mortality.³⁴ An arterial blood gas assessment may be helpful in identifying and classifying acid-base disturbances. Metabolic acidosis suggests inadequate tissue perfusion. Liver function tests can be used to identify liver failure or dysfunction. An elevated bilirubin level may suggest the gallbladder as a cause of sepsis. An elevated lipase level may represent pancreatitis as the cause of noninfectious SIRS.

Microbiology

Proper blood, sputum, urine, cerebrospinal fluid, and other tissue culture samples are important in guiding therapy. Although the results of culture are not helpful in the initial management, culture samples should be obtained before or soon after the administration of antibiotics in the patient with sepsis syndrome. The initiation of antibiotic therapy should not be delayed while waiting for culture samples to be obtained. The literature suggests that the yield of initial blood cultures is low (5-10%), but this is probably an artifact of the lack of reliable discriminatory guidelines for obtaining blood culture samples in the ED.⁶ Among patients with clinical sepsis, only 30 to 60% of patients will have positive cultures.^5,7

The results of initial microbiologic tests, including Gram’s stains whenever possible, will help guide subsequent antibiotic treatment. Initial empirical therapy should be broad spectrum to allow early treatment of all likely organisms.

Special Procedures

The comprehensive protocol of early goal-directed therapy published by Rivers and associates³⁵ showed that a resuscitation strategy guided by the central venous pressure (CVP), an arterial line, and a central venous catheter with continuous central venous oxygen saturation (ScvO₂) monitoring capability can reduce mortality in patients with sepsis. However, whether each of these components is required is unclear. A CVP line may guide fluid resuscitation, with a low CVP indicating the need for continued fluid repletion. The ScvO₂ monitor measures venous blood saturation, which is a marker of tissue metabolism and extraction. It has been suggested that intermittent monitoring of ScvO₂ may suffice, although there are no firm data to support this. The goal is to ensure proper tissue oxygenation, which is routinely achieved by adjustment of oxygen delivery, keeping the ScvO₂ greater than 70%. An arterial line can be useful for close monitoring of hypotensive patients, especially when one or more vasopressors are being titrated to maintain an adequate blood pressure; however, no evidence exists to show that it is mandatory. Finally, the Swan-Ganz catheter is not a mandatory part of the sepsis management, although the physiologic measurements may be useful in identifying the cause of shock and guiding fluid and inotropic therapy. Low systemic vascular resistance and high cardiac output are most commonly associated with sepsis, although this may vary with the stage of shock and the individual patient. There are data to suggest that the use of Swan-Ganz catheters does not have a substantial impact on mortality.³⁶

Radiology

A chest radiograph should be obtained in all patients with suspected sepsis syndrome, looking not only for a focal infiltrate representing pneumonia but also for the fluffy, bilateral infiltrates indicative of ARDS. The pathophysiologic changes of ARDS are often delayed as much as 24 hours after radiographic identification. An upright chest radiograph should be obtained for suspected bowel perforation to detect free air under the diaphragm. The presence of pneumomediastinum is suggestive of esophageal perforation and current or impending mediastinitis.

Soft tissue plain radiographs of infected areas should be obtained, looking for air in the soft tissues associated with necrotizing or gas-forming infection. Periosteal thickening or bone erosion may be seen on plain radiographs of patients with osteomyelitis; a bone scan may be diagnostic. Computed tomography (CT) of superficial infections may be helpful to further quantify the extent of infection and to identify abscesses that are not readily evident on physical examination. A CT scan of the abdomen and pelvis may identify abdominal or pelvic pathologic lesions, provided there is no clear clinical indication for immediate operative intervention. Suspected disease, such as diverticulitis, appendicitis, necrotizing pancreatitis, microperforation of the stomach or bowel, or formation of an intra-abdominal abscess, may be best diagnosed by CT scan. A head CT scan can identify septic emboli from endocarditis or increased intracranial pressure from a mass and should be considered before a lumbar puncture is performed. An abdominal ultrasound examination may be indicated for suspected cholecystitis, and a pelvic ultrasound examination for tubo-ovarian abscess or endometritis. If endocarditis is suspected, a transesophageal cardiac ultrasound study should be obtained for the detection of any valvular vegetations. A magnetic resonance imaging scan can be useful to identify soft tissue infection, such as necrotizing fasciitis or epidural abscess.

