914

75 Sepsis, Severe Sepsis, and Septic ShockRobert S. Munford and Anthony F. Suffredini

E  Sepsis

INTRODUCTION AND DEFINITIONSSepsis, severe sepsis, and septic shock are the terms most often used to describe the body’s harmful systemic responses to infection. Lacking a precise biochemical characterization of the syndromes or a certain understanding of their causation, experts have defined them by apply-ing clinical and laboratory findings to a likely framework of pathogen-esis. For 2 decades, most authors and clinicians used definitions developed by a consensus committee of experts in 19921 and revised in 2001.2 The “systemic inflammatory response syndrome” (SIRS) was defined by the presence of two or more acute findings (tachycardia, leukocytosis or leukopenia, fever or hypothermia, tachypnea); if SIRS was thought to be caused by presumed or proven infection, the patient was said to have “sepsis” (from the Greek for “putrefaction”). In the most recent version of consensus definitions, published in 2012 by the Surviving Sepsis Campaign, the SIRS nomenclature was discarded in favor of much more fluid criteria.3 “Sepsis is a systemic, deleterious host response to infection leading to severe sepsis (acute organ dys-function secondary to documented or suspected infection) and septic shock (severe sepsis plus hypotension not reversed with fluid resuscita-tion).” A patient is said to be “septic” if infection is documented or suspected and “some” additional criteria are met (Table 75-1).

The SIRS definitions were derived by experts who assumed that even the early systemic responses to infection, such as tachycardia, leukocytosis, and fever, are inflammatory. “We characterize SIRS as an abnormal generalized inflammatory reaction in organs remote from the initial insult.”4 Many authors criticized these definitions, noting that the criteria for SIRS are both nonspecific and insufficiently predic-tive. The 2001 definitions2 attempted to provide a more comprehensive overview of the clinical manifestations of sepsis by considering predis-posing factors (i.e., comorbidities), infection characteristics (organism, virulence, sensitivity to antimicrobial drugs, site of infection), host responses (clinical and laboratory markers), and types and degree of organ dysfunction (i.e., laboratory and functional correlates). A large retrospective analysis published in 2012 found that both the 1992 and 2001 consensus definitions were highly sensitive but that the

specificity of the newer definition was only 61%.5 The fact that these definitions were still commonly used after almost 2 decades reflects several points: they seemed to describe a clinically-observable con-tinuum,6,7 they were easy to use, and, perhaps most important, a more precise or clinically helpful set of definitions has not appeared. The popularity of the definitions and the changes incorporated into the most recent version reflect an important deficiency: there are still no reliable biochemical (laboratory-based) criteria for knowing when a patient’s systemic responses to infection have become harmful. The operational definitions for “severe sepsis” and “septic shock” have nonetheless become standard nomenclature and will be used through-out this chapter.

EPIDEMIOLOGYEstimates of the incidence of severe sepsis and septic shock are compromised by the absence of both standardized definitions and population-based prospective cohort studies.8 Most of the available estimates are based on hospital discharge diagnoses, which do not use standardized definitions, and the results may differ substantially when different recovery methods are used.9 A review of data in the National Hospital Discharge Survey (United States) found that the incidence of sepsis increased by almost fourfold to 240 cases per 100,000 population per year during the interval 1979 to 2000.10 The incidence was higher in men than in women and in nonwhite persons than in white persons. Over the same period, the in-hospital case-fatality rate for patients with a sepsis-related diagnosis fell from 28% to 18%. A newer analysis, from the Nationwide Inpatient Sample (United States), found that the incidence of severe sepsis increased from 200 cases per 100,000 popu-lation 18 years of age or older in 2003 to 300 cases per 100,000 in 2007, or 711,736 cases in the United States.11 The authors of the report stated that the increase in the number of hospitalizations for severe sepsis likely represents changes in “documentation and hospital coding prac-tice that could bias efforts to conduct national surveillance.”11

Although the median age for patients with a sepsis-related hospital discharge diagnosis is approximately 60 years, the attack rate is very

Definitions• Sepsisisthebody’sharmfulsystemicreaction

tomicrobialinfection.• Severesepsisisorgandysfunctioncomplicating

infection;ifhypotensionispresent,itcanbereversedwithintravenousfluids.

• Septicshockissepsis-associatedhypotensionthatrequirespharmacologicreversal.

Epidemiology• Mostcommonlyaffectedpersonsarethevery

youngandolderadults.• Majoroutcomedeterminantsareageand

comorbidity,especiallyimmunosuppressiveillness.

Microbiology• Virtuallyanymicrobecantriggersepsis.• Mostcasestodayoccurinpreviously

morbidindividualsandarecausedbyopportunistsfromthepatient’sownmicrobiome.

Diagnosis• Usualsigns:tachycardia,tachypnea,

leukocytosisorleukopenia,feverorhypothermia

• Frequent,suggestive:thrombocytopenia,lactatemia,delirium,respiratoryalkalosis,hyperbilirubinemia

Therapy• Promptadministrationofantimicrobialdrugs

thatcankillthepatient’soffendingmicrobe(seeTable75-4)

• Intravenousfluidsandpressorsupportasneeded

• Sourcecontrol:eliminatingthelocalsiteofinfection,usuallywithsurgeryorremovalofindwellingdevice

• Supportivecare

Prevention• Hospitalinfectioncontrol• Vaccination

SHORT VIEW SUMMARY

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

914.e1

KEYWORDSadrenal insufficiency; bacteremia; coagulopathy; delirium; disseminated intravascular coagulation; endotoxemia; endotoxin; glycolysis; lactate; LPS; mitochondria; septic shock; sepsis; severe sepsis

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

915

In surveys performed in intensive care units (ICUs) in the United States and Europe during the 1990s and 2000s, approximately 70% to 80% of the cases of severe sepsis in adults occurred in individuals who were already hospitalized for another reason.14-16 In 30% to 50% of the cases in these and other series, a definite microbial etiology was not found.6,14,16,17 Moreover, the microorganism cultured from blood or an infected local site was often one that does not usually cause disease in otherwise healthy people (in most cases, it was from the patient’s microbiome and may have been acquired during hospitalization),14,17 and in approximately one fifth of the patients, more than one isolate was found (polymicrobial infection).

Respiratory infections most often induce severe sepsis, followed by abdominal and urinary tract infections. Severe sepsis caused by classic bacterial pathogens, such as Neisseria meningitidis or Streptococcus pyogenes, is much less frequently encountered today than is sepsis trig-gered by commensal microbes that infect individuals whose epithelial barriers or other antimicrobial defenses have been compromised by injury or illness. Although for many years gram-negative bacteria were isolated from the majority of culture-positive patients with severe sepsis, the fraction of cases associated with gram-positive bacteria has steadily increased, and now Staphylococcus aureus, coagulase-negative staphylococci, and enterococci account for approximately 30% to 50% of the cases in most clinical series. Another recent trend is the emer-gence of fungi (in particular, Candida spp.) as etiologic agents of severe sepsis; in some recent series, Candida spp. have caused 5% to 20% of the microbiologically documented cases.15,16

PATHOGENESISMuch of what is known about sepsis pathogenesis has been learned using animal models. None of these models completely mimics human responses to infection.18,19 The usefulness of mouse models has recently been challenged by a report that humans and mice have strikingly dif-ferent systemic (blood leukocyte) responses to endotoxin, trauma, and burn injury.20 Mouse models have been extraordinarily helpful for dis-secting the contributions of individual genes and molecular pathways to various immune responses, and the responses of individual cell types (e.g., macrophages) may be quite similar in humans and labora-tory mice. Human and murine systemic responses to acute stress, such as a bolus injection of endotoxin, may be strikingly different in many respects, but not all.21 Understanding these differences is essential for translating the results of animal models into clinical research and practice.

Early Host Responses to InfectionLocal Defenses: Walling Off and Killing Invading MicrobesWhen a microbe breaches an epithelial barrier and enters the underly-ing tissue, it quickly encounters tissue-resident macrophages, mast cells, and dendritic cells. These cells sense the invader and react by secreting mediators that mobilize the local inflammatory response. Their sensory ability is conferred by protein receptors that bind to highly conserved microbial molecules. The Toll-like receptor system for the lipopolysaccharides (LPS) made by some gram-negative bac-teria is perhaps best understood, but others exist for sensing the pres-ence of bacterial peptidoglycan, DNA, lipopeptides, flagella, viral double-standard RNA, and other conserved microbial molecules.22,23 In the case of LPS, a plasma protein (LPS-binding protein [LBP]) can transfer LPS from bacterial membranes to another host protein, CD14, which is expressed on the surfaces of phagocytes. CD14 then passes the LPS to a signaling complex that has two members: an extra-cellular protein called myeloid differentiation protein-2 (MD-2), which binds the lipid A moiety of LPS,24 and the transmembrane receptor protein, Toll-like receptor 4 (TLR4). LPS binding to MD-2 triggers dimerization of TLR4; this transmits the LPS recognition signal to the interior of the cell, where signal transduction and gene tran-scription pathways promote the production and/or secretion of numer-ous molecules that mediate the inflammatory response (Fig. 75-1). These mediators include cytokines (in particular, tumor necrosis factor [TNF], interleukin [IL]-12), chemokines (IL-8, macrophage inflamma-tory protein [MIP]-1α), lipid mediators (prostaglandins, leukotrienes), and other mediators, and they result in the familiar elements of local

TABLE 75-1  Definitions

TERM DEFINITION COMMENTInfection Presence of

microorganisms in a normally sterile site

May be confused with “colonization,” which is the presence of microorganisms on an epithelial surface

Bacteremia Cultivatable bacteria in the bloodstream

May be transient and inconsequential; inconsistent correlation with severe sepsis

Systemic inflammatory response syndrome (SIRS)

Old term for the systemic response to a wide range of stresses. Criteria included two or more of the following:

Temperature >38° C or <36° C

Heart rate >90 beats/min Respiratory rate >20

breaths/min or PaCO2 <32 mm Hg

WBC >12,000 cells/mm3 or <4,000 cells/mm3, or >10% immature (band) forms

A potentially misleading term. The evidence that the body’s early responses to infection cause systemic inflammation is controversial. See “Pathogenesis” in text.

Sepsis The systemic response to infection. If associated with proven or clinically suspected infection, SIRS was called “sepsis.”1

With the exceptions of leukopenia and hypothermia, these changes are among the body’s normal systemic responses to infection and do not necessarily imply a poor prognosis

In clinical parlance, the term septic is often used informally to describe patients with severe sepsis or septic shock

Hypotension A systolic blood pressure of <90 mm Hg, MAP <70 mm Hg, or a reduction of >40 mm Hg from baseline

To be considered sepsis-related, hypotension must have no other cause

Severe sepsis Sepsis associated with dysfunction of organ(s) distant from the site of infection, hypoperfusion, or hypotension. The term sepsis syndrome had a similar definition.

Abnormalities may include lactic acidosis, oliguria, acutely altered mental status, and acute lung injury.2 To be considered severe sepsis, hypotension must be reversible by administering fluids. Organ dysfunction may be defined according to Marshall et al,446 the SOFA score,447 or the Surviving Sepsis Campaign criteria.3

Septic shock Sepsis with hypotension that, despite adequate fluid resuscitation, requires pressor therapy. In addition, there are perfusion abnormalities that may include lactic acidosis, oliguria, altered mental status, and acute lung injury.

If septic shock lasts for >1 hr and does not respond to pressor administration, the term refractory septic shock is often used

MAP, mean arterial pressure; PaCO2, partial pressure of arterial carbon dioxide; SOFA, sequential organ failure assessment; WBC, white blood cell count.

high among infants (more than 500 cases/100,000 population/year), with low-birth-weight newborns experiencing particularly high risk.12 Sepsis-related mortality decreases after the first year of life and then increases gradually with increasing age. Age and comorbidity are major determinants of outcome: in persons with no known comorbidity, the case-fatality rate is less than 10% from age 3 to 5 years through the third decade, after which it slowly increases to reach approximately 60% by the seventh decade.13 In all age groups, mortality is strongly enhanced by comorbid conditions such as cancer, diabetes, and immunosuppression.

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

916

eat, limiting this activity to local tissues minimizes the release of diges-tive enzymes and oxidants into the circulating blood. The major inher-ited mechanisms for killing microbes in the blood are soluble molecules: the mannose-binding lectin and C-reactive protein (CRP) pathways for activating complement, the alternative complement pathway, anti-bacterial proteins (such as bactericidal permeability-increasing protein [BPI]), and natural immunoglobulin M (IgM) antibodies. Increased capillary permeability allows these molecules to diffuse into tissues where there is local inflammation.

In most instances, invading microbes are eliminated by phagocytes, complement, antimicrobial peptides, NETs, and perhaps natural anti-bodies, and the invaded tissue returns to normal. These rapidly-activated, broadly-applicable host defenses are hardwired in the genome. They have been collectively called “innate” immune mecha-nisms to distinguish them from the “acquired” mechanisms that

inflammation: increased capillary permeability and blood flow, infiltra-tion of neutrophils, and pain. In addition, local deposition of fibrin, initiated by the expression of tissue factor on activated macrophages and endothelial cells, helps wall off the infected tissue and, along with vasoconstriction, provides an important impediment to bloodstream invasion.

Although neutrophils circulate in the bloodstream, they carry out phagocytosis largely in tissue spaces, where they can attach to extracel-lular matrix, spread out, get traction, and ingest. They also may die and expel strands of DNA to form “neutrophil extracellular traps” (NETs). Extracellular neutrophil DNA, decorated with antimicrobial proteins, such as histones, myeloperoxidase, cathepsin G, and neutrophil elas-tase, has been shown to kill a wide variety of microbes.25 Because phagocytes may also release reactive oxygen species (ROS; superoxide, hydrogen peroxide) and the other contents of their lysosomes as they

FIGURE 75-1 Mechanisms that sense bacteria and initiate host responses. Several cell-surface Toll-like receptors (TLRs) sense conserved bacterial cell wall components. For example, most bacterial lipopolysaccharides that have a hexa-acyl, bis-phosphorylated lipid A structure that binds to MD-2, inducing a conformational change in TLR4. Via its cytosolic TIR domain, TLR4 initiates transmembrane signaling through two pathways: the MyD88-dependent path that leads to activation of NF-κB and production of tumor necrosis factor (TNF) and the slower MyD88-independent, TRIF-dependent pathway that increases interferon-β (IFN-β) synthesis. The TRIF pathway may be activated after TLR4 has been internalized in an endosome and engaged TRAM (not shown). Peptidoglycan (PG) and bacterial lipoproteins (LP) interact with TLR2, which oligomerizes with TLR6 or TLR1 to form signaling com-plexes that activate the MyD88-dependent pathway. Other MyD88-dependent sensors are TLR5, which recognizes bacterial flagellin; TLR9, which senses DNA with unmethylated CpG motifs; and TLR7 and TLR8, which recognize single-stranded RNA. TLRs 7, 8, and 9 are found within endosomes, as is TLR3, which recognizes double-stranded RNA and signals via TRIF to induce interferon IFN-β production but also can elicit TNF via TRAF2. The members of the TLR family of transmembrane proteins thus sense diverse microbial molecules and use both combinatorial interactions and different, yet overlapping, downstream pathways to activate the inflammatory response. A second family of intracellular sensory proteins exists for components of peptidoglycan. Nucleotide oligomerization domains NOD1 and NOD2 recognize diaminopimelic acid (only found in gram-negative peptidoglycan) and muramyl dipeptide (found in both gram-negative and gram-positive peptidoglycan), respectively, and initiate signaling via RICK to activate NF-κB. Proteins that contain a NOD may have one or more CARDs, linking them structurally to the members of the inflammasome, intracellular “stress” sensors that activate caspase-1, the enzyme that cleaves pro-IL-1β and pro-IL-18 to release the mature cytokines. Not shown are cytosolic sensors for RNA and DNA. See Baccala and co-workers448 for details. AP-1, activator protein-1; ASC, apoptosis-associated speck-like protein containing a CARD; CARD, caspase activation and recruit-ment domain; IL, interleukin; IRF, interferon regulatory factor; LP, lipoprotein; LPS, lipopolysaccharide; LRR, leucine-rich repeat; MD-2, myeloid differentia-tion protein-2 (also called lymphocyte antigen 96 [LY96]); MyD88, myeloid differentiation primary response gene 88; NALP3, NACHT-LRR-PYD domains-containing protein 3; NLRP, nucleotide-binding oligomerization domain leucine-rich repeat–containing receptors; PG, peptidoglycan; PYD, pyrin domain; RICK, receptor-interacting serine/threonine kinase; TIR, Toll/interleukin-1 receptor; TLR, Toll-like receptors; TRAF2, TNF receptor-associated factor 2; TRAM, Toll-like receptor 4 adaptor molecule; TRIF, TIR domain-containing adaptor-inducing interferon-β). (Modified from Ishi KJ, Koyama S, Nakagawa A, et al. Host innate immune receptors and beyond: making sense of microbial infections. Cell Host Microbe. 2008;3:352-363.)

Sensors for nucleic acids

NOD

NOD

NOD

CARD

CARD

CARD

CARD

RICK

NOD1

NOD2

Muramyl peptides

TLR4 MD-2TLR2/1 TLR2/6

LPS

MyD88

TLR3

TLR7, 8, 9

endosome

TRIF

MyD88

PG, LP

Pro-IL1/18

Mature IL-1, IL-18

Caspase-1

PYD

PYDDanger (stress) sensors

ASC

NALP3

Sensors for bacterial wall structures

Proinflammatorymediators

AP-1 NF-κB

IRF1 IRF3/7IFN-α, -β

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

917

that caspase-1 also controls the production of HMGB-1, a proinflam-matory “alarmin,” in mice.38

Early Systemic Responses: Keeping Infection and Inflammation LocalizedMuch evidence supports the notion that the body’s systemic responses to injury, infection, and other stresses generally suppress inflammation within the bloodstream.39 Systemic responses enhance local defenses by providing antimicrobial molecules (Table 75-3) and effector leuko-cytes (neutrophils, natural killer [NK] cells, monocytes), and they prevent systemic damage by minimizing leukocyte-endothelial adhe-sion in uninvolved tissues and neutralizing chemical mediators (e.g., oxidants and proteases) that enter the blood from inflamed sites.

The early systemic responses are controlled principally by the brain and prominently involve the liver. Their regulation has much in common with that of the body’s systemic “flight or fight” reaction to noninfectious stresses. They are important here because they are acti-vated in response to stress and local infection and thus, in most instances, they form the physiologic stage on which sepsis progresses to severe sepsis and shock.

Central Nervous System Regulation of Systemic ResponsesThe central nervous system (CNS) receives news about microbial inva-sion via at least two routes. First, afferent impulses along nociceptive and vagal nerves rapidly transmit signals from infected or inflamed local tissues to the hypothalamus and brainstem, where they can acti-vate the hypothalamic-pituitary-adrenal (HPA) axis, the autonomic nervous system, and the hypothalamic thermoregulatory center. Toll-like receptors have been identified in dorsal root,40 nociceptive,41 enteric,42 and trigeminal43 neurons and shown to respond to LPS in mice.41 Second, bloodborne mediators (IL-1β, TNF, IL-6, interferons [IFNs], prostaglandin E2) can cross the blood-brain barrier and/or be transported passively through the fenestrated capillaries in the cir-cumventricular organs to reach the hypothalamus.44 Remarkably, the

TABLE 75-2  Innate and Acquired Immunity

Innate ImmunitySenses microbes through proteins that bind highly conserved microbial

molecules (e.g., lipopolysaccharide, peptidoglycan)

Hardwired, that is, inherited in the genome; shaped by evolution

Responds rapidly to microbial invasion

Elements: mannose-binding lectin, alternative complement pathway, “natural” antibodies (gene rearrangements present at birth), pattern-recognition proteins (e.g., LBP, MBP), cellular sensors (Toll-like receptors, NODs, inflammasome proteins), the “professional” phagocytes, mast cells, natural killer cells, Th17 cells

Acquired ImmunityRecognizes microbial epitopes using T- and B-cell receptors → cellular and

humoral immunity

Requires gene rearrangements during the life of the individual

Develops slowly after microbial invasion

Protects the body from subsequent exposure to same and some related (cross-reactive) microbes

Is the basis for vaccine-induced immunity

Elements: antibodies, cytotoxic and helper T cells

LBP, LPS-binding protein; MBP, mannose-binding protein; NODs, nucleotide oligomerization domains.