Preload

The first step in goal-directed therapy is to provide a proper filling pressure to ensure adequate cardiac preload. Patients with sepsis and septic shock are often in a state of substantial fluid deficit. Rivers and colleagues showed that patients required an average of 5 L of fluids in the first 6 hours of their resuscitation. A CVP measurement should be obtained, and the patients should receive crystalloid or colloid resuscitation to keep the CVP at a level of 8 to 12 mm Hg. In intubated patients with positive-pressure ventilation, pathophysiologic principles support raising the target level to 12 to 16 mm Hg. Whereas CVP monitoring is one approach, other approaches may be used. For example, if cardiac output monitoring is available, an approach of administering intravenous fluids until the point at which cardiac output no longer responds to fluid boluses is certainly reasonable.

Perfusion Pressure

The next step in goal-directed therapy is to maintain an adequate blood pressure. A mean arterial blood pressure (⅔ diastolic + ⅓ systolic) should be kept between 60 and 90 mm Hg. If the mean arterial blood pressure is below 65 mm Hg in the presence of adequate preload, treatment with a vasopressor agent should be started; norepinephrine and dopamine are considered first-line agents.

Oxygen Delivery

The initial management of all critically ill patients should include supplemental oxygen with a goal to maintain a peripheral oxygen saturation of 95% (or the patient’s baseline oxygen saturation in the case of preexisting underlying lung disease). Once preload and pressure are normalized, the next focus is on oxygen delivery. A mixed venous saturation or assessment of lactate clearance can be used to guide this aspect of resuscitation. Organ hypoperfusion is a result of global and distributive changes in both systemic blood flow and the microvasculature. As a result of marrow suppression and dilution with crystalloid fluids, many patients with sepsis have a hemoglobin between 8 and 10 g/dL after initial resuscitation. A hematocrit between 27 and 30% has been reported to be most effective,³⁹ although other authorities have challenged this dogma.⁴⁰

The therapeutic goal is to maintain a mixed venous saturation (ScvO₂) greater than 70% or to reduce lactate level by at least 10%. As oxygen demand due to sepsis increases, the ScvO₂ starts to fall below 70%. This should be corrected by first ensuring that preload and pressure are adequate.⁴¹ Next, the arterial oxygen saturation should be optimized with non-rebreather oxygen delivery or intubation, as needed. Once these are optimized, one should ensure that the patient has adequate oxygen-carrying capacity by transfusion of the patient to a hematocrit greater than 30%. If this is normalized and the patient is not tachycardic, dobutamine can be added to increase cardiac contractility and to increase cardiac output, resulting in an overall improved oxygen delivery. The protocol should be run continuously with an effort toward normalization of the parameters during the entire resuscitation (Table 138-2).

Respiratory Support

Altered mental status is common in patients in septic shock, and they may require rapid airway protection. Patients with a respiratory rate of more than 30 breaths/minute are likely to develop respiratory collapse, irrespective of arterial oxygenation. In 1999, Wheeler and Gordon reported that 85% of patients with “severe sepsis” required mechanical ventilation during their hospital course.⁴² Because patients with impending respiratory failure use a disproportionately large amount of energy for the muscles of respiration, improved oxygen delivery to other organs is achieved by mechanical ventilation, sedation, and paralysis. Although there are no clear intubation guidelines, hypercapnia, persistent hypoxemia, airway compromise, and profound acidosis are valid indicators.