TABLE 75-3  Normal Systemic Responses to Infection and Injury: Presumed Contributions to Host Defense

Leukocytosis Mobilizes neutrophils into the circulation

Tachycardia Increases cardiac output, blood flow to injured tissue

Fever Raises core temperature; peripheral vasoconstriction shunts blood flow to injured tissue. Occurs much more often when infection is the trigger for systemic responses.

Acute-Phase Responses (Categorized According to Possible Roles in Defense)Anti-infective Increases synthesis of complement factors, microbe

pattern-recognition molecules (mannose-binding lectin, LBP, CRP, CD14, others)

Sequesters iron (lactoferrin, hepcidin) and zinc (metallothionein)

Anti-inflammatory Releases anti-inflammatory neuroendocrine hormones (cortisol, ACTH, epinephrine, α-MSH)

Increases synthesis of proteins that help prevent inflammation within the systemic compartment

Cytokine antagonists (IL-1Ra, sTNF-Rs)Anti-inflammatory mediators (e.g., IL-4, IL-6, IL-6R,

IL-10, IL-13, TGF-β)Protease inhibitors (e.g., α1-antiprotease)Antioxidants (haptoglobin)Reprograms circulating leukocytes (epinephrine, cortisol,

PGE2, ? other factors)

Procoagulant Walls off infection, prevents systemic spreadIncreases synthesis or release of fibrinogen, PAI-1, C4bDecreases synthesis of protein C, antithrombin III

Metabolic Preserves euglycemia, mobilizes fatty acids, amino acidsEpinephrine, cortisol, glucagon, cytokines

Thermoregulatory Inhibits microbial growthFever

ACTH, adrenocorticotropic hormone; CRP, C-reactive protein; IL, interleukin; LBP, lipopolysaccharide-binding protein; MSH, melanocyte-stimulating hormone; PAI-1, plasminogen activator inhibitor-1; PGE2, prostaglandin E2; sTNF-R, soluble tumor necrosis factor receptor; TGF, transforming growth factor.

develop more slowly and use gene rearrangements to achieve exquisite specificity in the recognition of foreign molecules (Table 75-2).26 Shaped by evolution, the innate system normally provides effective protection from the host’s commensal flora and from those pathogens that can be sensed via innate immune receptors. The innate defenses are significantly less effective toward those pathogens that escape recognition by the host. For example, Yersinia pestis adapts to grow at mammalian body temperature by modifying its lipid A so that it can no longer be sensed by MD-2–TLR4. Engineering Y. pestis to make it produce LPS that activates host cells via TLR4 rendered the bacteria nonpathogenic in mice.27 Avoiding host recognition by pro-ducing a nonrecognizable lipid A is probably a key feature of the patho-genesis of plague, a disease in which bacteria grow to high density in blood. The same phenomenon has been shown for other gram-negative bacteria, including another designated biothreat agent, Francisella tularensis.28

Remarkably, many innate immune mechanisms may become non-essential for host defense as adaptive immune defenses mature with age.29,30 In contrast, activation of inflammation by innate immune mechanisms plays a major role in the pathogenesis of sepsis through-out life. In addition to the TLR-initiated pathways for mobilizing host defenses, which can be activated by both microbial and endogenous molecules (e.g., high-mobility group box-1 protein [HMGB-1], high-mobility group nucleosome-binding–1 protein [HMGNB-1]),31 and certain glycosphingolipids32), there are networks that sense microbial and/or endogenous ligands via nucleotide oligomerization domain (NOD) proteins (see Fig. 75-1). NOD1 and NOD2 are cytosolic sensors for bacterial peptidoglycan fragments, whereas the constituents of the intracellular “inflammasome” respond to certain endogenous ligands (uric acid and adenosine triphosphate [ATP], for example) and regu-late the release of IL-1 and IL-18 from cells by activating the cleavage enzyme caspase-1. The roles played by the NOD proteins and inflam-masome activation in the pathogenesis of severe sepsis are uncertain. Most patients with inherited deficiencies or mutations in these proteins have experienced local tissue inflammation without shock; monocytes from patients with severe sepsis had low levels of caspase-1,33 and, in another study, monocytes from patients with early sepsis had dimin-ished messenger RNA (mRNA) levels for several inflammasome pro-teins, in keeping with the “deactivation state” noted for many other monocyte genes (see later). On the other hand, one study noted a high frequency of pyrin mutations in Turkish patients with severe sepsis,34 high monocyte pyrin mRNA levels were associated with death in pedi-atric patients,35 and mice deficient in caspase-1 were protected from endotoxic shock.36 High levels of IL-18 were found in the plasma of patients with acute respiratory distress syndrome (ARDS) and were associated with ICU morbidity and mortality.37 Other studies found

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

918

the infected tissue. Enhanced surface expression of CD11b/CD18 on neutrophils57 may promote their adhesion to intercellular adhesion molecule-1 (ICAM-1) on the surfaces of activated endothelial cells. In addition to mobilizing neutrophils into the circulation and delivering them to sites of infection, the acute-phase response also involves increased production of several proteins that bind conserved microbial molecules and assist in recognizing and/or killing microbes (LBP, mannose-binding lectin, CRP, CD14, BPI, and complement compo-nents [C3, C4b]). Lactoferrin release from neutrophils and the produc-tion of hepcidin by the liver promote sequestration of iron in the reticuloendothelial system,58 whereas zinc is retained within cells by binding to metallothionein.

Anti-inflammatory ResponsesThe mechanisms that induce neutrophils to demarginate also seem to inhibit their ability to adhere to noninflamed vascular endothelium, thereby preventing unnecessary accumulation of neutrophils in uninfected tissues. Other responses that may prevent inflammation within the systemic compartment include increases in the blood levels of cytokine antagonists (IL-1Ra, soluble TNF receptors), other anti-inflammatory mediators (epinephrine, cortisol, α-melanocyte-stimulating hormone [α-MSH], adrenocorticotropic hormone [ACTH], IL-4, IL-6, IL-10, IL-13, transforming growth factor-β [TGF-β], CRP), protease inhibitors, and antioxidants.

Another phenomenon that may dampen local or systemic inflam-mation occurs after animals have reacted to certain microbial mole-cules. Known as innate immune “tolerance” or “reprogramming,” it may follow exposure to either bacterial or viral molecules.59,60 In general, a second exposure to the same or another microbial agonist fails to elicit the usual proinflammatory response (production of TNF, IL-1, IL-12), whereas the production of some anti-inflammatory mol-ecules (IL-10, IL-1Ra) is maintained. This adaptation, which generally lasts a few days after the primary infection or exposure, is thought to prevent untoward inflammation. Produced by complex cellular pro-cesses that include chromatin remodeling (gene silencing),61-64 toler-ance may contribute to postinfection immune suppression. At least in the case of LPS, inactivation of the inciting ligand by a host enzyme, acyloxyacyl hydrolase, may be necessary to allow restoration of normal host defenses.65,66 A novel way to reverse tolerance in the late stages of sepsis was recently reported in a murine model of peritonitis: adoptive transfer of CD34+ hematopoietic stem-progenitor cells enhanced bac-terial clearance and improved long-term survival,67 presumably by providing nontolerant immune cells.

In an important experiment, van der Poll and co-workers68 showed how systemic responses may be modulated by the sympathetic nervous system during periods of stress. They infused a bolus of endotoxin into volunteers and then measured blood levels of TNF and IL-10 over several hours. Another group of volunteers received an infusion of epinephrine for 3 hours before the endotoxin bolus. Epinephrine dra-matically shifted the response from proinflammatory (TNF ≫ IL-10) to anti-inflammatory (IL-10 ≫ TNF). Epinephrine-induced repro-gramming of cellular responses to endotoxin can also be demonstrated in whole blood ex vivo; similar reprogramming can be caused by his-tamine, prostaglandin E2, and other agonists that raise intracellular cyclic adenosine monophosphate (cAMP). Infusing hydrocortisone also has dramatic (though somewhat less predictable) effects on the cytokine response to an endotoxin bolus.69 A similar reprogramming of circulating blood cells seems to occur in vivo in response to rela-tively minor stresses.70

There is strong evidence that the body’s inflammatory responses to infection and other stresses are compartmentalized. In patients with pancreatitis, for example, Dugernier and co-workers71 found that the concentrations of proinflammatory and anti-inflammatory cytokines decreased from the peritoneal fluid to the lymph to the blood; net anti-inflammatory activity was measured in virtually all lymph and blood samples, whereas net proinflammatory activity was detected only in ascites. Similarly, in patients with acute appendicitis, TNF and IFN-γ were found in peritoneal fluid but not in plasma, whereas high concen-trations of IL-10 and IL-4 were found in plasma, which inhibited the ability of LPS to stimulate monocytes ex vivo.72 Studies performed in rabbits suggested that, with low bacterial inocula, the proinflammatory

output of three major CNS efferent pathways (the HPA, the sympa-thetic nervous system, and the parasympathetic nervous system) inhibits inflammation within the circulating blood (see later and Table 75-3), and the thermoregulatory center may enhance antimicrobial activity by elevating body temperature.45

The Liver: Essential Roles in Systemic Responses to InfectionThe liver is anatomically situated to remove microorganisms that cross the gut mucosa and enter the portal circulation; whether or not trans-locating microbes or their products enter the blood without first travel-ing through the lymphatic system is uncertain.46 The liver is nonetheless the major filter for endotoxin and for nonopsonized particulate matter, including many microbes, whereas the spleen is the major trap for opsonized microorganisms. Impairment of the hepatic filter (e.g., by cirrhosis) predisposes to bacteremic infections with Vibrio vulnificus and certain other gut bacteria, and individuals with hyposplenism may experience overwhelming bloodstream infections, usually with encap-sulated bacteria (e.g., Streptococcus pneumoniae, Haemophilus influen-zae type b or Neisseria meningitidis). Whereas hepatic macrophages (Kupffer cells) and sinusoidal endothelial cells can extract bacterial LPS from blood and inactivate it,47 they also release mediators that induce hepatocytes to produce many of the body’s acute-phase and metabolic responses to injury and infection. IL-6, which is released by many types of cells in response to injury or other stimuli, is the major trigger for most elements of the acute-phase response; IL-1β may also be an important stimulus.48 Whereas a minor population of circulating monocytes (CD14+CD16+DR++) seems to account for most of the TNF produced by blood cells,49 most of the TNF that circulates in the blood after an intravenous (IV) injection of endotoxin is produced within the splanchnic bed, largely in the liver.50

The liver may also play a role in the sensory system that informs the CNS that microbes have invaded. In rodents, the ability of low intrave-nous doses of endotoxin or IL-1β to induce fever and activate the HPA axis can be blocked by cutting or poisoning the hepatic branches of the vagus nerve.51 Activation of vagal afferents by pyrogens is thought to involve prostaglandin E2.51 Conversely, stimulation of vagus nerve efferents suppresses endotoxin-induced TNF production in rats via a “cholinergic anti-inflammatory pathway” that involves inhibition of macrophage cytokine synthesis by acetylcholine.52 This inhibitory action is most potent in the spleen,53 which lacks cholinergic innerva-tion; in an unexpected twist, it was reported recently that adrenergic stimulation induces T and B lymphocytes to synthesize and release acetylcholine in the spleen and other tissues.54 Because much of the TNF produced in response to intravenous challenge with endotoxin arises from the splanchnic bed,50 which is richly innervated with cho-linergic neurons and hosts choline acetyltransferase–expressing T cells in Peyer’s patches,55 acetylcholine may be very important for regulating systemic responses to microbial agonists within the bloodstream.

The liver is thus a key player in the systemic response to infection—as a blood filter that collects and kills bloodborne microbes and inac-tivates bacterial endotoxins, as a “listening station” that senses low concentrations of circulating cytokines and transmits this information to the CNS, as a factory for the production of many (acute-phase) ele-ments of the systemic response, and as a major site of infection-associated metabolic adaptations.56

Acute Phase ResponsesIt is convenient to think of the acute systemic responses to injury, infection, and other stresses in five categories (see Table 75-2): anti-infective, anti-inflammatory, procoagulant, metabolic, and thermo-regulatory. The scavenging and wound-healing responses will not be discussed here.

Anti-infective ResponsesAcute leukocytosis, which largely reflects the demargination of neutro-phils, is brought about by epinephrine, cortisol, and possibly IL-10 and other mediators. Mobilization of marrow neutrophils by granulocyte colony-stimulating factor (G-CSF) and other cytokines also plays an important role. Circulating neutrophils adhere to inflammation-activated endothelium at sites of infection and move by diapedesis into

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

919

can supply the brain and immune cells until levels of ketone bodies (produced by the liver from fatty acids) and lactate, which can be used for food by the CNS, rise later in the course of starvation or critical illness.78 Treatment of critically ill patients with intravenous glucose solutions and parenteral and/or enteral feeding largely satisfies the body’s need to maintain euglycemia and may induce hyperglycemia in many individuals.

It is important to note that the systemic responses that help restrict inflammation to infected sites and optimize metabolism are carried out by many of the same molecules, most prominently, catecholamines and glucocorticoids. The effects of cortisol on metabolism—promoting gluconeogenesis, glycogenolysis, insulin resistance, and lipolysis—are dose related in a monogenic fashion. In contrast, cortisol’s effects on various aspects of inflammation may be permissive at normal concen-trations (allowing acute-phase protein synthesis, for example) and either suppressive (inhibiting cytokine and acute-phase protein pro-duction) or stimulatory (increasing IL-10 production) at high con-centrations within the physiologic concentration range.79 It is also important that the regulated mechanisms that diminish the impact of glucocorticoids (e.g., receptor downregulation, the actions of macro-phage migration-inhibitory factor [MIF], others) affect both immune and nonimmune cells. In a similar way, the tachyphylaxis or desensi-tization that reduces responses to catecholamines affects both vascular smooth muscle and inflammatory cells.68,80,81 Mechanisms that may have evolved to prevent stress-induced hypertension and dampen sys-temic anti-inflammation may also contribute to the pathogenesis of severe sepsis and septic shock.

Procoagulant ResponsesInflammation and coagulation are closely linked.82,83 Inflammation-induced procoagulant responses contribute to abscess formation and delayed hypersensitivity reactions in humans.

In individuals who have sustained physical trauma, activation of coagulation and inhibition of fibrinolysis occur roughly in proportion to the severity of injury.84 Briefly, inflammation-induced expression of tissue factor on the surfaces of monocytes and endothelial cells is thought to initiate the production of thrombin via factors VIIa and Xa, whereas increased production of plasminogen activator inhibitor-1 (PAI-1) inhibits fibrinolysis (Fig. 75-2).85

cytokine responses to infection are compartmentalized to the lung73 or peritoneum,74 depending upon the site of inoculation.

Another view was suggested by a report that the initial response to major trauma is a “genomic storm” in which mRNA abundance for both proinflammatory and anti-inflammatory molecules is upregu-lated simultaneously in peripheral blood leukocytes.75 The findings—which actually suggested that the body’s acute responses to trauma are well organized and relatively consistent from person to person, whether or not patients subsequently succumb—were cited as evidence to refute the old notion4 that the initially proinflammatory systemic response to injury is followed by a “compensatory” anti-inflammatory response. Because others have found that circulating leukocytes do not produce most of the cytokines found in the blood (see earlier), interpreting the significance of events occurring within peripheral blood leukocytes is difficult; perhaps they could be used as evidence for “immune disso-nance,” another old concept,76 but the reported changes were far from being chaotic, dissonant, or stormlike. Studies of blood leukocytes may provide a misleading view of the body’s overall responses to infection and other stresses.

Metabolic ResponsesSeveral mechanisms maintain euglycemia in the face of stress. Epi-nephrine, glucagon, cortisol, and possibly other hormones stimulate glycogenolysis and gluconeogenesis in the liver; the major precursors for hepatic gluconeogenesis are lactate, which is derived from glucose via glycolysis in immune cells and muscle, and alanine, which is largely produced by transamination of pyruvate in muscle and other tissues. Insulin resistance reduces glucose uptake by muscle and contributes to muscle catabolism.77 The counterregulatory hormones also induce lipolysis (so blood levels of free fatty acids and glycerol increase) and muscle proteolysis. Lipolysis, which occurs principally in adipocytes, involves the actions of adipose triglyceride lipase and hormone-sensitive lipase; lipoprotein lipase is inhibited, accounting in part for the increase in circulating triglycerides that often occurs. Muscle pro-teolysis releases amino acids (alanine, glutamine, and others) that are used for hepatic gluconeogenesis and for producing acute-phase proteins.

A long-accepted rationale for increased glucose production is the requirement to maintain blood glucose levels at concentrations that

FIGURE 75-2 Inflammation-activated coagulation. Inflammation-associated coagulation begins when cytokines, Toll-like receptor agonists, or other stimuli induce tissue factor expression on the surfaces of monocytes and vascular endothelial cells. Increased concentrations of plasminogen activator inhibitor-1 (PAI-1) prevent the formation of plasmin, thus decreasing fibrinolysis. Tissue factor pathway inhibitor (TFPI) modulates thrombin activation by blocking the activity of the tissue factor/factor VIIa/factor Xa complex. Hepatic synthesis of protein C and antithrombin III (AT-III) decreases during the acute-phase response, whereas an increase in the plasma concentration of complement factor C4b binds more of the available protein S, reducing its ability to inhibit clotting. Factors may also be consumed during clot formation. Despite these procoagulant and antifibrinolytic changes, clinically apparent intravascular thrombosis is unusual. IL-6, interleukin-6. (Modified from Levi M, ten Cate H. Disseminated intravascular coagulation. N Engl J Med. 1999;341:586-592. Copyright © 1999 Massachusetts Medical Society.)

Tissuefactor+VIIa

IXa(+ VIII)

Xa(+ V)

Thrombin

Fibrinogen Fibrin

Fibrinolysis

Plasminogenactivators

Plasminogen

PlasminPAI-1

FibrinD-dimers

Fibrin

Inflammatorymediators(IL-6 + )

Low levels ofAT-III, TFPI,

proteins C and S

AnticoagulationThrombin

Fibrin formation Inadequate fibrin removal

Thrombosis of smalland midsize vessels

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

920

view is that cytokines and other molecules, produced either locally or systemically, circulate via the blood and induce injury to the vascular endothelium and/or microcirculation in different organs. A newer view invokes an extension of the body’s normal neuroendocrine responses to stress, in essence, an exhaustion of ATP in critical organs when the inflammatory stimulus is too strong or too prolonged. In both concepts, microvascular derangement and mitochondrial dys-function are thought to play central roles.