In addition to airway protection, intubation and mechanical ventilatory support provide positive-pressure ventilation. The pattern of injury in ARDS is such that normal lung parenchyma is adjacent to affected tissue. Therefore, increased airway pressures are required to maintain normal oxygen delivery. Current recommendations are to maintain transalveolar pressures (measured as plateau pressures) below 35 cm H₂O because increased pressures are associated with ventilator-induced lung injury. Maintenance of relatively low transalveolar pressure with increasing end-expiratory pressure is an effective way to increase arterial oxygen delivery.⁴³ The ARDSNet trial established the benefit of low tidal volumes in acute lung injury (6 mL/kg), which has translated to a recommended approach of lower tidal volumes in mechanically intubated patients to prevent subsequent iatrogenic lung damage.⁴⁴

Cardiovascular Support

Vasopressin

Vasopressin is a naturally occurring nonapeptide that is synthesized as a large prohormone in the hypothalamus. In states of septic shock, there is an early surge of vasopressin followed by a profound drop in circulating vasopressin levels.²³ This is the foundation for use of vasopressin as an adjunct therapy for patients with severe sepsis. Vasopressin should not be used as the sole initial therapy for refractory septic shock. In a well-designed randomized trial, investigators demonstrated no change in mortality for patients with severe sepsis when vasopressin was added to catecholamine vasopressors.⁴⁶

Inotropic Agents

Bicarbonate

Bicarbonate supplementation was previously the standard treatment for patients with presumed lactic acidosis. Current consensus is that it should be reserved for severe acidemia (pH < 7.0-7.2) as there may be a paradoxical decrease in intracellular pH as a result of diffusion of soluble carbon dioxide across the cell membrane. Alternatively, hyperventilation has been suggested to help increase systemic pH.

Antibiotics

Early and appropriate antibiotic therapy should target the nidus of infection; this reduces mortality, perhaps by as much as 30 to 50%.⁵⁴ If the patient’s condition permits, appropriate culture specimens should be obtained before the administration of broad-spectrum antibiotics (Table 138-4). Surgically correctable conditions, such as intra-abdominal abscesses, perforated viscus, retained products of conception, or retained foreign body (such as a tampon), should be treated concurrently.

In the absence of an obvious source of infection, the use of broad-spectrum antibiotics is recommended. The specific agent depends on many variables, including institutional preference and local resistance patterns. As results from cultures become available, therapy should be modified. There is no consensus about the need for double or triple antibiotic coverage for particular organisms, although it is common practice to double-cover virulent organisms, such as Pseudomonas aeruginosa, as well as areas that are commonly infected with multiple organisms, such as the peritoneum. With increasing rates of methicillin-resistant organisms, combinations that include nonpenicillin choices may be warranted.

Novel Therapies

1. Bone, R, American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med. 1992;20:864–874.

2. Shapiro, NI, et al. The association of sepsis syndrome and organ dysfunction with mortality in emergency department patients with suspected infection. Ann Emerg Med. 2006;48:583–590.

3. Howell, M, et al. Proof of principle: The predisposition, infection, response, organ failure sepsis staging system. Crit Care Med. 2011;39:322–327.

4. Levy, MM, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med. 2003;31:1250–1256.

5. Rangel-Frausto, M, et al. The natural history of the systemic inflammatory response syndrome (SIRS). JAMA. 1995;273:117–123.

6. Bates, D, Cook, E, Goldman, L, Lee, T. Predicting bacteremia in hospitalized patients: A prospectively validated model. Ann Intern Med. 1990;113:495–500.

7. Brun-Buisson, C, et al. Incidence, risk factors, and outcome of severe sepsis and septic shock in adults. JAMA. 1995;274:968–974.

8. Perl, T, Dvorak, L, Hwang, T, Wenzel, R. Long-term survival and function after suspected gram-negative sepsis. JAMA. 1995;274:338–345.