Cytokines and Other MediatorsThe discovery of the major proinflammatory cytokines (TNF and IL-1β) was quickly followed by evidence that these proteins, individu-ally or in combination, can induce severe sepsis and septic shock in experimental animals. Moreover, higher-than-normal levels of these cytokines were found in the blood of many septic patients, and, in some studies, these levels correlated with risk of dying. During the 1990s, however, investigators found that the blood of severely septic patients contains not only these and other proinflammatory mediators (e.g., IL-12, platelet-activating factor [PAF]) but also a broad array of anti-inflammatory molecules (e.g., IL-4, IL-10, IL-1Ra, soluble TNF receptors). Remarkably, attempts to define the “dominant” molecules in the plasma of septic patients concluded that the anti-inflammatory cytokines, in particular, IL-10 and IL-4, were most active.93-95 In other studies, peripheral blood monocytes from patients with severe sepsis responded to LPS and other agonists ex vivo by producing lower-than-normal amounts of TNF, IL-12, and IFN-γ, yet normal or high amounts of IL-1Ra and IL-10.96-99

It has been especially challenging to understand the role of IL-6.100 Although early studies suggested that this abundant cytokine is pro-inflammatory, there is also evidence suggesting that it is an “SOS” signal that can be produced by most cells in response to injury. Indeed, epinephrine induces IL-6 production in vivo, whereas IL-6 infusion enhances blood levels of IL-1Ra, IL-10, and cortisol.101 Experiments in genetically manipulated mice suggest that IL-6 is the most important activator of the HPA axis in response to stress,102 and IL-6–deficient mice have exaggerated inflammatory responses to bacterial infec-tions.103,104 Studies in volunteers and in chimpanzees suggest that IL-6 is the major procoagulant cytokine.82 More than any other cytokine, IL-6 seems to direct the body’s systemic acute-phase responses.100,105

Two proinflammatory mediators may become important late in the course of the response to severe infection. MIF is a product of T lym-phocytes and macrophages that is induced by, and opposes the actions of, glucocorticoids.106 Although MIF normally circulates at a low, basal level, its plasma concentration increases during infection and stress, and very high levels have been found in the plasma of patients with severe sepsis.107,108 The role played by MIF in endotoxin-induced sepsis has been disputed, however.107,109 A second “late” proinflammatory molecule is a transcription factor, HMGB-1, that appears in the blood several hours after infection begins and contributes to death in a mouse endotoxin challenge model.110 Because recombinant HMGB-1 has little proinflammatory activity itself, its ability to bind to various bacterial products may be important for its activities in vivo111; it is considered an alarmin and may be critical for LPS-induced responses in mice, for example.112 However, there has been poor correlation between HMGB-1 blood levels and clinical severity or outcome.113,114

For many years, it seemed that TNF and other proinflammatory mediators are produced at a local site of infection, diffuse into the blood stream, initiate systemic inflammation, and are then opposed by a “counterregulatory” anti-inflammatory response. An imbalance in these opposing forces (“immune dissonance”76) was thought to cause severe sepsis. This view found support in the results of numerous studies that measured the cytokine responses of healthy subjects to a bolus injection of endotoxin,115 as well as in observations made in patients with fulminant meningococcemia. It is difficult to envision the same sequence of events in patients who, as a consequence of trauma or illness, are already experiencing acute-phase systemic responses at the time that infection begins. In such patients, systemic anti-inflammatory mechanisms may dominate throughout the clinical course. Indeed, an early anti-inflammatory systemic response, as reflected in a high ratio of IL-10 to TNF in plasma at the time of hospital admission, has been associated with a poor outcome.116,117

Protein C is a natural anticoagulant that is converted to its activated form (aPC) when thrombin binds thrombomodulin, an endothelial cell surface protein. Activated protein C then dissociates from its own receptor, endothelial protein C receptor (EPCR), before binding soluble protein S to produce a complex that inactivates factors Va and VIIIa, thereby blocking the activation of thrombin.86 During acute-phase responses, depletion of protein C and antithrombin III parallels the fall in serum albumin concentration, suggesting that these anticoagulants are negative acute-phase reactants.87 Protein C may also be degraded by elastase. The available molecules of protein S decrease as it binds to its circulating partner, C4b, a positive acute-phase reactant. Paradoxi-cally, generation of low amounts of thrombin may inhibit clotting by activating protein C.

Thermoregulatory ResponsesThe adaptations to stress just described often occur in the absence of fever; in fact, there is evidence that fever and the other elements of the acute-phase response are independently regulated.88 Because TNF, IL-6, and other putative pyrogens can be found in the blood in the absence of fever, and because the evidence that endogenous pyrogens circulate in human blood is surprisingly limited,44,89 it is possible that infection-related thermogenesis is induced when local inflammation activates neural afferent signals to the thermoregulatory center, either via nociceptive neurons40,41,90 or the vagus.51 The physiologic responses that increase body temperature include shivering (rhythmic contrac-tions of skeletal muscles) and redirection of blood flow from the skin and extremities to internal organs by means of vasoconstriction.

An increase in body temperature may favor host survival in several ways.45 It may help inhibit bacterial growth and increase the bacteri-cidal activities of neutrophils and macrophages,91 for example, and the redistribution of blood flow that occurs during thermogenesis may increase blood delivery to infected tissues.

SummaryThere is considerable overlap between the body’s acute systemic responses to infection, injury, and many other stresses. Indeed, many of the same responses occur after strenuous exercise and during periods of psychological stress. As shown best in studies performed in patients who have sustained trauma, the intensity of the response generally increases with increasing stimulus severity,92 and there is individual variability in the expression of its different elements. The evidence that the body’s early systemic responses to stress can be immunosuppres-sive, increasing risk for acquiring viral as well as bacterial infections, is intriguing but still largely circumstantial. This phenomenon may account, in part, for the increased susceptibility to nosocomial bacterial infection and recrudescence of viral infections (e.g., cytomegalovirus, herpes simplex; see later) experienced by patients who have sustained major trauma or are experiencing critical illness.

In most patients in whom severe sepsis occurs today, many of the acute-phase responses discussed here would probably have been acti-vated by injury or illness before the infectious challenge. Local inflam-mation would provide a further stimulus to these responses, broadening and intensifying their expression. If uncontrolled infection then induces severe sepsis and septic shock, it does so when anti-inflammatory influences may dominate in the peripheral blood; as will be discussed later, a net anti-inflammatory advantage seems to be maintained during severe sepsis, and immunosuppression becomes even more prominent.

Harmful Responses to Infection: Severe Sepsis and Septic ShockSevere SepsisIf severe sepsis is organ hypofunction or dysfunction occurring as a result of infection, what is organ “hypofunction” or “dysfunction,” and how does it develop? “Hypofunction” implies an inadequate level of activity, whereas “dysfunction” suggests that organ performance is in some way abnormal. Are harmful changes in organ activity level or function precipitated by specific pathologic event(s) or triggers, or do they develop when normally adaptive stress responses are pushed beyond their ability to be protective? Two general ideas have domi-nated thinking about pathogenesis. The older and more widely held

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

921

noted microthrombi in up to 20% of cases of shock,132 the thrombi were often discontinuous and may have undergone perimortem fibrinolysis. It is conceivable that fibrin deposition and microthrombi compromise blood flow without producing overt ischemia; downstream cells might then survive in a state of hibernation (see later). Of importance, a recent study of histopathologic changes in the heart and kidney did not reveal significant cell necrosis, apoptosis, microthrombus deposi-tion, or ischemic changes.136 Perhaps the strongest evidence against a prominent role for coagulopathy in sepsis-induced organ dysfunction is the fact that three different anticoagulant drugs (tissue factor pathway inhibitor, antithrombin III, activated protein C) did not reproducibly improve organ function or increase survival in patients with severe sepsis.137,138

The best-documented and most dramatic exception is fulminant meningococcemia. Autopsy studies have found meningococci within the lumina of dermal vessels, as well as in endothelial cells, and thrombi have been noted in small vessels of many tissues, most notably the skin and adrenal glands. Arterial thrombosis, especially of midsized to small-sized vessels, also occurs with other etiologies of purpura fulminans and can precipitate digital ischemia and extremity loss, especially when combined with intense vasoconstriction.139 Impressive evidence for early activation of the vascular endothelium was also obtained in a study of volunteers infected with Rickettsia rickettsii, which also infects endothelial cells.140 In contrast, in patients with viral hemorrhagic fevers (e.g., Ebola virus), DIC typically occurs late in the course and the clotting factor abnormalities, triggered in part by expression of tissue factor on infected monocytes, may be surprisingly modest; mucosal hemorrhage may be largely a consequence of platelet dysfunction,141 and there is little evidence that DIC contributes to organ injury.

Activation or Injury of the Vascular EndotheliumThe vascular endothelium is involved in three processes that play major roles in sepsis pathophysiology: vascular tone, vascular permeability, and coagulation.129 The evidence that endothelial activation and/or injury contributes to the observed abnormalities in blood pressure, fluid extravasation, and coagulation is almost entirely derived from studies in endotoxin-challenged animals or from in vitro investiga-tions.142,143 In humans, there is ample clinical evidence that capillary permeability increases in the dysfunctional organs of patients with severe sepsis; it is most obvious in patients with acute lung injury (see later). Similarly, a role for endothelial activation in the initiation or propagation of intravascular coagulopathy seems likely, although direct evidence for this in humans is limited almost entirely to patients with fulminant meningococcemia.86 Mutunga and co-workers144 found that patients with severe sepsis and septic shock had higher concentra-tions of von Willebrand factor–positive (endothelial) cells in their plasma than did healthy human control subjects, suggesting that endo-thelial damage occurs in human sepsis. Although the source of these circulating endothelial cells was not determined, others have found both high circulating levels of immunoreactive von Willebrand factor and decreased von Willebrand factor staining in dermal vessels of septic patients, suggesting a generalized endothelial response to inflammation.145 One important endothelial activator is vascular endo-thelial growth factor (VEGF); in mice, sepsis-induced mortality could be prevented by treatment with soluble VEGF receptor (FLT).146 There is also much interest in angiopoietin-2, which circulates in high levels in patients with severe sepsis and can disrupt the endothelial barrier.147

In one study of muscle biopsies from patients with severe sepsis, endothelial activation was suggested by increased expression of ICAM-1 and by an increased proportion of capillaries expressing P-selectin and E-selectin.148 On the other hand, others have found that the skin microcirculation of patients with septic shock has normal endothelium-dependent responses to vasodilators149 and that circulat-ing levels of soluble adhesion molecules do not correlate well with endothelial activation in skin biopsies.150 There is considerable hetero-geneity in the structure and function of endothelial cells in different tissue beds,151 so it is quite possible that the changes noted in the tissues that are most accessible to examination (skin, skeletal muscle, tongue) are not representative of those in the organs that are affected most significantly during severe sepsis (lungs, kidneys, liver, brain).

To date, no theory proposed adequately accounts for the transition to severe sepsis or septic shock, however. Prospective studies that combine clinical observation with serial measurements of mediator biosynthesis, cellular responsiveness, and tissue metabolism in patients at risk (e.g., previously healthy young individuals who sustain major trauma) might provide the information needed to produce an accurate account.

Other Candidate TriggersSeveral kinds of molecules have been proposed to trigger severe sepsis by circulating in the bloodstream and either eliciting inflammation or disrupting the vascular endothelium. They include proinflammatory cytokines, as discussed earlier; microparticles derived from platelets, vascular endothelial cells, and other cell types118; mitochondrial DNA119,120; S100A8 and S100A9, the most abundant cytoplasmic proteins of neutrophils and monocytes; HMGB-1112; and formyl pep-tides120,121 derived from mitochondria. Higher-than-normal concentra-tions of each of these molecules have been found in the blood of patients with severe sepsis, and each can induce organ dysfunction when injected into experimental animals. With some exceptions,122 the ability of these molecules to induce severe sepsis has been studied almost exclusively in animal models, usually performed in mice, using artificial challenges such as bolus injections of endotoxin or hyperacute peritonitis. Their role in human disease remains uncertain.

Complement ActivationThe ability of normal human serum to kill bacteria is largely conferred by the complement system, which can be activated by antigen-antibody complexes or CRP (classical pathway), certain bacterial surface sugars (mannose-binding lectin pathway), or (lipo)polysaccharides (alterna-tive pathway). There is evidence that each of these pathways can be activated in the serum of patients with sepsis,123-125 and at least two complement proteins may contribute to the septic response. Activation of both the complement and the contact systems is regulated by C1-esterase inhibitor, an acute-phase protein that undergoes proteo-lytic inactivation in patients with severe sepsis125; administration of the inhibitor had only modest effects on the outcome of severe sepsis in a nonhuman primate model,126 yet it had impressive efficacy in models of septic peritonitis in mice.127 Factor C5a is a potent chemoattractant that also can induce vasodilation, increase vascular permeability, and augment the release of granule enzymes from phagocytes.128 The blood neutrophils of humans with early sepsis may lose responsiveness to C5a. Studies in septic rodents suggest that antibody-mediated neutral-ization of C5a or its receptor may prevent death.19

CoagulopathyActivation of the coagulation cascade and inhibition of anticoagulation and fibrinolysis are commonly observed during sepsis.129 In most patients, these changes may simply be extensions of the normal acute-phase response to infection (see earlier), but in others, they may be triggered by vascular endothelial injury or dysfunction either at a local site of infection or more diffusely. The endothelium is pivotal in promoting coagulation via its expression of tissue factor and von Willebrand factor and its association with activated platelets. Key anti-coagulant molecules (antithrombin, protein C, tissue factor pathway inhibitor) also interact with the endothelial surface and can be com-promised during activation of coagulation by inflammatory media-tors.82 In addition, cell fragments or microparticles derived from activated or apoptotic cells may express tissue factor and contribute to the procoagulant state.118,129,130

The extent to which coagulopathy contributes to organ dysfunction in septic humans is controversial.82,83 The occurrence of disseminated intravascular coagulation (DIC) is a strong predictor of death in patients with severe sepsis or septic shock.131 Although there is an experimental basis for suspecting that microthrombi (comprising platelet or fibrin aggregates as well as fibrin thrombi) form in a general-ized fashion when patients develop DIC, numerous autopsy studies have not found a convincing link between fibrin deposition and organ failure.132-136 When thrombosis was found, it was usually associated with an indwelling catheter or device, and hemorrhage was usually more life-threatening than was thrombosis.133 Although some studies

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

922

co-workers166 proposed that multiorgan failure is “an attempt by the body to ensure cell survival in the face of sustained critical illness, with affected cells entering a dormant state analogous to hibernation... .” A cautionary note has been raised by Jeger and co-workers,173 whose review of the literature concluded that “the current data from mostly young and otherwise healthy animals does not support the view that mitochondrial dysfunction is the general denominator for multiple organ failure in severe sepsis and septic shock.” “Whether this is true if underlying comorbidities are present, especially in older patients, should be addressed in further studies.” As they point out, an accurate noninvasive method to measure mitochondrial function in vivo would be a major advance.

ImmunosuppressionSick individuals often lose their ability to exhibit cutaneous delayed hypersensitivity to recall antigens, becoming anergic174; this change is temporally associated with decreased surface expression of class II (human leukocyte antigen [HLA]-DR) and co-stimulatory175 mole-cules on circulating monocytes. When compared with monocytes from normal controls, peripheral blood monocytes from septic patients produce less TNF when stimulated ex vivo with LPS but maintain or increase their production of IL-1Ra and/or IL-10.93,95,97 There is evi-dence that IL-10 both inhibits TNF production97 and promotes intra-cellular sequestration of HLA-DR.176 These changes normalize as patients recover.96

Neutrophils obtained from severely septic patients show reduced IL-1β177 and IL-8178 production when stimulated with LPS or strepto-cocci; similar behavior can be induced in normal neutrophils by treat-ment with IL-10.178 These phenomena resemble, at least in part,179 the state of “tolerance” discussed earlier (“Acute Phase Responses”/ “Anti-inflammatory Responses”). In a study of blood leukocytes obtained from trauma victims, early changes in gene expression gradually resolved over time, with no hint of a “second hit” of inflammation when patients developed complications in the ICU,75 perhaps reflecting such tolerance. Prolonged tolerance also occurred in mice that were exposed to gram-negative bacteria and lacked the host lipase that inac-tivates LPS,65,66 raising the possibility that immunologic recovery from other infectious agents may also require inactivating the microbial stimuli that induce the host’s inflammatory responses.

Hotchkiss and co-workers180,181 have noted extensive apoptosis of CD4 and CD8 lymphocytes, dendritic cells, and B cells in the spleens of patients with severe sepsis who died. Intestinal epithelial cells may also undergo apoptosis.152 These changes were not noted in critically ill but nonseptic patients, suggesting that they do not contribute to the immunosuppressive state that follows injury and other stresses.152 Studies of inflammation-associated apoptosis in experimental animals suggest that glucocorticoids are more important inducers of apoptosis than are TNF or the Fas ligand.182 Hyperactivation of splenic sympa-thetic input may also play a role. How loss of these cells contributes to the pathogenesis of severe sepsis is uncertain, however. Remarkably, in the murine model of septic peritonitis, death could be prevented by administering a caspase-3 inhibitor or overexpressing Bcl-2, an anti-apoptotic protein, in lymphocytes183 or other cells.184 Unlike lympho-cytes, the life span of neutrophils is prolonged by proinflammatory mediators, such as TNF and G-CSF, and shortened by IL-10, which promotes neutrophil apoptosis.185

Septic ShockSeptic shock may have two distinguishable phases. Vasoconstrictive (cold) shock, characterized by low cardiac output and high peripheral resistance, occurs in patients who are hypovolemic; factors that con-tribute to decreased effective intravascular volume include redistribu-tion of blood flow, venous pooling, increased capillary permeability, increased insensible losses, and poor fluid intake. During this phase, the blood pressure is supported by peripheral vasoconstriction. Resto-ration of effective intravascular volume by the administration of fluids is usually followed by vasodilation. Vasodilation is not often seen with acute hemorrhagic shock or cardiogenic shock, and it does not usually occur in patients with shock caused by viral hemorrhagic fevers or hantavirus infections, which cause myocardial dysfunction and pro-found increases in capillary permeability.186,187 The clinical hallmarks

Mitochondrial and Microcirculatory DysfunctionInfection-associated abnormalities in organ function are often revers-ible. Moreover, there is often little or no detectable evidence for cell death in the microscopic appearance of tissues of patients or experi-mental animals who die from severe sepsis.134,152,153 Pathologists have found apoptosis of cells in the spleen and intestine,152 myopathic changes in skeletal muscle,148 and changes in blood vessel morphology (in meningococcal disease, for example86), but significant necrosis does not seem to occur in the major organs,136,152 arguing against a prominent general role for thrombosis or necrotic cell death in the early patho-genesis of organ dysfunction. On the other hand, much evidence points to the importance of microcirculatory dysfunction and abnormal oxygen use, and the long course of recovery and frequent occurrence of long-term sequelae (see later) suggest that biochemical derange-ments may persist much longer than has been generally suspected.

In septic patients, the readily measurable indices of macrocircula-tory function (mean arterial pressure, cardiac output, mixed venous oxygen saturation) often do not parallel the severity of organ dysfunc-tion.154 It is now widely believed that a critical cause of abnormal organ function in severe sepsis resides in the microcirculatory units (arteri-ole, capillary bed, venule) within tissues. Evidence for this hypothesis has been obtained using orthogonal polarization spectral (OPS) imaging of the most accessible muscle, the tongue. When they were compared with control subjects, patients with severe sepsis had signifi-cantly lower vessel density, and the proportion of perfused small vessels was also below normal155,156; in other studies, restoration of the sublingual microcirculation was associated with survival.157 Shunting of blood around “weak” microcirculatory units could account for the maintenance of relatively normal mixed venous oxygen saturation despite apparent tissue dysoxia.158 Mechanisms often invoked to account for changes in the microcirculation include reduced deform-ability of erythrocytes and activated neutrophils, neutrophil aggrega-tion, and microthrombosis.159 There is little evidence that any of these phenomena alters microcirculatory function in humans, however, and the aforementioned changes noted by spectral imaging were reversible by the application of acetylcholine155 or nitroprusside,156 indicating that they were not anatomically fixed. Maldistribution of blood flow at the level of the microcirculation thus may contribute to low oxygen use by affected tissues; paradoxically, tissue oxygen levels are often elevated, suggesting a defect in oxygen use at the cellular level.

When they are activated by exposure to microbial molecules, myo-cytes160 and most leukocytes (alternatively activated macrophages and regulatory CD4 T cells are exceptions) rely upon glycolysis rather than oxidative phosphorylation to provide ATP and maintain membrane potentials.161 This phenomenon is an example of “aerobic glycolysis” or the “Warburg effect,” noted initially in tumor cells: the use of glycolysis to produce ATP despite the availability of oxygen.162,163 As the stimulus increases or persists for long periods, ATP concentrations eventually fall, impairing cell function in part by allowing mitochondrial mem-brane potential collapse.164 This normal response to cell activation is conceptually similar to the state of “cytopathic hypoxia,” proposed by Fink165 to denote “diminished production of ATP despite normal (or even supranormal) po2 values in the vicinity of mitochondria within cells.” Many phenomena might contribute, including diminished entry of pyruvate into the tricarboxylic acid cycle161 and uncoupling of oxida-tion from phosphorylation because of collapse of the proton gradient across the mitochondrial membrane.164 Nitric oxide (NO) and ROS are thought to play major roles in triggering mitochondrial loss and hypofunction.166 The ability of peroxynitrite and NO to activate poly-(adenosine diphosphate [ADP]-ribose) polymerase (PARP) may also be important because PARP rapidly polymerizes cellular ADP and thus robs cells of ATP.167 It is likely that many influences intersect to alter mitochondrial function, including both cell-intrinsic mechanisms (e.g., mitochondrial loss via autophagy [mitophagy]168) and cell-extrinsic stimuli, such as thyroid hormone, cortisol, and hyperglycemia.169

Mitochondrial dysfunction in muscle170 and peripheral blood monocytes171 has correlated with sepsis severity in clinical studies, whereas mitochondrial preservation of tissue ATP concentrations170 and early activation of mitochondrial biogenesis172 have been associ-ated with survival. Noting that cell necrosis has been an uncommon finding in patients who died from severe sepsis,133,136,152 Singer and

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

923

either limited or no associations.202-205 A genome-wide association study (GWAS) evaluated treatment responses in severe sepsis and noted that the prognostic models were dominated by clinical variables.206 It is difficult to take into account all of the many gene interactions and clini-cal factors that influence the outcome of serious infections.

Microbial Triggers for Severe SepsisThe most commonly identified sites of primary infection in patients with severe sepsis are the lungs and the abdomen (Fig. 75-3). No obvious source of infection can be found in approximately one third to one half of the cases in most series.6,137,207 Culture-negative and culture-positive cases have had similar morbidity and mortality rates.6,208

In keeping with the discovery that conserved gram-positive (lipoproteins, lipoteichoic acid) and gram-negative (LPS) bacterial molecules are recognized by distinct receptors on leukocytes, with overlapping yet distinctive downstream signaling pathways (see Fig. 75-1), patients with severe sepsis caused by gram-positive and gram-negative bacteria may have somewhat different cytokine responses. In one study, IL-1β, IL-6, and IL-18 concentrations were higher in the plasma of patients with sepsis caused by gram-positive bacteria209; others have found higher levels of TNF or IL-6 in the plasma of patients with severe sepsis caused by gram-negative bacteria.210,211 Microarray analysis of mRNA from macrophages and whole blood stimulated ex vivo has also identified numerous differences in gene expression in response to gram-positive compared with gram-negative stimuli in most,209,212-214 but not all,215 studies. In septic children, a microarray-derived mRNA pattern in blood leukocytes discriminated successfully between influenza, gram-positive, and gram-negative etiologies.216 Although the clinical features of severe sepsis caused by different microbes are very similar, it is possible that these new molecular sig-natures will provide insights into pathogenesis and be useful both for diagnosis and for directing therapy.

BacteremiaHow local infection leads to multiorgan dysfunction and hypotension is uncertain. One long-favored hypothesis is that uncontrolled local infection eventuates in bacteremia or toxinemia; circulating bacteria or their products then stimulate inflammatory reactions within the vas-culature and distant organs that lead to organ dysfunction and hypoten-sion. Given the long history of this idea, it is surprising that strong evidence for it is actually limited to a few special situations. The most prominent example is fulminant meningococcemia, when N. menin-gitidis bacteria grow in the blood and induce DIC, severe sepsis, and shock. Other examples are the neutropenic or splenectomized indi-viduals who develop overwhelming bacteremia217 and patients with Yersinia pestis,218 Burkholderia pseudomallei,219 nontyphoidal Salmo-nella,220 Bacillus anthracis, or Capnocytophaga canimorsus bacteremia.

These examples are special because the responsible bacteria are usually human pathogens, able to cause disease in immunocompetent,

of vasodilation are decreased systemic vascular resistance and high cardiac output. The following factors may contribute to inflammation-induced vasodilation:• Tachyphylaxis to catecholamines, which diminishes the sensitivity

of vascular smooth muscle to catecholamines administered as pressors80

• The underproduction or peripheral resistance of glucocorticoids, which upregulate adrenergic receptors80

• The underproduction or ineffectiveness of aldosterone188,189

• The production of adrenomedullin, which has vasodilatory actions, increases renal blood flow, and inhibits aldosterone secretion190

• The release of NO from sites of inflammation191 and/or distant vas-cular endothelium

• The absence of the normal baroreflex response that increases circu-lating vasopressin levels (and depletion of neurohypophyseal vaso-pressin stores)192

• The release of platelet activating factor• The activation of potassium (K)ATP channels in arteriolar smooth

muscle cells by hypoxia and lactate193

• The generation of bradykinin, a vasodilator that also increases capil-lary permeability194

Remarkably, inflammation-induced vasodilation can occur when there are very high blood levels of several vasoconstrictor hormones (norepinephrine, epinephrine, endothelin-1, and angiotensin II). The best clue to its pathogenesis may be the ability of treatment with hydro-cortisone or vasopressin to improve the pressor effect of catechol-amines in many patients (see “Therapy”).

The basic mechanisms that compromise essential cell function from hypoperfusion are alterations in cell membrane permeability and mito-chondrial energy production. Hypoxic cells switch to anaerobic gly-colysis and accumulate lactate, hydrogen ion, and inorganic phosphates. Cellular ATP stores decrease because of diminished synthesis, contin-ued consumption, and the actions of ATPases.195 Energy-dependent sodium and calcium pumps in the plasma membrane are affected, resulting in the loss of cellular potassium and the accumulation of sodium, calcium, and water.196 This is apparent histologically as gener-alized cell swelling. In addition, protein synthesis is impeded, ribo-somes detach from the endoplasmic reticulum, and mitochondrial and lysosomal membranes are damaged. Protein misfolding further con-tributes to cell injury and death.196 High conductance, nonselective permeability channels develop in the mitochondrial inner membrane (the mitochondrial permeability transition); the loss of membrane potential impedes oxidative phosphorylation and allows leakage of cytochrome C into the cytoplasm, compromising electron transport and serving as a major signal to initiate apoptosis.197 Increases in intra-cellular calcium may initiate cell injury and an imbalance between free-radical generation and radical-scavenging systems may enhance oxidative stress. Reactive oxygen species (i.e., superoxide anion, hydro-gen peroxide, and hydroxyl ions) can accelerate cell injury via lipid peroxidation, damaging plasma and organelle membranes. Further, oxidative modification of proteins enhances their degradation by the proteosome and may result in the loss of vital protein components.

Infection Susceptibility and Outcome: Genetic InfluencesA major obstacle to understanding the pathogenesis of severe sepsis is the striking heterogeneity of the patient population that experiences the syndrome. Patients may differ in age, sex, ethnic group, underlying disease, inciting microbe, medications, and numerous other variables. Genetic epidemiology studies suggest that genetic variation contrib-utes to both susceptibility and outcome in infectious diseases.198 Many polymorphisms or haplotypes have been associated with meningococ-cal disease and/or pneumococcal disease.199,200

Several groups have described single nucleotide polymorphism (SNP)-outcome associations in small groups of critically ill patients with sepsis caused by diverse pathogens. However, a systematic review of the quality of 147 genetic association studies found that most of the reported SNP-sepsis associations were not reproducible.201 Further, studies using meta-analysis of polymorphisms of different inflamma-tory molecules (i.e., PAI-1, TLR4, TNF, lymphotoxin-α [LT-α]) and their relationship to the development of sepsis or survival has revealed

FIGURE 75-3 Presumed sites of infection in patients with culture-positive severe sepsis. Bars show the means of data from four studies.17,315,449,450 Brackets show the minimum and maximum values reported. Note that the lung and abdomen are the most common primary sites. IV, intravenous.

Site

of i

nfec

tion

(% o

f cas

es) 50

40

30

20

10

0

Lung

Abd

omen

Urin

ary

trac

t

Sof

t tis

sues

Oth

er

IV c

athe

ter

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

924

EndotoxemiaIt is also widely held that bacterial endotoxin, when it enters the bloodstream, triggers systemic inflammation. Fulminant meningo-coccemia has provided the most striking evidence for this idea; the high levels of circulating endotoxin found in patients with this syn-drome have correlated directly with plasma TNF levels and the risk of dying.242-244 Infusing endotoxin directly into the bloodstream of healthy volunteers readily confirms the ability of this substance to induce dramatic changes in circulating mediator concentrations and to cause symptoms.245,246

Fulminant meningococcemia is a very distinctive syndrome, however, in which a pathogenic bacterium invades a healthy individual and grows within the vasculature before the victim can mount an effec-tive defense. In most instances, there is no local (i.e., nasopharyngeal) inflammation to impede bacterial invasion or induce systemic responses. Moreover, meningococci can grow to very high density in the blood244 and shed membrane blebs that contain endotoxin and other molecules; these particles may serve as surfaces for activating complement and coagulation within the bloodstream. Meningococ-cemia may thus be a poor model for the typical case of endotoxemia, in which a commensal such as E. coli or Klebsiella invades an already-sick patient, triggers inflammation in a local tissue, and transiently invades the bloodstream to achieve low-level bacteremia. The endo-toxin infusion model resembles meningococcemia in important ways (healthy subjects, acute exposure to endotoxin, no local infection/inflammation), and it may thus be a better model for meningococce-mia than for other forms of endotoxemia.

For 4 decades, the Limulus amebocyte lysate (LAL) assay has been used to measure endotoxin in plasma. This method and a newer antibody-based assay247 measure both active endotoxin and inactive endotoxin that may circulate in plasma. It is thus uncertain that the plasma endotoxin detected by using these assays is truly able to induce inflammation in vivo. Moreover, there is now evidence that much of the endotoxin produced in tissues may be inactivated by acyloxyacyl hydrolase before it reaches the blood,248 and several other endotoxin-neutralizing mechanisms have increased activity in the plasma of sick humans.249-251,252 The factors that promote the binding of endotoxin to lipoproteins, which sequester the lipid A moiety and prevent cell acti-vation, are more effective.251 For example, and blood levels of BPI, a neutrophil-derived, endotoxin-neutralizing protein, increase.253 The low amounts of endotoxin that circulate in septic patients may thus be unable to stimulate cells within the systemic compartment. In addition, even relatively mild stress induces changes in circulating monocytes that “reprogram” them to produce less TNF when they are stimulated by endotoxin.70 Finally, the high concentrations of LBP that are found in plasma during the systemic response to infection may inhibit the ability of LPS to activate monocytes, possibly by preventing LPS trans-fer from CD14 to MD-2–TLR4.254,255

It is thus not surprising that the correlation between endotoxemia, gram-negative bacteremia, and severe sepsis has been inconsistent and often weak.256-258 This does not mean that endotoxin contributes little or nothing to the pathogenesis of severe sepsis, however. In most patients who develop severe sepsis caused by a gram-negative bacterium that produces hexaacyl LPS (and thus can be sensed by MD-2–TLR4),259 the major site for the stimulatory action of endo-toxin may be in an infected extravascular tissue, not the circulating blood. It is also possible that endotoxin is more active in the blood of acutely infected, previously healthy patients, in whom the enhanced endotoxin-inactivating mechanisms cited previously have yet to be induced and in whom circulating endotoxin has been associated with poor outcome.243,256

Other Bacterial ToxinsStaphylococcal and streptococcal toxic shock syndrome toxins (TSST-1, streptococcal pyrogenic exotoxins, streptococcal mitogenic exotoxin Z260) are superantigens that can activate large numbers of circulating T cells to release cytokines. They do so by cross linking major com-patibility complex (MHC) class II molecules on antigen-presenting cells with T-cell receptor Vβ domains, thus triggering the T cell to release proinflammatory cytokines.261-263 Although there is strong circumstantial evidence that these toxins play a central role in

previously healthy individuals. In contrast, the great majority of the cases of severe sepsis occur in previously ill persons and are associated with bacterial or fungal microorganisms that are acquired from the patient’s own microbiota.221 These commensal microbes—enteric gram-negative bacilli, coagulase-negative staphylococci, enterococci, Candida species, and others—infrequently cause disease in humans who have normal immune defenses. Individuals who develop serious disease caused by a commensal bacterium generally have a significant immune defect, most often, epithelial barrier disruption (e.g., cathe-ters, bites, cuts), obstruction of a drainage conduit, or immunosuppres-sion.221 Accordingly, it has been difficult to identify virulence factors in these bacteria that promote bloodstream invasion per se.222-224 Even geography and ethnicity may be important; for example, Yu and co-workers225 found that 94% of Klebsiella pneumoniae blood isolates from immunocompetent patients who developed community-acquired pneumonia in Taiwan and South Africa had a mucoid phenotype, whereas this association did not occur in patients with hospital-acquired K. pneumoniae or in patients from five other countries.

The evidence that circulating commensal gram-negative bacteria stimulate inflammation within the blood stream is limited to a few observations. Blood cultures were positive more commonly in patients who have severe sepsis than in those with sepsis,6 for example, and the fraction of patients who have a positive blood culture is even greater among those with septic shock.6,14 In patients with severe sepsis and documented infection, moreover, bacteremia has been associated with early mortality.17 Bloodstream infection with certain microorganisms, such as Candida albicans,226 methicillin-resistant S. aureus (MRSA),227 and vancomycin-resistant Enterococcus faecium,228 has also been asso-ciated with significant attributable mortality (an often-used but imper-fect surrogate outcome for severe sepsis/septic shock), and one study found that catheter-associated bacteremia was more often accompa-nied by hypotension when the infecting agent was a gram-negative aerobe.229

Other lines of evidence suggest that circulating commensal bacteria may not directly trigger severe sepsis or septic shock.208 First, bactere-mia is usually low grade and transient, provided that reseeding of the bloodstream does not occur. The body’s innate immune mechanisms for clearing commensal bacteria and fungi from the bloodstream are evidently very effective. Second, with some exceptions (such as S. aureus bacteremia, meningococcemia, plague218 and septicemic meli-oidosis219), the risk for developing severe sepsis has not correlated directly with the density of cultivatable bacteria in the blood. Third, bacteremia has no distinctive clinical features—at the bedside, bactere-mic patients with severe sepsis are indistinguishable from those whose cultures are negative230,231; a diagnostic algorithm that identified 88% of patients with bacteremia as a complication of community-acquired pneumonia was only 53% specific.232 Finally, the case-fatality rates for culture-positive and culture-negative patients with severe sepsis and septic shock have been very similar,6,17,233 suggesting that bacteremia may contribute little to outcome. In keeping with this idea, little or no excess mortality could be attributed to bacteremia associated with indwelling vascular catheters,234-236 antibiotic-resistant nosocomial bacteria,237 transfusion-related Serratia,238 nosocomial Enterobacter,239 or with bacteremia that occurred in a tertiary hospital population.240

A second hypothesis might account for the occurrence of severe sepsis in patients who have primary extravascular infections with com-mensal organisms (see discussion and references within Munford208). It holds that inflammatory mediators produced in the infected tissue activate the CNS by stimulating local nerves; when they enter the bloodstream, these mediators also may stimulate cells in the vascula-ture or in distant organs. In such patients, circulating bacteria would be a marker for uncontrolled local infection, not the direct trigger for severe sepsis.

The conclusion reached by Felty and Keefer241 in their 1924 review of patients with Escherichia coli bacteremia seems valid for many patients today: “We are inclined toward the view not only that the symptoms of the generalized infection are difficult to differentiate from those of the local process, but also that the prognosis depends largely on the character of the primary focus. In short, the essential feature is the extent, severity and amenability to treatment, of the local process rather than the sepsis itself.”

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

925

versus intravascular infection, infection with commensal versus patho-gen, or others? Answers to these questions may become possible when a quantitative description of the underlying biochemical mechanisms of severe sepsis and septic shock is achieved.

RECOVERY MECHANISMSRecovery follows successful host defense so regularly that its mecha-nisms have attracted little investigation, yet the time required for recovery is by no means predictable after host defenses have failed (see “Prognosis”). Renewed interest in understanding recovery has fol-lowed the discovery of novel arachidonic acid derivatives (lipoxins, resolvins) that help resolve inflammation in tissues. Anti-inflammation (preventing inflammation-induced damage) and resolution (clearing the battlefield and promoting return of homeostasis) are overlapping yet distinct processes with somewhat different regulatory mecha-nisms.268 Tissue resolution mechanisms include neutrophil apoptosis, macrophage emigration, efferocytosis of dead cells, and the production of lipoxins, resolvins (and other lipids), proteases, and gaseous signals that promote restoration of homeostasis in tissues.269 Little is known about how these processes are regulated during recovery from infec-tion, either in local tissues or in the blood and uninfected organs.

What happens when a sick patient “turns the corner” and begins to improve? What turns off the body’s anti-inflammatory (immunocom-promising) responses? Do hypofunctioning organs only recover when the inflammation-inducing stimulus has been removed? If killing the inciting microbes is necessary, is it sufficient, or must potent microbial signal molecules, such as LPS, also be inactivated?65,66 Does severe sepsis induce epigenetic or other changes that alter host physiology long term64,270? What is the nervous system’s role in recovery? Are late deaths and disability caused by damage incurred during the acute crisis,271 failure of some key recovery mechanism(s), or other pro-cesses? These are among the questions that await future students of the body’s harmful responses to infection. Answering them may uncover new ways to improve outcome.

CLINICAL MANIFESTATIONSPatients with severe sepsis and septic shock experience derangements in both of the body’s major communication networks, the nervous system, and the blood. The function of every organ may be affected.