9. Murray, S, Bates, DW, Ngo, L, Ufberg, J, Shapiro, NI. Charlson Index is associated with one-year mortality in emergency department patients with suspected infection. Acad Emerg Med. 2006;13:530–536.

10. Xu, J, Kochanek, K, Murphy, S, Tejada-Vera, B. Deaths: Final data for 2007. Natl Vital Stat Rep. 2010;58:1–51.

11. Wang, H, Shapiro, NI, Angus, D, Yealy, D. National estimates of severe sepsis in United States emergency departments. Crit Care Med. 2007;35:1928.

12. Angus, D, et al. Epidemiology of severe sepsis in the United States: Analysis of incidence, outcome, and associated costs of care. Crit Care Med. 2001;29:1303–1310.

13. Martin, G, Mannino, D, Eaton, S, Moss, M. The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med. 2003;348:1546–1554.

14. Strehlow, M, et al. National study of emergency department visits for sepsis, 1992-2001. Ann Emerg Med. 2006;48:326.

15. Marshall, JC, Reinhart, K, Forum, IS. Biomarkers of sepsis. Crit Care Med. 2009;37:2290–2298.

16. Russell, J. Management of sepsis. N Engl J Med. 2006;355:1699–1713.

17. Ferreira, F, et al. Serial evaluation of the SOFA score to predict outcome in critically ill patients. JAMA. 2001;286:1754.

18. Akira, S, Uematsu, S, Takeuchi, O. Pathogen recognition and innate immunity. Cell. 2006;124:783–801.

19. Schouten, M, Wiersinga, W, Levi, M, van der Poll, T. Inflammation, endothelium, and coagulation in sepsis. J Leukoc Biol. 2008;83:536–545.

20. Opal, S, Huber, C. Bench-to-bedside review: Toll-like receptors and their role in septic shock. Crit Care Med. 2002;6:125.

21. Zhao, B, Bowden, R, Stavchansky, S, Bowman, P. Human endothelial cell response to gram-negative lipopolysaccharide assessed with cDNA microarrays. Am J Physiol Cell Physiol. 2001;281:C1587.

22. Toussaint, S, Gerlach, H. Activated protein C for sepsis. N Engl J Med. 2009;361:2646–2652.

23. Holmes, CL, Walley, KR, Chittock, DR, Lehman, T, Russell, JA. The effects of vasopressin on hemodynamics and renal function in severe septic shock: A case series. Intensive Care Med. 2001;27:1416–1421.

24. Beasley, D, Schwartz, J, Brenner, B. Interleukin-1 induces prolonged L-arginine–dependent cyclic guanosine monophosphate and nitrite production in rat vascular smooth muscle cells. J Clin Invest. 1991;87:602–608.

25. Eidelman, L, Putterman, D, Putterman, C, Sprung, C. The spectrum of septic encephalopathy: Definitions, etiologies, and mortalities. JAMA. 1996;275:470–473.

26. Werdan, K, et al. Impaired regulation of cardiac function in sepsis, SIRS, and MODS. Can J Physiol Pharmacol. 2009;87:266–274.

27. Lipiner-Friedman, D, et al. Adrenal function in sepsis: The retrospective Corticus cohort study. Crit Care Med. 2007;35:1012–1018.

28. Faust, SN, et al. Dysfunction of endothelial protein C activation in severe meningococcal sepsis. N Engl J Med. 2001;345:408–416.

29. Bernard, GR, et al. Safety and dose relationship of recombinant human activated protein C for coagulopathy in severe sepsis. Crit Care Med. 2001;29:2051–2059.

29a. Ranieri, VM, et al. Drotrecogin alfa (activated) in adults with septic shock. N Engl J Med. 2012;366:2055.

30. Sheu, C, et al. A large-scale genetic association study identifies different susceptibility loci of sepsis-related and non–sepsis-related ARDS. Am J Respir Crit Care Med. 2010;181:A2702.