Nervous and Neuroendocrine SystemsCerebral FunctionIndividuals who experience relatively mild infectious illnesses may exhibit subtle abnormalities in cognitive performance. It is therefore not surprising that confusion and other alterations in higher cerebral function are often early manifestations of severe sepsis, particularly in older adult patients, or that the severity of these changes has correlated with patients’ overall severity of illness.272

Sepsis-associated encephalopathy (SAE) is defined as “diffuse cere-bral dysfunction that accompanies sepsis in the absence of direct CNS infection, structural abnormalities, or other types of encephalopa-thy.”273 Delirium is the most common clinical manifestation, occurring in 30% to 50% of patients with severe sepsis. In general, the severity of SAE parallels the severity of other manifestations of sepsis. Because there are no specific markers for SAE, the diagnosis relies upon exclud-ing primary CNS infections and other causes of encephalopathy. Pro-posed underlying mechanisms include microscopic brain injury, blood-brain barrier and cerebral microcirculation dysfunction, altered CNS metabolism, and impaired cholinergic neurotransmission. One study found white matter lesions, suggesting increased blood-brain barrier permeability, in five of nine septic patients evaluated using magnetic resonance imaging274; a more recent survey concluded that prolonged delirium may be associated with smaller brain volumes and long-term cognitive impairment.275 Another analysis found that patients who experience severe sepsis may have cognitive and func-tional defects that last for years.276

Hypothalamic-Pituitary-Adrenal AxisThe normally pulsatile pattern of pituitary hormone release (growth hormone, ACTH, prolactin) is often blunted in critically ill patients, as is the normal circadian variability in levels of cortisol, leptin, IL-6, and

producing the gram-positive bacterial toxic shock syndromes, proof of causation has not been attained. There is also evidence that bacterial cell wall lipoproteins, which can signal cells via TLR2, may circulate in the blood of septic animals and provoke inflammation.264

SummaryThe evidence reviewed here suggests that severe sepsis is a hetero-geneous disorder of tissue metabolism in which an altered micro-circulation plays a major role. Because tissue metabolism and the microcirculation are normally regulated via peripheral nerves and cir-culating hormones, it seems likely that neuroendocrine derangements are closely tied to organ dysfunction and septic shock. The most proxi-mal cause(s) remain unknown, however, and how the phenomena discussed here (e.g., complement activation, coagulopathy, mediator action or desensitization thereto, endothelial injury, microcirculatory dysfunction) interact to produce the syndromes is uncertain. On the other hand, it is possible to draw tentative conclusions regarding the interactions between certain microbial triggers and the human host. Three scenarios illustrate the spectrum:1. Opportunistic commensal bacteria typically invade across dis-

rupted epithelia, often into hosts in whom immunosuppressive acute-phase responses are already occurring because of illness, injury, or infection. A vigorous local inflammatory response, usually initiated by host sensing of conserved microbial molecules,265 is unable to kill the bacteria because of mechanical failure (obstructed drainage pathway), immunosuppression (neutropenia, “endoge-nous immunosuppression”), or other factors (including bacterial virulence determinants). These bacteria invade the bloodstream when local defenses are unable to kill or contain them; bacteremia, when it occurs, is often transient and may be less important than locally-produced mediators as a trigger for severe sepsis and septic shock.208 Outcome is strongly related to the patient’s underlying physiologic fitness. In the acute management of these patients, a diligent search for the primary focus of infection is essential.

2. At the other end of the spectrum are pathogenic microbes that can survive and multiply in previously healthy humans. They can invade without eliciting clinically significant inflammation other than, in some cases, pneumonia, lymphadenopathy, or a lesion at a cutane-ous entry site. If their growth is not controlled by innate immune defenses, the microbes may enter the bloodstream, infect vascular endothelial cells and/or blood cells, and release toxins or other molecules that stimulate inflammation or induce damage within the blood and tissues. The circulating microbes may provoke both shock and profound coagulopathy that not uncommonly results in hemorrhage and/or arterial thrombosis. Examples include S. pneu-moniae, N. meningitidis, R. rickettsii, Y. pestis, Salmonella Typhi, B. anthracis and probably V. vulnificus and C. canimorsus. With many of these, the absence of an early proinflammatory (local) host defense is an important key to pathogenesis.27,265 Certain viruses may also be in this category; there is evidence that symptomatic infection with filoviruses (e.g., Ebola), for example, which invade without provoking local inflammation and infect monocyte-macrophages in many tissues, may be prevented by an early proin-flammatory systemic response.266 Recent research has associated inherited defects in innate and/or acquired immune function with susceptibility to some of these pathogens.267

3. Other stimuli, such as gram-positive bacterial superantigens, may be produced by extravascular bacteria and diffuse into the blood or be released into it by circulating bacteria. They activate T lympho-cytes in the blood and tissues to release cytokines; in poorly under-stood ways, these cytokines induce organ dysfunction and cause shock.262,263

Although each of these microbe-host interactions leads to the syn-dromes known now as severe sepsis and septic shock, they are suffi-ciently different from one another that they force the question: Is the apparent continuum from sepsis to septic shock truly a single process, a “final common path” that can be induced by many different initiating events, or do different microbe-host interactions produce severe sepsis and septic shock in different ways? If the latter, which are the most important determinants: susceptibility genes, underlying disease, age, physiologic state at the time infection occurs, primary extravascular

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

926

necrosis; a mononuclear cell infiltrate may be present.295 Glucocorti-coids and neuromuscular blocking agents may contribute to the patho-genesis and further complicate the clinical picture.296

Blood CompartmentBlood CellsThe composition of the circulating blood has usually been viewed through a limited window, the peripheral vein. A neutrophilic leuko-cytosis is the normal response to bacterial or fungal infection. It is produced by mobilizing neutrophils from the marginal pool as well as the marrow (see acute-phase and anti-infective responses under “Pathogenesis”). Failure to mount a neutrophilic leukocytosis has been associated with a poor outcome.

At the onset of sepsis, peripheral blood lymphopenia occurs, reflecting diminished numbers of T, B, and NK cells.297-299 The apparent effects of sepsis on different T-lymphocyte subsets has varied among studies, resulting in part from differences in the timing of blood sam-pling, patient selection, and presentation of the data as either absolute cell numbers (usually decreased) or relative percentages of the total lymphocyte population (often unchanged). At least in part, the reduc-tion in circulating CD4+ cells may be due to apoptosis.298,300 In contrast, the numbers of circulating B-lymphocytes may increase301 despite apoptotic cell death. Cytokine production by circulating T cells often has had a helper T-cell 2 (Th2) predominance.302 T-regulatory lym-phocyte (CD4+CD25+Foxp3+) numbers have been unchanged or increased.303,304 The percentage of T-helper lymphocytes producing IL-17 increases, and the fraction that produces IFN-γ decreases at the onset of sepsis; this pattern may reverse after 1 week.304 NK-cell (CD3+, CD56+) absolute numbers fall and their IFN-γ production is reduced.299

Monocyte numbers do not change substantially, but their cellular function is altered. Decreased cytokine responses and cell-surface expression of HLA-DR are common and have been used as biomarkers of immunosuppression in severely ill patients.305 Increased expression of CD163 and CD206, markers of alternative activation phenotypes, occurs within the first week of severe sepsis.304

Thrombocytopenia is a frequent finding in patients with severe sepsis.83,207 Although thrombocytopenia often accompanies DIC, it may be the only routinely measured clotting parameter that is abnor-mal; on the other hand, many patients with low-grade DIC do not have thrombocytopenia. The basis for isolated thrombocytopenia in septic patients is probably multifactorial, with peripheral nonimmune destruction,306 hemophagocytic histiocytosis,307 and marrow suppres-sion playing variable roles.

Plasma LipidsStriking changes occur in the circulating lipids and lipoproteins.308 High-density lipoprotein (HDL) and low-density lipoprotein (LDL) levels decrease, whereas triglyceride, free fatty acid, and very low-density lipoprotein (VLDL) levels increase. The decrease in serum cholesterol is almost entirely accounted for by lower concentrations of cholesterol esters in circulating HDL and LDL.251

GlucoseHypoglycemia is a relatively uncommon manifestation of sepsis. Although many of the reported cases have occurred in patients with hepatic or renal disease or malnutrition, hypoglycemia has also been observed in patients with no definable cause other than sepsis.309 The pathogenesis of hypoglycemia is not well understood, but adrenal insufficiency should be considered in such patients. Moderate or severe hypoglycemia was associated with an increased risk of death from distributive shock in patients who received either intensive or conven-tional glucose control.310 The body’s acute metabolic responses to infec-tion maintain the blood sugar concentration through gluconeogenesis, glycogenolysis, and insulin resistance (see acute-phase and metabolic responses under “Pathogenesis”); hyperglycemia may result, especially in diabetics or when glucose-containing fluids are administered.

LactateIncreased blood lactate concentrations and an increased lactate-to-pyruvate ratio are often seen in patients with severe sepsis, even in the

other hormones. As patients develop septic shock, high plasma con-centrations of vasopressin are followed by relatively low levels, possibly reflecting both loss of baroreflex feedback regulation and vasopressin depletion from the posterior pituitary.192,277,278 The diagnosis and sig-nificance of “inappropriately low” plasma vasopressin levels have been controversial.272

Adrenal InsufficiencyActivation of the hypothalamic pituitary adrenal axis is essential for survival from severe stresses. Adrenalectomized animals are unable to survive septic or traumatic shock without corticosteroid replace-ment. Rarely, infectious agents cause primary adrenal insufficiency by directly inducing adrenal hemorrhage or necrosis. The most frequently implicated microbes are N. meningitidis, Mycobacterium tuberculosis, cytomegalovirus (CMV), and Histoplasma capsulatum. CMV-related adrenalitis has been common in patients with end-stage human immu-nodeficiency virus (HIV) infection, but its significance has been uncer-tain. Among the other factors that may contribute to hypoadrenalism in septic patients are hypoperfusion, cytokine-induced dysfunction of the adrenals, drug-induced steroid hypermetabolism (rifampin, phe-nytoin) or inhibition of steroidogenesis (ketoconazole, etomidate), and desensitization to glucocorticoid responsiveness at the cellular level.279,280 Adrenal suppression by prior therapy with glucocorticoid or megestrol should also be considered. Secondary adrenal insufficiency, caused by pituitary infection or apoplexy, is quite rare. Adrenal respon-siveness to ACTH infusion typically returns to normal in patients who recover from septic shock.281

Traditionally, adrenal insufficiency has been diagnosed if the plasma cortisol level was less than 10 µg/dL in the setting of significant stress or if the level did not increase more than 9 µg/dL in response to ACTH stimulation (250 µg synthetic ACTH [1-24] [cosyntropin]).282-284 Unfortunately, these criteria do not adequately measure the reversible dysfunction of the HPA axis that occurs in critically ill septic patients. A clinical entity, critical illness–related corticosteroid insufficiency (CIRCI), has been proposed to describe this state, which reflects inad-equate cellular corticosteroid activity resulting from adrenal insuffi-ciency, tissue corticosteroid resistance, or both.285 CIRCI is associated with exaggerated and prolonged inflammatory responses to infection. Whereas the aldosterone response to exogenous ACTH seems to be maintained in most patients with severe sepsis,284 a state of hyperrenin-emic hypoaldosteronism has been described in critically ill individuals, most of whom have been hypotensive.188,189

The textbook manifestations of adrenal insufficiency (hyponatre-mia with hyperkalemia, hypothermia, eosinophilia, hyperpigmenta-tion, nausea, vomiting) are not often attributed to adrenal dysfunction in septic patients. Hypotension and hypoglycemia may be the most commonly recognized manifestations.

Autonomic DysfunctionHeart rate variability286 is influenced by the balance of vagal and sym-pathetic inputs to the sinoatrial node. Autonomic reflexes can modu-late these inputs, as can the central (vasomotor and respiratory centers) and peripheral (arterial pressure and respiratory movement) oscilla-tors. Studies have found that abnormalities in heart rate characteristics, measured using spectral analysis, precede (in neonates287,288) or coin-cide with (in adults289) the onset of septic shock, and that they may predict in-hospital mortality in some settings.290 Although the precise basis for these changes is uncertain, in general they seem to reflect a decrease in sympathetic input to the cardiac pacemaker. They may reflect an uncoupling of the biologic oscillations in heart rate, blood pressure, respiration, temperature, and other functions that are nor-mally connected through neural networks.291,292

Peripheral Nerves, MusclesCritical illness polyneuropathy and myopathy may occur in patients who have been ill for a week or more.293 The clinical features include difficulty in weaning from a ventilator, generalized wasting of the limbs, and diffuse weakness (tetraparesis). The diagnosis is usually made when electromyographic examination294 reveals denervation potentials compatible with axonal polyneuropathy, predominantly of distal motor fibers. A muscle biopsy may show edema, atrophy, and

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

927

and (4) hypoxemia. The descriptive term “acute lung injury” has been supplanted by new definitions (the Berlin Definition): ARDS is classi-fied as mild (200 mm Hg < partial pressure of arterial oxygen (Pao2)/fractional inspired oxygen (Fio2) < 300 mm Hg), moderate (100 mm Hg < Pao2/Fio2 ≤ 200 mm Hg), or severe (Pao2/Fio2 ≤ 100 mm Hg).322 The underlying pathology is diffuse alveolar epithelial injury, with increased barrier permeability and exudation of protein-rich fluid into the inter-stitial and airspace compartments.321 Elevated plasma angiopoietin-2 levels have been found to presage onset of ARDS in high-risk patients,323 supporting an important role for endothelial injury in pathogenesis. Neutrophils and monocytes accumulate in the lungs and may form cellular aggregates in pulmonary vessels. Significant right-to-left shunting occurs. Dead space volume increases and compliance decreases, augmenting the work of breathing and often necessitating mechanical ventilation. Indeed, a common indication for mechanical ventilation is respiratory muscle fatigue; in patients who are obtunded or have impaired gag reflexes, intubation may also be used to prevent aspiration of oropharyngeal or gastric contents.

Patients who recover from ARDS may have significant functional impairment related to the healing process, which can produce restric-tive defects and diminish diffusing capacity.

Renal DysfunctionSevere sepsis is often accompanied by azotemia and oliguria. The renal abnormalities range from minimal proteinuria to profound renal failure; postmortem studies have found focal acute tubular injury and minimal glomerular damage.136 The pathogenetic mechanisms include hypovolemia, hypotension, renal vasoconstriction, and toxic drugs (in particular, aminoglycosides).324 In most cases, sepsis-induced renal injury is largely reversible. On the other hand, sepsis can also occur in patients who have acute kidney injury of other etiologies and then acquire nosocomial infection. Predictors of sepsis after acute kidney injury included oliguria, higher fluid accumulation, higher severity of illness score, nonsurgical procedures after acute kidney injury, and dialysis.325 Sepsis that follows acute kidney injury is associated with high mortality and a high rate of dependency on dialysis at the time of hospital discharge.

Gastrointestinal Tract InjuryPeripheral vasodilation redistributes the cardiac output so that visceral organs are underperfused; the morbidity and mortality of septic shock have correlated with the degree of tissue (e.g., gastric) hypoperfusion. Several mechanisms have been proposed for gut failure that results in sepsis and multiple organ hypofunction: disruption of an intact intes-tinal epithelium, reperfusion injury, and translocation into the blood-stream via mesenteric lymphatics of bacteria, bacterial products, and inflammatory mediators.326 One study of critically ill patients in an ICU

absence of shock. Contrary to long-standing dogma, the accumulation of lactate and pyruvate in the blood is not simply a consequence of limited tissue oxygenation (hypoxia).311 Rather, there is evidence that it results from the marked increases in aerobic glycolysis that occurs in muscle, phagocytes, and other cells (see earlier), triggered by cytokine-induced glucose uptake and/or catecholamine-stimulated increases in Na+-K+ pump activity.312-314 Increases in blood lactate may also result from decreased lactate clearance by the liver, mitochondrial dysfunction, and (respiratory) alkalosis, which decreases lactate uptake by cells. Correction of hypotension with vasopressors does not always correct lactic acidosis, possibly because tissue perfusion remains com-promised by vasoconstriction or abnormal microcirculatory vasoregu-lation. Intravenous ibuprofen significantly lowered blood lactate levels in one clinical trial without improving outcome.315

Clotting FactorsThe prevalence of DIC increases as the inflammatory response intensi-fies,6 reaching approximately 30% to 50% in patients with severe sepsis. The patient’s underlying condition (e.g., infection, solid cancers, hema-tologic malignancies, obstetric diseases, trauma, liver disease) can influence diagnostic laboratory tests. In addition, all of these condi-tions can be complicated by the development of sepsis, making the diagnosis and treatment of DIC dependent on the clinical context rather than on any one specific laboratory parameter.316

Commonly used screening assays for DIC include (1) a reduced or downward trend in the platelet count (usually < 100,000/mm3); (2) the presence of fibrin-related markers, including fibrin degradation prod-ucts, D-dimers, or soluble fibrin in plasma; (3) prolongation of the prothrombin time or the activated partial thromboplastin time (>1.2 times the upper limit of normal); and (4) low plasma levels of endog-enous anticoagulants, such as antithrombin III and protein C.85,317 Reduced levels of ADAMTS13 activity and/or elevation of soluble thrombomodulin, plasminogen activator inhibitor, von Willebrand factor, and von Willebrand factor propeptide may be seen, although none of these tests offers a secure diagnosis of DIC or predicts its outcome.318 Serial measurements of these tests may improve the diag-nostic certainty of suspected DIC.316 The use of low levels of antithrom-bin and protein C for the diagnosis of DIC was challenged by Asakura and co-workers,87 who found no differences in the plasma activity of antithrombin and protein C between septic and control patients after stratification for plasma albumin levels.

The most common adverse consequence of DIC is hemorrhage, which is most often apparent as oozing from wounds or as gastroin-testinal (GI) bleeding. Thrombosis of large and small vessels may also occur, usually in relationship to local tissue infection or indwelling catheters.133,319 In general, thrombosis-induced tissue injury is most apparent when there is cutaneous necrosis and especially when blood flow to a distal structure (finger, toe, hand, foot, tip of the nose) is interrupted. There is evidence that vasoconstriction contributes to the pathogenesis of arterial thrombosis, and most attempts to restore blood flow have involved interfering with the sympathetic nerve supply to the affected extremity.320 In one instructive case, a young boy with purpura fulminans developed gangrene of three extremities; in the spared limb, vasoconstriction was impaired because of a brachial plexus injury acquired at birth.139

Dysfunction of Other OrgansPatients entered into studies of severe sepsis must have evidence for one or more dysfunctional organ systems. As is shown in Figure 75-4, hypotension is the most commonly noted abnormality; the frequency with which other organ systems are affected ranges from 5% to 50%. There is considerable patient-to-patient variability in the manifesta-tions of severe sepsis.

Acute Lung InjuryHyperventilation, with respiratory alkalosis, can be one of the earliest manifestations of sepsis. Similarly, pulmonary dysfunction typically occurs early in the course of severe sepsis.321 The clinical diagnosis of ARDS is based upon four elements: (1) the occurrence of lung injury within 1 week of a known clinical insult, (2) chest imaging that shows bilateral opacities, (3) the absence of cardiac failure or fluid overload,

FIGURE  75-4 Organ dysfunction at entry into studies of severe sepsis. Bars show means of data from six representative studies.14,315,449,451-453 Brackets show the minimal and maximal values reported. The definitions used to diagnose organ dysfunction varied some-what from study to study. Renal dysfunction was uncommon in some series and very common in others. CNS, central nervous system; DIC, dis-seminated intravascular coagulation.

Per

cent

of c

ases

(at

stu

dy e

ntry

)

75

50

25

0

Pul

mon

ary

Ren

al

Hep

atic

CN

S

DIC

Sho

ck

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

928

Desquamation of the skin of the distal extremities does not usually occur until the second week of illness.

Ischemic changes (dusky or pallid color, coldness, loss of pulses) usually occur in the hands and feet, where they may follow thrombosis of small-sized to midsized arteries. Inflammation-induced coagulopa-thy and vasoconstriction both contribute to their pathogenesis, as noted earlier.

Septic ShockIn prospective studies of the natural history of critical illness,6,340 patients have progressed from sepsis to severe sepsis to septic shock, suggesting that these syndromes are part of a continuum. In the study by Rangel-Frausto and co-workers, 71% of the patients who developed septic shock had been previously classified as having severe sepsis, sepsis, or SIRS.6

The cardiovascular impact of sepsis has two components: myocar-dial dysfunction and relative hypovolemia resulting from vasodilation. A pattern of sepsis-associated myocardial dysfunction was recognized during the 1980s. It includes reduced left and right ventricular ejection fractions, increased left and right ventricular end-diastolic volumes, and an elevated heart rate and cardiac output.341 This pattern typically follows fluid resuscitation, occurs 24 to 48 hours after the onset of severe sepsis, and is reversible in patients who survive 5 to 10 days after its onset. The cardiac depression associated with septic shock reflects the effects of inflammatory mediators on cardiac myocyte and micro-circulatory function, is not caused by ischemia, and does not usually require inotropic therapy. However, a small fraction of patients with septic shock may develop profound myocardial depression in conjunc-tion with vasodilatory shock and require inotropic support (i.e., dobu-tamine) in addition to vasopressor therapy.

Mechanisms implicated in the development of sepsis-induced myo-cardial depression include alterations in calcium homeostasis, mito-chondrial dysfunction, apoptosis, circulating cardiosuppressant mediators, nitric oxide, and peroxynitrite. Some authors have posited that sepsis-induced cardiodepression is a form of cardiac hiberna-tion.342 A postmortem study found that cardiac myocyte cell death was rare, whereas sepsis-induced focal mitochondrial injury was present in many cells.136 Connexin-43, a gap junction protein that forms electrical synapses between myocytes, was altered in a way that suggested car-diomyocyte injury. These findings were said to be in keeping with the reversible nature of the myocardial injury induced by sepsis.136,343

Hypovolemic shock can usually be reversed by administering intravenous fluids. In the normovolemic patients with vasodilatory (warm, hyperdynamic) shock, studied prospectively by Abraham and co-workers,344,345 the first noticeable change was a fall in oxygen con-sumption, which was followed by compensatory increases in cardiac output and oxygen delivery; peripheral vascular resistance decreased progressively over the 24-hour period before the onset of overt hypo-tension. The lowest blood pressure was recorded when the cardiac output failed to compensate for low vascular resistance.

Severe Sepsis and Septic Shock: Is There a “Tipping Point”?When they compared whole-blood gene expression profiles in patients with “sterile SIRS” and early sepsis (infection-induced SIRS), Johnson and co-workers346 found that upregulation of mitogen-activated protein kinase-14 (MAPK-14) (p38) occurred up to 48 hours before the diagnosis of clinical sepsis. Other mRNAs that increased in abun-dance in early sepsis were those for members of the IL-1 receptor and IL-22 receptor families. A subsequent study by the same group found that many TLR-related effector molecules were upregulated (less than twofold) in peripheral blood cells 24 hours before the diagnosis of clinical sepsis347; MAPK-14 was again among the harbinger molecules. Using mass spectrometry to identify plasma proteins, the authors also reported that 134 unique plasma proteins were overrepresented in sepsis patients compared with ICU patients who had only SIRS; most were coagulation or complement proteins.348

Another potential marker for the transition to severe sepsis is a decrease in the natural variability in heart rate, respiratory rate, or temperature. There may also be loss of the normal circadian variability in plasma cortisol, glucose, iron, and cytokine levels. Heart rate

found that increased GI permeability preceded the onset of multiple organ hypofunction.327

In addition, aspiration of the microbial and chemical contents of the upper GI tract into the tracheobronchial tree may initiate nosoco-mial pneumonia. Small erosions of the gastric and duodenal mucosa predispose to upper GI bleeding. Ileus, a common feature of septic shock, may persist for a day or two after shock resolves.

Hepatic DysfunctionThe principal sepsis-associated abnormality is cholestatic jaundice, characterized by elevations in conjugated and unconjugated bilirubin (<10 mg/dL). These changes occur in patients with and without pre-existing liver disease and may precede recognition of infection.328 In patients with severe sepsis, elevated alkaline phosphatase, bilirubin, and amino transferase levels are common, but frank hepatic failure (“shock liver”) is unusual. If the duration of septic shock is prolonged, however, a massive rise in serum transaminases may follow hypoxic necrosis of centrilobular liver cells.329 Hypoxic hepatitis has a poor prognosis.

Immune DysfunctionReactivation of latent herpes simplex and CMV infections occurs in approximately 35% of critically ill patients,330-333 and CMV viremia has been described in a similar fraction of patients with severe sepsis.334 The extent to which CMV contributes to immunosuppression in these patients is not known, yet patients with CMV antigenemia have had higher rates of nosocomial infection, prolonged hospitalization, and mortality.332,335 Although clinical experience suggests that patients with severe sepsis are at increased risk for secondary infections,336 a quan-titative documentation of this risk has not been published. In one autopsy study of patients who died in an ICU, 80% had an undiagnosed infectious focus.337

Cutaneous ManifestationsA wide range of skin lesions may occur in patients with severe sepsis. They include the cutaneous reaction at a local inoculation site (pustule, eschar), lesions that appear at sites of hematogenous seeding of the skin or underlying soft tissue (petechiae, pustules, ecthyma gangrenosum, cellulitis), diffuse eruptions caused by bloodborne toxins (e.g., TSST), and hemorrhagic or necrotic lesions. Recognition of certain character-istic lesions can greatly assist etiologic diagnosis.

Musher338 distinguished three patterns of tissue involvement by gram-negative enteric bacilli.1. Cellulitis and thrombophlebitis are associated with intense local

inflammation. Bacteria implicated in case reports include Campy-lobacter fetus, Vibrio species, and Aeromonas hydrophila.

2. When the inflammatory response is impaired, usually by neutro-penia, ecthyma gangrenosum or bullous lesions may occur (see later); Pseudomonas aeruginosa is the most commonly isolated microorganism.

3. In symmetrical peripheral gangrene associated with DIC, fibrin thrombi are seen in small vessels, but neither inflammatory cells nor bacteria are found.Palpable petechiae or purpura suggests leukocytoclastic vasculitis,

which may be caused by N. meningitidis, R. rickettsii, S. pneumoniae, H. influenzae, and occasionally S. aureus.339 Pustules often contain S. aureus or C. albicans. Cellulitis is most often caused by S. pyogenes but may, in unusual settings, result from bacteremia caused by clos-tridial species or one of the gram-negative bacilli mentioned earlier.

The term ecthyma gangrenosum (“necrotic blister”) is used for lesions that begin as papules surrounded by erythema and edema and evolve into hemorrhagic, necrotic ulcers. They typically appear between the umbilicus and the knees. Although often considered pathogno-monic for P. aeruginosa bacteremia, ecthyma gangrenosum has also been observed in patients whose blood cultures grew Klebsiella, Ser-ratia, A. hydrophila, or E. coli. Pathologic examination reveals direct invasion of venules by bacteria and local thrombosis. Almost all patients with ecthyma gangrenosum are neutropenic at the time the lesions develop.

Diffuse erythema (erythroderma) is a characteristic finding in toxic shock syndrome caused by either S. aureus or S. pyogenes.

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

929

In patients with indwelling vascular catheters, the rate of microbial growth in a culture of blood drawn through the catheter may be com-pared with that of blood drawn from a peripheral vein; a difference in the “time to positivity” of 2 or more hours suggests catheter infec-tion.354 Careful preparation of the skin is essential to avoid false posi-tive cultures from skin contaminants. Chlorhexidine (2%) has a short (15 to 30 seconds) drying time compared with 10% povidone-iodine or 1% to 2% tincture of iodine (approximately 2 minutes)355; in one recent trial, cleansing with 2% chlorhexidine in 70% alcohol was supe-rior to 10% aqueous povidone-iodine for preventing culture contami-nation.356 Using these agents properly can reduce costs and preserve resources.352

Cultures and microscopic examination of urine, sputum (including tracheal aspirate if done within a few hours of intubation), likely infected fluids, purulent wound drainage, and skin lesions should also be obtained. Gram-stained material obtained from biopsies or needle aspirates of petechial lesions can provide a rapid diagnosis in patients with meningococcemia.357

New molecular and other non–culture-based methods have the potential to speed diagnosis of serious infections. These include nucleic acid–based diagnostics applied to cultures from clinical samples or used directly on blood or serum.353 Detection of microbial nucleic acids does not necessarily indicate that a viable organism was present or that transient bacteremia was related to a true infection.353

Diagnostic ImagingAxial tomography is an important complement to routine chest and abdominal radiography to assess for unrecognized sources of infection in the sinuses, lungs, liver, and abdomen. Ultrasonography and cho-lescintigraphy (hepatobiliary iminodiacetic acid [HIDA] scanning) may be useful for evaluating gallbladder function.

Cytokine and Biomarker LevelsAlthough there have been many attempts to identify a cytokine profile that would distinguish infected patients from those with systemic responses to other stimuli, none has been very successful. A review of 178 biomarkers used for diagnostic and prognostic purposes noted that all lack sufficient sensitivity and specificity to be used in routine clinical practice.358 Studies of biomarkers in severe sepsis have often been limited by the absence of immunosuppressed patients or patients with non–respiratory tract infections. Variability in physi-cian adherence to treatment algorithms may also compromise their validity.

Adrenal Insufficiency in Patients with Septic ShockThe clinical and laboratory diagnoses of sepsis-associated relative adrenal insufficiency are inexact and controversial (see adrenal insuf-ficiency under “Clinical Manifestations”). Unresolved issues include the need to measure free cortisol, the diagnostic significance of cortisol-binding proteins, the utility of salivary cortisol levels, and the quantitation of tissue glucocorticoid resistance.359 The most useful clinical definition of relative adrenal insufficiency may be based simply upon the response to hydrocortisone administration (see “Other Ther-apies”). The presence of “pressor-dependent hypotension that responds to the administration of 50 to 100 mg hydrocortisone every 6 hours” would strongly support the diagnosis.360 Many experts now recom-mend obtaining a baseline serum cortisol level before initiating hydro-cortisone therapy. A value less than 15 µg/dL in a patient with septic shock should encourage careful evaluation for adrenal insufficiency after recovery from the septic episode. The diagnostic value of the ACTH stimulation test is less clear285; of importance, the reversal of vasopressor-dependent shock by low dose steroids is not predicted by the response to the ACTH stimulation test.361

THERAPYSepsis, severe sepsis, and septic shock are medical emergencies. Despite intensive effort, including more than three dozen clinical trials, optimal therapy has changed little since the 1960s.362 As sum-marized by Young,363 “Early clinical suspicion, rigorous diagnostic measures, aggressive initiation of appropriate antimicrobial therapy,

variability (HRV) has been studied most intensively; although a decrease in HRV may herald the onset of systemic organ hypofunction, particularly in neonates, many therapeutic interventions may influence HRV in sick patients. Using a composite measure of variability has been proposed.349

DIAGNOSISNo bedside or laboratory test provides a definitive diagnosis. There is also considerable inter-individual and time-dependent variability in the expression of the body’s responses to infection, so a diagnostically useful “profile” of laboratory tests is not possible. On the other hand, certain findings are sufficiently suggestive that they should prompt further evaluation. In addition to the signs that comprised SIRS (tachy-cardia, tachypnea, leukocytosis or leukopenia, and fever or hypother-mia; see Table 75-1), findings such as altered mental status, unexplained hyperbilirubinemia, lactatemia, metabolic acidosis or respiratory alka-losis, and thrombocytopenia can be useful clues. The appearance of new lesions on the skin or mucosae may also be suggestive.

One normal response to infection is a neutrophilic leukocytosis in the peripheral blood. Infections that are typically associated with peripheral blood leukopenia include typhoid fever, brucellosis, Rocky Mountain spotted fever, Colorado tick fever, and ehrlichiosis; in indi-viduals with severe sepsis induced by bacteria, leukopenia is more common among children than adults.306 Fever is also a normal response to infection, and an increase in body temperature above a certain level (usually 38.0° or 38.3° C) is often the trigger for initiating a diagnostic evaluation. Some septic patients may be euthermic or hypothermic, however. These include older adults, patients with open wounds or large burns, and patients taking anti-inflammatory or antipyretic drugs. In patients with comorbid conditions or immunosuppression, the clinical manifestations of sepsis may also be atypical: for example, the fever response may be blunted (concomitant glucocorticoid use, continuous renal replacement therapy), the white blood cell count may be normal (depressed bone marrow reserves resulting from chemo-therapy or stem cell transplantation), and the heart rate may be normal (β-blockers, sick sinus syndrome).

Differential DiagnosisNumerous noninfectious conditions can mimic sepsis by presenting with hypotension and/or organ failure. They include burns, trauma, adrenal insufficiency, pancreatitis, pulmonary embolism, dissecting or ruptured aortic aneurysm, myocardial infarction, occult hemorrhage, cardiac tamponade, and drug overdose. Fever and hypotension can also be caused by a number of noninfectious processes, including adrenal insufficiency, thyroid storm, pancreatitis, drug hypersensitivity reactions, malignant hyperthermia, serotonin syndrome, and heat-stroke. Vasodilatory shock can be a manifestation of anaphylaxis. A sepsis-like syndrome may follow cardiopulmonary bypass; there is cir-cumstantial evidence that the trigger is either pump trauma to circulat-ing leukocytes or bacterial endotoxin absorbed from the gut. Indeed, sepsis and septic shock may complicate any of the disease states described above. One should always consider occult sepsis in the dif-ferential diagnosis of fever, acutely altered mental status, thrombocy-topenia, or hypotension.350

CulturesCultures are essential for identifying the likely microbial invaders and ascertaining antimicrobial susceptibility patterns. For optimal sensitiv-ity and specificity, blood cultures (split between aerobic and anaerobic bottles) should be drawn from two or three different venipuncture sites.351 The volume of blood drawn (adults 20 to 30 mL/venipuncture, children no more than 1% of total blood volume) is the most important variable in detecting bacteremia.352

Bacteremia may be categorized as transient, intermittent, or con-tinuous (reviewed by Mancini and co-workers353). Transient bactere-mia lasts minutes to hours and may occur with manipulation of either anatomic sites colonized by normal flora (i.e., colonoscopy) or local infected sites. Intermittent bacteremia is associated with closed-space infections (e.g., abscesses) or focal infections (e.g., pneumonia). Per-sistent low-grade bacteremia is associated with an intravascular focus, such as endocarditis or vascular graft infection.

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

930

therapy administered within 24 hours of the onset of shock reduced mortality from candidemia by almost 50% when combined with ade-quate source control.373

Patients with severe sepsis should thus receive a broad-spectrum, intravenous regimen that is effective for both gram-negative and gram-positive bacteria, and they should receive it as quickly as possible. The choice of drugs should be modified according to the patient’s own microbiologic culture data and the resistance patterns prevalent in the patient’s community or hospital. Under some circumstances, such as when a patient is neutropenic or develops severe sepsis after having received a broad-spectrum regimen for another infection, empirical therapy for Candida species may be warranted (Candida species now account for 10% or more of the cases of severe sepsis in such patients). Anti-Candida therapy should also be considered in patients with severe sepsis who have indwelling catheters. Specific recommendations are provided in Table 75-4.

With the possible exception of neutropenic patients or those with P. aeruginosa disease, there is little evidence to support the use of combination therapy using an aminoglycoside instead of mono-therapy with a broad-spectrum drug such as a carbapenem.374 Many experts consider an extended-spectrum penicillin, combined with a β-lactamase inhibitor, to be effective empirical therapy for most patients in locales where multiresistant gram-negative bacteria are not prevalent. This recommendation was questioned recently by a meta-analysis that found that combination antimicrobial therapy improved survival in patients with septic shock but may be harmful in those with less severe sepsis.375 Like so many other therapeutic interventions in septic patients,376 antibiotic benefit seems greatest in the sickest individuals.

In much of the world, S. aureus isolates are so frequently methicillin-resistant that, if S. aureus is a potential causative organism, a drug active against MRSA is recommended for empirical therapy in pref-erence to oxacillin or nafcillin. Fortunately, S. aureus resistance to

comprehensive supportive care, and measures aimed at reversing pre-disposing causes are the cornerstones of successful management.”

Antimicrobial DrugsNumerous analyses have concluded that early treatment of bacteremic patients with an appropriate antimicrobial drug improves sur-vival.364-366,367,368 In most of these studies, a drug was considered appro-priate if it was able to inhibit the patient’s microbial isolate(s) in vitro and was administered within 24 to 48 hours of the onset of bacteremia or severe sepsis. One study found that appropriate antimicrobial therapy, as defined by the recommendations of an authoritative anti-biotic guidebook, was equally beneficial to patients with septic shock who did and did not have positive culture results.233 Among patients with septic shock who received appropriate antimicrobial therapy, ICU-acquired infection and severity of illness (Acute Physiology and Chronic Health Evaluation [APACHE] II) were the most significant determinants of outcome.369 In another case series, inappropriate antimicrobial treatment was most common (45%) in patients with nosocomial infections that developed after treatment for a community-acquired infection.364

There is also strong evidence now that survival is more likely if septic patients receive appropriate antimicrobial drugs as soon as possible after the diagnosis is suspected. In keeping with the important role for rapid antibiotic administration promoted by Greis-man370 almost 30 years ago, a retrospective review by Kumar and co-workers371 found a strong relationship between the delay in effective antimicrobial initiation and in-hospital mortality of septic shock. Each hour of delay in antimicrobial administration was associated with an average decrease in survival of 8%. The time to initiation of effective antimicrobial therapy was the single strongest predictor of outcome. Others found that antimicrobial administration within 4 hours of arrival at the hospital was associated with decreased mortality in older patients with community-acquired pneumonia,372 and antifungal

TABLE 75-4  Empirical Antibiotic Options for Patients with Severe Sepsis or Septic Shock

SUSPECTED SOURCELung Abdomen Skin/Soft Tissue Urinary Tract Source Uncertain

Major Community-Acquired Pathogens

Streptococcus pneumoniaeHaemophilus influenzaeLegionellaChlamydia pneumoniae

Escherichia coliBacteroides fragilis

Streptococcus pyogenesStaphylococcus aureusPolymicrobial

E. coliKlebsiella speciesEnterobacter speciesProteus spp.Enterococci

Empirical Antibiotic Therapy

Moxifloxacin or levofloxacin or azithromycin plus cefotaxime or ceftazidime or cefepime or piperacillin-tazobactam

Imipenem or meropenem or doripenem or piperacillin-tazobactam ± aminoglycoside

If biliary source: piperacillin-tazobactam, ampicillin-sulbactam, or ceftriaxone with metronidazole

Vancomycin or daptomycin plus either imipenem or meropenem or piperacillin-tazobactam; ± clindamycin (see text)

Ciprofloxacin or levofloxacin (if gram-positive cocci, use ampicillin or vancomycin ± gentamicin)

Vancomycin plus either doripenem or ertapenem or imipenem or meropenem

Major Commensal or Nosocomial Microorganisms

Aerobic gram-negative bacilli

Aerobic gram-negative rods

AnaerobesCandida spp.