31. Dolinay, T, et al. Gene expression profiling of sepsis patients reveals inflammasome-mediated susceptibility to acute respiratory distress syndrome development. Am J Respir Crit Care Med. 2011;183:A2896.

32. Shapiro, NI, et al. Mortality in Emergency Department Sepsis (MEDS) score: A prospectively derived and validated clinical prediction rule. Crit Care Med. 2003;31:670–675.

33. Sankoff, J, et al. Validation of the Mortality in Emergency Department Sepsis (MEDS) score in patients with the systemic inflammatory response syndrome (SIRS). Crit Care Med. 2008;36:421–426.

34. Shapiro, NI, et al. Serum lactate as a predictor of mortality in emergency department patients with infection. Ann Emerg Med. 2005;45:524–528.

35. Rivers, E, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345:1368–1377.

36. Shah, M, et al. Impact of the pulmonary artery catheter in critically ill patients: Meta-analysis of randomized clinical trials. JAMA. 2005;294:1664–1670.

37. Bernard, GR, et al. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med. 2001;344:699–709.

38. Dellinger, RP, et al. Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock. Crit Care Med. 2008;36:296–327.

39. Czer, L, Shoemaker, W. Optimal hematocrit value in critically ill patients. Surg Gynecol Obstet. 1978;147:363–368.

40. Hebert, P, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med. 1999;340:409–417.

41. Pope, J, et al. Multicenter study of central venous oxygen saturation (ScvO₂) as a predictor of mortality in patients with sepsis. Ann Emerg Med. 2010;55:40–46.

42. Wheeler, A, Gordon, R. Treating patients with severe sepsis. N Engl J Med. 1999;340:207–214.

43. Dellinger, RP, et al. Surviving Sepsis Campaign guidelines for management of severe sepsis and septic shock. Intensive Care Med. 2004;30:536–555.

44. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network. N Engl J Med. 2000;342:1301–1308.

45. De Backer, D, et al. Does hepato-splanchnic /DO₂ dependency exist in critically ill septic patients? Am J Respir Crit Care Med. 1998;157:1219–1225.

46. Russell, J, et al. Vasopressin versus norepinephrine infusion in patients with septic shock. N Engl J Med. 2008;358:877–887.

47. De Backer, D, et al. SOAP II Investigators: Comparison of dopamine and norepinephrine in the treatment of schock. N Engl J Med. 2010;362:779–789.

48. Vasu, T, et al. Norepinephrine or dopamine for septic shock: A systematic review of randomized clinical trials. J Intensive Care Med. 2012;27:172–178.

49. Bellomo, R, Chapman, M, Finfer, S, Hickling, K, Myburgh, J. Low-dose dopamine in patients with early renal dysfunction: A placebo-controlled randomised trial. Australian and New Zealand Intensive Care Society (ANZICS) Clinical Trials Group. Lancet. 2000;356:2139–2143.

50. Kellum, JA, M Decker, J. Use of dopamine in acute renal failure: A meta-analysis. Crit Care Med. 2001;29:1526–1531.

51. Gregory, J, et al. Experience with phenylephrine as a component of the pharmacologic support of septic shock. Crit Care Med. 1991;19:1395–1400.

52. Morelli, A, et al. Phenylephrine versus norepinephrine for initial hemodynamic support of patients with septic shock: A randomized, controlled trial. Crit Care. 2008;12:R143.

53. Kumar, A, et al. Cardiovascular response to dobutamine stress predicts outcome in severe sepsis and septic shock. Crit Care. 2008;12:R35.

54. Aube, H, Milan, C, Bletery, B. Risk factors for septic shock in the early management of bacteremia. Am J Med. 1992;93:283–288.

55. 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.

56. Sprung, CL, et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med. 2008;358:111–124.

57. Marik, PE, Zaloga, GP. Adrenal insufficiency during septic shock. Crit Care Med. 2003;31:141–145.