Staphylococcus aureus (? MRSA)

Aerobic gram-negative rods

Aerobic gram-negative rods

Enterococci

Consider MDRO if in area of high prevalence

Consider echinocandin if neutropenic or indwelling intravascular catheter

Empirical Antibiotic Therapy

Imipenem or meropenem or doripenem or cefepime (if Acinetobacter baumanii or carbapenem-resistant Klebsiella in ICU, add colistin)

Imipenem or meropenem ± aminoglycoside (consider echinocandin)

Vancomycin or daptomycin plus imipenem-cilastatin or meropenem or cefepime, ± clindamycin

Vancomycin plus imipenem or meropenem or cefepime

Cefepime plus vancomycin ± caspofungin

Dosages for intravenous administration (normal renal function):*Imipenem-cilastatin, 0.5-1.0 g q6-8h*Meropenem, 1-2 g q8h*Doripenem, 0.5 g q8hPiperacillin-tazobactam, 3.375 g q4h or 4.5 g q6hVancomycin, load 25-30 mg/kg, then 15-20 q8-12hCefepime, 1-2 g q8hLevofloxacin, 750 mg q24h

Ciprofloxacin, 400 mg q8-12hMoxifloxacin, 400 mg qdCeftriaxone, 2.0 g q24hCaspofungin, 70 mg, followed by 50 mg q24hColistin: loading dose = 5 mg/kg body weight. For maintenance dosing,

see University of California, Los Angeles Dosing Protocol: www.infectiousdiseases-ucla-affiliated.org/Intranet/FILES/ColistinDosing.pdf

*Carbapenems are less susceptible to extended-spectrum β-lactamases; base choice on local resistance pattern.ICU, intensive care unit; MDRO, multidrug-resistant organisms; MRSA, methicillin-resistant Staphylococcus aureus.For MDRO, resistance usually includes carbapenems.

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

931

Intravenous Fluids and VasopressorsFluid resuscitation is the mainstay of hemodynamic support in patients with septic shock. Whereas antibiotics or fluids alone did not improve outcome in animal models of septic shock, these interventions were found to act synergistically to improve survival.382 Crystalloid is gener-ally used for resuscitation instead of colloid.383 The Saline versus Albumin Fluid Evaluation (SAFE) trial384 compared fluid resuscitation with 4% albumin versus saline in a heterogeneous population of ICU patients who required fluids for intravascular volume depletion. Sur-vival was similar in the two arms at 28 days, and the two fluids were considered comparable in patients with severe sepsis. Although a meta-analysis found that albumin was associated with lower mortality when compared with other fluids,385 a recent trial found that patients who received crystalloid plus albumin to maintain a serum albumin concentration of 30 g/L were no more likely to survive than were patients who received only crystalloid.385a The use of hydroxyethyl starch preparations for resuscitation caused an increase in mortality and acute kidney injury in critically ill patients386 and in patients with septic shock.387 It should be not be used.

The goals of fluid resuscitation in severe sepsis and septic shock are to ensure adequate tissue perfusion by restoring effective intravascular volume (depleted by vasodilation and increases in vascular permeabil-ity) and to optimize cardiac output by enhancing venous return and cardiac filling. A reasonable goal in general is maintenance of mean arterial pressure (MAP) greater than 65 mm Hg, although in some patients with long-standing hypertension, a higher MAP may be required. In most patients, 15 to 30 mL/kg or up to 4 to 6 L of crystal-loid may be required in the early phases of resuscitation. Too little fluid may cause tissue hypoperfusion and worsen organ function, whereas excessive fluid administration may impair organ function resulting from tissue edema. When no further benefit is apparent from addi-tional fluid administration (i.e., restoration of mean arterial blood pressure), vasopressor administration is essential to maintain adequate tissue perfusion. However, considerable controversy exists regarding which hemodynamic parameters should guide resuscitation in septic patients.388-390 Three international multicenter trials are currently addressing critical end points in resuscitation.391 Protocol-based resus-citation using an oximetric central venous catheter to monitor central venous pressures and oxygen saturation did not improve survival of patients with septic shock diagnosed in the emergency room. Using this catheter to guide the administration of fluids, vasopressors, dobu-tamine, or packed red cell transfusions to achieve specific hemody-namic targets was no better than usual care composed of early antibiotics, fluid resuscitation, and, if needed for venous access, a central venous catheter.391a

For many years, dopamine was considered the drug of choice for restoring normotension in patients with septic shock. When used at low doses (<5 µg/kg/min), its preferential interaction with dopaminer-gic receptors was thought to produce renal and splanchnic vasodila-tion. This notion has been challenged by recent studies and analyses. In particular, a randomized, controlled clinical trial found that low-dose dopamine infusion did not improve survival or prevent renal failure in critically ill patients at risk for renal dysfunction,392 and a similar conclusion was reached by a retrospective analysis of patients with septic shock in a large clinical trial.393 Many experts now favor using norepinephrine over dopamine for septic shock.394 A trial of norepinephrine plus dobutamine (added if needed) compared with epinephrine found no difference in 28-day all-cause mortality.395

Vasopressin levels initially rise as patients develop shock, then they fall with more prolonged hypotension. Continuous infusion of argi-nine vasopressin (AVP) may help restore normotension in patients with catecholamine-resistant vasodilatory shock. The doses of vaso-pressin that increase blood pressure in septic patients are lower than those required in normal individuals. Landry and Oliver193 speculated that the increase in vasopressor potency may be caused by unoccupied vascular receptors for vasopressin, the coexistence of autonomic failure (which potentiates vasopressin action), or vasopressin’s ability to enhance the vasoconstrictor effect of norepinephrine, which is present in markedly elevated concentrations in patients with septic shock. Vasopressin also directly inactivates KATP channels in vascular smooth muscle and inhibits the inducible form of NO synthase. Vasopressin is

vancomycin or daptomycin continues to be rare in most communities. If a vancomycin-resistant enterococcus (VRE) is the likely trigger for severe sepsis, linezolid should be included. Daptomycin is another alternative drug for MRSA, but its efficacy in patients with severe sepsis has not been tested, and it should not be used for the treatment of pneumonia because its activity is inhibited by pulmonary surfactant. Linezolid’s toxicities (i.e., thrombocytopenia, anemia, rare serotonin syndrome) limit its attractiveness as a substitute for vancomycin.

Although there are few data on the use of once-daily dosing of aminoglycosides in critically ill patients, the available evidence sug-gests that this administration method is safe.377 The volume of dis-tribution of both tobramycin and gentamicin is higher in critically ill patients with septic shock than in those without shock. A once- daily dosage of 7 mg/kg produced maximal peak concentration (Cmax) to minimal inhibitory concentration (MIC) ratios in most patients. This should be avoided if possible in patients with unstable renal func-tion, anuria, or an increased volume of distribution (i.e., ascites, anasarca) because of the difficulty in predicting peak and trough drug levels. Therapeutic drug monitoring is warranted, and more than one drug level may be needed to determine the appropriate dosing interval.

The available data also generally support using continuous infu-sions of β-lactam drugs to maximize the time that blood concentra-tions remain above a target bacterium’s MIC, yet adequately sized and controlled clinical trials have not been reported to date.378 The most recent study found that continuous infusion only benefited the sickest patients.379

Antimicrobial Chemotherapy for Specific Etiologies of Severe SepsisAlthough empirical broad-spectrum regimens will provide agents that are active against most pathogens, some situations may warrant differ-ent or additional coverage. For example, tick exposure might warrant treatment with doxycycline (Rocky Mountain spotted fever) or atovaquone-azithromycin for babesiosis (with clindamycin if critically ill) in different exposure environments. Travel to the tropics might suggest the need for antimalarial therapy. In patients with suspected or proven streptococcal myositis/fasciitis or toxic shock syndrome, clindamycin should be given in addition to penicillin G to reduce toxin production. If staphylococcal toxic shock syndrome is considered, clindamycin should be given with either oxacillin (if methicillin-susceptible S. aureus [MSSA]), vancomycin, or linezolid. In patients who may have eaten raw oysters and acquired V. vulnificus bacteremia, intravenous doxycycline should be used along with ceftazidime or a fluoroquinolone. Severe sepsis that follows a dog bite may be due to C. canimorsus, which is usually susceptible to cephalosporins, carbapen-ems, β-lactam–β-lactamase inhibitor combinations, and quinolones but resistant to trimethoprim-sulfamethoxazole and aminoglycosides. Cefotaxime or ceftriaxone is preferred for asplenic patients, who may have overwhelming bacteremia with S. pneumoniae, N. meningitidis, H. influenzae type b, or C. canimorsus.

Surgical Drainage (Source Control)Recovery from severe sepsis or septic shock is unlikely, even with appropriate antimicrobial therapy and diligent ICU care, if the patient has an undrained abscess or obstructed viscus.380 An excellent discus-sion of surgical and nonsurgical “source control” measures was pub-lished by Marshall and others.381 In patients with community-acquired infections, the most common occult sources are in the lungs and the urinary tract. Intra-abdominal infections (e.g., diverticulitis, cholecys-titis, pylephlebitis), septic arthritis, endocarditis, and osteomyelitis should also be sought. Nosocomial infections often arise at sites of epithelial barrier disruption and thus frequently involve intravascular catheters, endotracheal tubes (pneumonia and paranasal sinusitis), urinary catheters, and operative wounds or other sites of traumatic injury.

In general, when a patient develops severe sepsis, all intravascular and bladder catheters should be removed, with reinsertion at new sites as needed. It is not often necessary to do surgical exploration of an infected thrombus because medical management usually suffices.350

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

932

dominant), whereas some benefit can be derived from blocking even a single proinflammatory mediator in patients at the more severe end of the spectrum. To date, modifying inflammation with novel agents has not been shown to be beneficial.391

Antagonists to Gram-Negative Bacterial Endotoxin and Related MoleculesClinical trials have evaluated antiendotoxin polyclonal antiserum (six studies) and monoclonal antibodies against common epitopes of endo-toxin (five studies) for their ability to improve outcome from severe sepsis induced by gram-negative bacteria. None succeeded as either prophylaxis or therapy.401 Similarly, a monoclonal antibody against enterobacterial common antigen, a surface antigen closely linked to endotoxin, did not benefit septic patients.402 The recombinant form of a naturally occurring endotoxin-neutralizing molecule, recombinant BPI (rBPI), may have reduced the need for amputation in children with fulminant meningococcemia, but it did not improve survival.403 A potent lipid A analogue that competes with LPS for binding MD-2 (Eritoran24) failed to improve outcome in severe sepsis, as did a cyclo-hexene antagonist of TLR4 signaling (TAK-242).391,404 A trial of a phos-pholipid emulsion that neutralizes endotoxin was stopped early because of toxicity,405 whereas a preliminary trial of hemoperfusion through a matrix containing polymyxin B was stopped prematurely after it achieved its primary end points.391 Its usefulness must be dem-onstrated in larger trials.

AnticoagulantsThree recombinant anticoagulant drugs have been tested for their ability to increase the survival of patients with severe sepsis or septic shock. None was consistently successful in improving outcomes. Tissue factor pathway inhibitor (TFPI) and antithrombin III did not improve outcomes in severe sepsis.137,406 Activated protein C (aPC, drotrecogin-alfa [activated]) reduced mortality from 31% (placebo control) to 25% (treatment group) (P < .005)207 and was licensed for the treatment of severe sepsis. The basis for the apparent impact of aPC on sepsis sur-vival was questioned by some,407 and concerns were raised about a significant increase in the incidence of severe bleeding, including intra-cranial hemorrhage. The efficacy of aPC was not confirmed in trials performed in children408 and in adults with low risk of dying (APACHE II score ≤25)409; each of these trials was stopped when an interim analysis revealed that the drug would not reduce mortality. Because of concerns in Europe regarding the drug’s efficacy and toxicity, a confir-matory trial was done in patients with septic shock. aPC failed to reduce mortality at 28 or 90 days, compared with placebo,410 and it was withdrawn from the market in 2011. A second confirmatory trial also found no benefit in patients with septic shock.411 The protein C mol-ecule has recently been modified so that it is not an anticoagulant yet retains anti-inflammatory potency.412 It is conceivable that such mutated versions will be useful anti-inflammatory drugs and have less risk of hemorrhage than does aPC.

It is noteworthy that, in each of the three large trials of anticoagu-lant drugs, administration of heparin in a nonrandomized fashion to patients in the placebo group was associated with a reduction in mor-tality. The basis for these results is not clear; because most ICU patients receive low-dose heparin as prophylaxis for deep venous thrombosis, unless there is a contraindication to its use, it is possible that the observed outcome difference simply reflects selection bias. More recently, a large retrospective review found significantly lower 28-day mortality in patients who received unfractionated heparin intrave-nously during the first 48 hours of septic shock.413 Prospective con-trolled trials of this inexpensive, widely-available agent are needed.

Boosting Host DefensesDuring the 1990s, several studies addressed the ability of recombinant IFN-γ to prevent severe sepsis in patients who had recently undergone major surgery or sustained major trauma. Unfortunately, prophylactic administration of IFN-γ did not significantly reduce the incidence of nosocomial infection and severe sepsis,414,415 even though an impact of the drug on monocyte function was observed.416

Attempts to enhance myeloid cell function in septic patients by giving growth factors have also not met with success. G-CSF did not

often used as a catecholamine-sparing agent in pressor-dependent shock, but a mortality benefit has not been seen.391

OTHER THERAPIESGlucocorticoidsMany patients with septic shock exhibit a rightward shift in the dose-response relationship between blood pressure and catecholamines. Annane and co-workers80 found that this occurs most often in patients with impaired adrenal function and that, in such patients, administer-ing hydrocortisone could return the dose-response curve to normal. Several factors probably contribute to reduced sensitivity to catechol-amines, including downregulation of adrenergic receptors and NO- induced vasopressor resistance. Hydrocortisone increases adrenergic receptor expression. There is thus a plausible theoretical and experi-mental basis for using glucocorticoids to treat patients with septic shock.

Investigators noted several decades ago that very high doses of corticosteroids were beneficial in animal models of septic shock. Accordingly, high doses of corticosteroids were widely given as adjunc-tive therapy for human sepsis and septic shock. Unfortunately, ran-domized clinical trials of high doses of corticosteroids for sepsis (the median dose was equivalent to 23,975 mg hydrocortisone given over 24 hours) showed that this high-dose approach was harmful,361 and clinicians abandoned it.

Although the ACTH stimulation test may be useful to identify patients with shock and overt primary or secondary adrenal insuffi-ciency, it has not been adequate to identify the patients with septic shock and CIRCI who will respond to physiologic doses of ste-roids.285,361 As such, some experts now recommend that the decision to treat with low doses of glucocorticoids should be based on clinical criteria and not on the results of adrenal function testing.285 In patients with septic shock who have not responded to fluid and vasopressor resuscitation, hydrocortisone treatment should be initiated (50 mg IV every 6 hours or with a loading dose of 100 mg, followed by a continu-ous infusion of 10 mg/hr), continued for 7 days, and then slowly tapered over 5 to 6 days. Confirmation of this approach awaits further large clinical trials and new methods to assess the integrity of the HPA axis and tissue responsiveness to corticosteroids. Whether there is added benefit to the addition of a mineralocorticoid, fludrocortisone (50 µg PO daily), remains uncertain,359,396 but recent trials have been negative.397

Anti-inflammatory DrugsDuring the 1990s, clinical trials were performed to test the ability of numerous immunomodulatory drugs to improve survival in patients with severe sepsis.391,398,399 They included large doses of glucocorticoids, antiendotoxin agents, antibodies to TNF and TNF-immunoglobulin fusion proteins that trap TNF, IL-1 receptor antagonist and antagonists to PAF, bradykinin, phospholipase A2, NO synthase, cyclooxygenase, bradykinin, and others. Although many of these agents appeared prom-ising in preliminary trials, none reproducibly improved 28-day all-cause survival, and some (e.g., a NO synthase inhibitor) caused harm. Explanations offered for the failure of this approach have included using the wrong drugs, doses, or duration of therapy; administration of the drug too late in the clinical course; heterogeneity in the clinical population treated; and ineffectiveness of single interventions. One group of experts recommended limiting clinical trials in septic patients to individuals with specific infectious diseases or sites of infection.400

A meta-regression analysis of 23 clinical trials concluded that the efficacy of anti-inflammatory drugs in patients with severe sepsis depends on the risk of dying.376 Although many of the agents studied work through different biologic mechanisms, their efficacy was consis-tently greater in patients with a high risk of dying, whereas they were ineffective or harmful in those with low mortality risk. Studies in experimental animal models showed similar trends.376 Some of the same drugs have been very effective in the treatment of rheumatoid arthritis and other rheumatologic diseases, yet they have also predis-posed patients to reactivation of tuberculosis and other infections. By analogy with this experience, perhaps interfering with the proinflam-matory response impairs antimicrobial defenses in patients with less severe sepsis (in whom systemic anti-inflammation may already be

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

933

receiving mechanical ventilation.429 A recent study investigated the benefits of fever control by using external cooling in sedated patients with septic shock. The short-term benefits included less vasopressor requirement and more rapid shock reversal.430 A nonsignificant trend toward more nosocomial infections was noted at 2 weeks, however.

Preventing Secondary InfectionsBecause patients with severe sepsis are immunosuppressed and sub-jected to invasive procedures, they may be at risk for secondary infec-tions. Measures that decrease acquisition of hospital-associated microorganisms include hand washing431 and the use of barrier precau-tions when examining patients colonized with resistant bacteria. Updated guidelines for preventing intravascular catheter infections were published in 2011 (see Chapter 302).432

The risk for nosocomial pneumonia is greatest in patients who receive mechanical ventilation for longer than 1 week. Randomized trials have shown that the semirecumbent body position reduces the risk for nosocomial pneumonia, especially in patients who receive enteral nutrition.433 Maintaining an adequate intracuff pressure and effective aspiration of subglottic secretions may also be important.434 Avoiding nasal gastrointestinal tubes decreases the risk for developing sinusitis. A closed urinary drainage system is essential. Trials are needed to evaluate prophylaxis to prevent CMV reactivation in criti-cally ill patients.

PROGNOSISWhereas previously healthy young humans almost always (>90%) survive severe sepsis if their disease-causing microbes can be killed and supportive care is provided,435,436 severe sepsis and septic shock have case-fatality rates of approximately 30% and 50%, respectively, in older patients with comorbidities. As initially noted by McCabe and Jackson,435 outcome is significantly (and most profoundly) influenced by the patient’s underlying disease.240,436 Bacteremia with certain microbes (e.g., S. aureus) may also be independently related to mortal-ity in multivariate analyses.437 Of the many studied biologic markers, low monocyte HLA-DR expression, plasma IL-6 levels, and a high IL-10/TNF ratio116,438,439 may correlate best with risk of dying; circulat-ing levels of numerous other parameters (free DNA, thrombomodulin, CRP, procalcitonin, soluble phospholipase A2, others) have also been associated with increased mortality in human severe sepsis. Auto-nomic dysfunction detected by measuring heart rate variability may predict mortality for as long as 60 days.440,441 Prognostic scores based on bedside evaluations, such as the APACHE II, Simplified Acute Physiology Score (SAPS II), and the sequential organ failure assess-ments (SOFA), are easier to use in the typical ICU setting, although their mortality predictions can differ substantially.442,443 Moreover, as noted by Annane and co-workers,444 organ failure scores may have difficulty quantitating the contribution that preexisting organ dysfunc-tion (comorbidity) adds to risk. One group found that battery of endo-crine tests (thyroxine, thyrotropin, cortisol) was a better discriminator of outcome than the APACHE II score.445

Although most clinical trials of sepsis therapies have used 28-day all-cause mortality as the outcome variable, Perl and co-workers436 found that the median day of death was 30.5 days after the onset of sepsis. Another study concluded that patients who survive an episode of severe sepsis have significantly decreased life expectancy over the ensuing 5 years.271 Experiencing severe sepsis may also diminish an individual’s subsequent quality of life271,436 and cognitive function276 for several years.

improve outcome when it was used as adjunctive therapy in non-neutropenic patients with sepsis caused by community-acquired pneu-monia or hospital-acquired pneumonia.417,418 A similar lack of benefit occurred in septic patients given granulocyte-macrophage colony-stimulating factor (GM-CSF).419 A meta-analysis of the clinical trials of G-CSF and GM-CSF found that both agents were associated with recovery from infection without decreasing all-cause or in-hospital mortality.420

Although there is evidence that both passive421 and active422 immu-nization can benefit patients at risk for hospital-acquired infection, neither of these approaches can be recommended for general clinical use. A meta-analysis noted that intravenous polyclonal immunoglobu-lin may decrease the risk of death in sepsis but that most of the clinical trials had methodologic concerns.423 In specific infections, such as S. pyogenes toxic shock syndrome, intravenous immunoglobulin is often used as adjunctive therapy. The available immunoglobulin prepa-rations vary in their content of neutralizing antibodies to streptococcal exotoxins and superantigens, however,424 and the efficacy of this approach is not firmly established.

Although measures that prevent B cell, CD4+ cell, and follicular dendritic cell apoptosis can improve survival in animal models of sepsis,180 this approach has not been translated to the clinical arena. A small prospective trial noted an improvement in mortality when ator-vastatin was continued in prior statin users but not when it was started acutely during an episode of sepsis.425 In contrast, another study found that atorvastatin prevented the progression of sepsis to severe sepsis but did not change mortality when it was started acutely.426

SummaryThree elements are the cornerstones of therapy for septic shock: the rapid administration of antibiotics that are broad in spectrum and target both the species and antibiotic sensitivity of the likely patho-gen(s), prompt removal or drainage of the source of the infection, and the use of fluids and vasopressors to reverse hypotension and tissue hypoperfusion.391 Progress during the last decade has made it possible to consider a trial of low-dose hydrocortisone in patients with pressor-dependent septic shock. Although improvements in clini-cal trial design have reduced the impact of patient heterogeneity by using more restrictive entry criteria and enrolling larger numbers of patients, the nonreproducibility of clinical trial results remains a major problem.400 Using all-cause 28-day mortality as the arbitrary primary end point ignores the long-term consequences of severe infections and intensive care (see “Prognosis”). Clinical trials of combination therapy, using drugs with different mechanisms of action, have not been attempted.427

Nutrition and Other Supportive MeasuresMuch evidence now supports the use of enteral, instead of intravenous, nutrition in critically ill patients. Prophylaxis for GI bleeding, deep venous thrombosis, and decubitus ulcers should be routine. Decubitus ulcers may be prevented by avoiding prolonged skin exposure to stool and urine, by frequent repositioning, and by adequate nutrition. Patients with low bleeding risk should receive low doses of heparin, whereas intermittent compression devices should be applied to the lower extremities of those at risk for bleeding. H2-receptor antagonists are superior to sucralfate or antacids for preventing GI bleeding428; proton-pump inhibitors would be expected to be similarly effective. Sedation should be interrupted at least daily in patients who are

Key ReferencesThe complete reference list is available online at Expert Consult.

3. Dellinger RP, Levy M, Rhodes A, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med. 2013;39:165-228.

6. Rangel-Frausto MS, Pittet D, Costigan M, et al. The natural history of the systemic inflammatory response syndrome (SIRS). JAMA. 1995;273:117-123.

23. Baccala R, Gonzalez-Quintial R, Lawson BR, et al. Sensors of the innate immune system: their mode of action. Nat Rev Rheumatol. 2009;5:448-456.

25. Brinkmann V, Zychlinsky A. Neutrophil extracellular traps: is immunity the second function of chromatin? J Cell Biol. 2012;198:773-783.

26. Medzhitov R, Janeway CA Jr. Decoding the patterns of self and nonself by the innate immune system. Science. 2002;296:298-300.

39. Munford RS, Pugin J. Normal responses to injury prevent systemic inflammation and can be immunosuppressive. Am J Respir Crit Care Med. 2001;163:316-321.

52. Borovikova LV, Ivanova S, Zhang M, et al. Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin. Nature. 2000;405:458-462.

54. Reardon C, Duncan GS, Brustle A, et al. Lymphocyte-derived ACh regulates local innate but not adaptive immu-nity. Proc Natl Acad Sci U S A. 2013;110:1410-1415.

61. McCall CE, Yoza BK. Gene silencing in severe systemic inflammation. Am J Respir Crit Care Med. 2007;175:763-767.

64. Carson WF, Cavassani KA, Dou Y, Kunkel SL. Epigenetic regulation of immune cell functions during post-septic immunosuppression. Epigenetics. 2011;6:273-283.

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

934

65. Lu M, Varley AW, Ohta S, et al. Host inactivation of bac-terial lipopolysaccharide prevents prolonged tolerance following gram-negative bacterial infection. Cell Host Microbe. 2008;4:293-302.

66. Munford R, Lu M, Varley A. Chapter 2: kill the bacteria … and also their messengers? Adv Immunol. 2009;103:29-48.

68. van der Poll T, Coyle SM, Barbosa K, et al. Epinephrine inhibits tumor necrosis factor-a and potentiates interleukin 10 production during human endotoxemia. J Clin Invest. 1996;97:713-719.

71. Dugernier TL, Laterre PF, Wittebole X, et al. Compartmen-talization of the inflammatory response during acute pan-creatitis: correlation with local and systemic complications. Am J Respir Crit Care Med. 2003;168:148-157.

79. Sapolsky RM, Romero LM, Munck AU. How do glucocor-ticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocr Rev. 2000;21:55-89.

82. Levi M, van der Poll T, Schultz M. Systemic versus local-ized coagulation activation contributing to organ failure in critically ill patients. Semin Immunopathol. 2012;34:167-179.

88. Mackowiak PA, Bartlett JG, Borden EC, et al. Concepts of fever: recent advances and lingering dogma. Clin Infect Dis. 1997;25:119-138.

93. Brandtzaeg P, Osnes L, Ovstebo R, et al. Net inflammatory capacity of human septic shock plasma evaluated by a monocyte-based target cell assay: identification of interleukin-10 as a major functional deactivator of human monocytes. J Exp Med. 1996;184:51-60.

112. Chan JK, Roth J, Oppenheim JJ, et al. Alarmins: awaiting a clinical response. J Clin Invest. 2012;122:2711-2719.

136. Takasu O, Gaut JP, Watanabe E, et al. Mechanisms of cardiac and renal dysfunction in patients dying of sepsis. Am J Respir Crit Care Med. 2013;187:509-517.

143. Aird WC. The role of the endothelium in severe sepsis and multiple organ dysfunction syndrome. Blood. 2003;101:3765-3777.

165. Fink M. Cytopathic hypoxia in sepsis. Acta Anaesthesiol Scand. 1997;110:87-95.

169. Fullerton JN, Singer M. Organ failure in the ICU: cellular alterations. Semin Respir Crit Care Med. 2011;32:581-586.

173. Jeger V, Djafarzadeh S, Jakob SM, Takala J. Mitochondrial function in sepsis. Eur J Clin Invest. 2013;43:532-542.

180. Hotchkiss RS, Nicholson DW. Apoptosis and caspases regulate death and inflammation in sepsis. Nat Rev Immunol. 2006;6:813-822.

193. Landry DW, Oliver JA. The pathogenesis of vasodilatory shock. N Engl J Med. 2001;345:588-595.

197. Exline MC, Crouser ED. Mitochondrial mechanisms of sepsis-induced organ failure. Front Biosci. 2008;13:5030-5041.

200. Brouwer MC, de Gans J, Heckenberg SG, et al. Host genetic susceptibility to pneumococcal and meningococcal

disease: a systematic review and meta-analysis. Lancet Infect Dis. 2009;9:31-44.

201. Clark MF, Baudouin SV. A systematic review of the quality of genetic association studies in human sepsis. Intensive Care Med. 2006;32:1706-1712.

208. Munford RS. Severe sepsis and septic shock: the role of gram-negative bacteremia. Annu Rev Pathol. 2006;1:467-496.

214. Boldrick JC, Alizadeh AA, Diehn M, et al. Stereotyped and specific gene expression programs in human innate immune responses to bacteria. Proc Natl Acad Sci U S A. 2002;99:972-977.

233. Kumar A, Ellis P, Arabi Y, et al. Initiation of inappropriate antimicrobial therapy results in a fivefold reduction of sur-vival in human septic shock. Chest. 2009;136:1237-1248.

241. Felty AR, Keefer CS. Bacillus coli sepsis. JAMA. 1924;82:1430-1433.

248. Lu M, Munford RS. The transport and inactivation kinetics of bacterial lipopolysaccharide influence its immunological potency in vivo. J Immunol. 2011;187:3314-3320.

251. Kitchens RL, Thompson PA, Munford RS, O’Keefe GE. Acute inflammation and infection maintain circulating phospholipid levels and enhance lipopolysaccharide bind-ing to plasma lipoproteins. J Lipid Res. 2003;44:2339-2348.

267. Alcaïs A, Abel L, Casanova JL. Human genetics of infec-tious diseases: between proof of principle and paradigm. J Clin Invest. 2009;119:2506-2514.

269. Nathan C, Ding A. Nonresolving inflammation. Cell. 2010;140:871-882.

273. Gofton TE, Young GB. Sepsis-associated encephalopathy. Nat Rev Neurol. 2012;8:557-566.

276. Iwashyna TJ, Ely EW, Smith DM, Langa KM. Long-term cognitive impairment and functional disability among sur-vivors of severe sepsis. JAMA. 2010;304:1787-1794.

285. Marik PE, Pastores SM, Annane D, et al. Recommenda-tions for the diagnosis and management of corticosteroid insufficiency in critically ill adult patients: consensus state-ments from an international task force by the American College of Critical Care Medicine. Crit Care Med. 2008;36:1937-1949.

312. Levy B. Lactate and shock state: the metabolic view. Curr Opin Crit Care. 2006;12:315-321.

316. Wada H, Thachil J, Di Nisio M, et al. Guidance for diagno-sis and treatment of DIC from harmonization of the recommen dations from three guidelines. J Thromb Haemost. 2013;11:761-767.

321. Matthay MA, Ware LB, Zimmerman GA. The acute respira-tory distress syndrome. J Clin Invest. 2012;122:2731-2740.

326. Puleo F, Arvanitakis M, Van Gossum A, Preiser JC. Gut failure in the ICU. Semin Respir Crit Care Med. 2011;32:626-638.

337. Torgersen C, Moser P, Luckner G, et al. Macroscopic post-mortem findings in 235 surgical intensive care patients with sepsis. Anesth Analg. 2009;108:1841-1847.

338. Musher D. Cutaneous and soft-tissue manifestations of sepsis due to gram-negative enteric bacilli. Rev Infect Dis. 1980;2:854-866.

341. Kumar A, Haery C, Parrillo JE. Myocardial dysfunction in septic shock: Part I. Clinical manifestation of cardiovas-cular dysfunction. J Cardiothorac Vasc Anesth. 2001;15:364-376.

352. Riedel S, Carroll KC. Blood cultures: key elements for best practices and future directions. J Infect Chemother. 2010;16:301-316.

353. Mancini N, Carletti S, Ghidoli N, et al. The era of molecular and other non-culture-based methods in diagnosis of sepsis. Clin Microbiol Rev. 2010;23:235-251.

358. Pierrakos C, Vincent JL. Sepsis biomarkers: a review. Crit Care. 2010;14:R15.

361. Minneci PC, Deans KJ, Eichacker PQ, Natanson C. The effects of steroids during sepsis depend on dose and sever-ity of illness: an updated meta-analysis. Clin Microbiol Infect. 2009;15:308-318.

366. Leibovici L, Paul M, Poznanski O, et al. Monotherapy versus beta-lactam-aminoglycoside combination treat-ment for gram-negative bacteremia: a prospective, obser-vational study. Antimicrob Agents Chemother. 1997;41:1127-1133.

370. Greisman SE, DuBuy JB, Woodward CL. Experimental gram-negative bacterial sepsis: prevention of mortality not preventable by antibiotics alone. Infect Immun. 1979;25:538-557.

371. Kumar A, Roberts D, Wood KE, et al. Duration of hypoten-sion before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med. 2006;34:1589-1596.

376. Eichacker PQ, Parent C, Kalil A, et al. Risk and the efficacy of antiinflammatory agents: retrospective and confirma-tory studies of sepsis. Am J Respir Crit Care Med. 2002;166:1197-1205.

378. Drusano GL, Lodise TP. Editorial commentary: saving lives with optimal antimicrobial chemotherapy. Clin Infect Dis. 2013;56:245-247.

381. Marshall JC, Maier RV, Jimenez M, Dellinger EP. Source control in the management of severe sepsis and septic shock: an evidence-based review. Crit Care Med. 2004;32:S513-S526.

382. Natanson C, Danner RL, Reilly JM, et al. Antibiotics versus cardiovascular support in a canine model of human septic shock. Am J Physiol. 1990;259:H1440-H1447.

385a. Caironi P, Tognoni G, Masson S, et al. Albumin replace-ment in patients with severe sepsis or septic shock. N Engl J Med. 2014 Mar 18. [Epub ahead of print]

391. Suffredini AF, Munford RS. Novel therapies for septic shock over the past 4 decades. JAMA. 2011;306:194-199.

391a. The ProCESS Investigators. A randomized trial of proto-col-based care for early septic shock. N Engl J Med. 2014 Mar 18. [Epub ahead of print]

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

934.e1

References1. American College of Chest Physicians/Society of Critical

Care Medicine Consensus Conference Committee. Defini-tions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med. 1992;20:864-874.

2. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med. 2003;31:1250-1256.

3. Dellinger RP, Levy M, Rhodes A, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med. 2013;39:165-228.

4. Bone RC, Grodzin CJ, Balk RA. Sepsis: a new hypothesis for pathogenesis of the disease process. Chest. 1997;112:235-243.

5. Zhao H, Heard SO, Mullen MT, et al. An evaluation of the diagnostic accuracy of the 1991 American College of Chest Physicians/Society of Critical Care Medicine and the 2001 Society of Critical Care Medicine/European Society of Intensive Care Medicine/American College of Chest Physicians/American Thoracic Society/Surgical Infection Society sepsis definition. Crit Care Med. 2012;40:1700-1706.

6. Rangel-Frausto MS, Pittet D, Costigan M, et al. The natural history of the systemic inflammatory response syndrome (SIRS). JAMA. 1995;273:117-123.

7. Alberti C, Brun-Buisson C, Chevret S, et al. Systemic inflammatory response and progression to severe sepsis in critically ill infected patients. Am J Respir Crit Care Med. 2005;171:461-468.

8. Angus DC, Wax RS. Epidemiology of sepsis: an update. Crit Care Med. 2001;29:S109-S116.

9. Gaieski DF, Edwards JM, Kallan MJ, Carr BG. Benchmark-ing the incidence and mortality of severe sepsis in the United States. Crit Care Med. 2013;41:1167-1174.

10. Martin GS, Mannino DM, Eaton S, Moss M. The epidemi-ology of sepsis in the United States from 1979 through 2000. N Engl J Med. 2003;348:1546-1554.

11. Lagu T, Rothberg MB, Shieh MS, et al. Hospitalizations, costs, and outcomes of severe sepsis in the United States 2003 to 2007. Crit Care Med. 2012;40:754-761.

12. Minino AM, Murphy SL, Xu J, Kochanek KD. Deaths: final data for 2008. Natl Vital Stat Rep. 2011;59:1-126.

13. Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiol-ogy of severe sepsis in the United States: analysis of inci-dence, outcome, and associated costs of care. Crit Care Med. 2001;29:1303-1310.

14. Sands KE, Bates DW, Lanken PN, et al. Epidemiology of sepsis syndrome in 8 academic medical centers. JAMA. 1997;278:234-240.

15. Alberti C, Brun-Buisson C, Goodman SV, et al. Influence of systemic inflammatory response syndrome and sepsis on outcome of critically ill infected patients. Am J Respir Crit Care Med. 2003;168:77-84.

16. Vincent JL, Sakr Y, Sprung CL, et al. Sepsis in European intensive care units: results of the SOAP study. Crit Care Med. 2006;34:344-353.

17. Brun-Buisson C, Doyon F, Carlet J, et al. Incidence, risk factors, and outcome of severe sepsis and septic shock in adults. A multicenter prospective study in intensive care units. JAMA. 1995;274:968-974.

18. Zanotti-Cavazzoni SL, Goldfarb RD. Animal models of sepsis. Crit Care Clin. 2009;25:703-719.

19. Ward PA, Bosmann M. A historical perspective on sepsis. Am J Pathol. 2012;181:2-7.

20. Seok J, Warren HS, Cuenca AG, et al. Genomic responses in mouse models poorly mimic human inflammatory dis-eases. Proc Natl Acad Sci U S A. 2013;110:3507-3512.

21. Copeland S, Warren HS, Lowry SF, et al; Inflammation and the Host Response to Injury Investigators. Acute inflam-matory response to endotoxin in mice and humans. Clin Vaccine Immunol. 2005;12:60-67.

22. Ishii KJ, Koyama S, Nakagawa A, et al. Host innate immune receptors and beyond: making sense of microbial infec-tions. Cell Host Microbe. 2008;3:352-363.

23. Baccala R, Gonzalez-Quintial R, Lawson BR, et al. Sensors of the innate immune system: their mode of action. Nat Rev Rheumatol. 2009;5:448-456.

24. Kim HM, Park BS, Kim JI, et al. Crystal structure of the TLR4-MD-2 complex with bound endotoxin antagonist Eritoran. Cell. 2007;130:906-917.

25. Brinkmann V, Zychlinsky A. Neutrophil extracellular traps: is immunity the second function of chromatin? J Cell Biol. 2012;198:773-783.

26. Medzhitov R, Janeway CA Jr. Decoding the patterns of self and nonself by the innate immune system. Science. 2002;296:298-300.

27. Montminy SW, Khan N, McGrath S, et al. Virulence factors of Yersinia pestis are overcome by a strong lipopolysac-charide response. Nat Immunol. 2006;7:1066-1073.

28. Hajjar AM, Harvey MD, Shaffer SA, et al. Lack of in vitro and in vivo recognition of Francisella tularensis subspecies

lipo polysaccharide by Toll-like receptors. Infect Immun. 2006;74:6730-6738.

29. von Bernuth H, Picard C, Jin Z, et al. Pyogenic bacterial infections in humans with MyD88 deficiency. Science. 2008;321:691-696.

30. Ku CL, von Bernuth H, Picard C, et al. Selective predisposi-tion to bacterial infections in IRAK-4-deficient children: IRAK-4-dependent TLRs are otherwise redundant in pro-tective immunity. J Exp Med. 2007;204:2407-2422.

31. Yang D, Postnikov YV, Li Y, et al. High-mobility group nucleosome-binding protein 1 acts as an alarmin and is critical for lipopolysaccharide-induced immune responses. J Exp Med. 2012.

32. Kondo Y, Ikeda K, Tokuda N, et al. TLR4-MD-2 complex is negatively regulated by an endogenous ligand, globote-traosylceramide. Proc Natl Acad Sci U S A. 2013;110:4714-4719.

33. Giamarellos-Bourboulis EJ, van de Veerdonk FL, Mouk-taroudi M, et al. Inhibition of caspase-1 activation in Gram-negative sepsis and experimental endotoxemia. Crit Care. 2011;15:R27.

34. Koc B, Oktenli C, Bulucu F, et al. The rate of pyrin muta-tions in critically ill patients with systemic inflammatory response syndrome and sepsis: a pilot study. J Rheumatol. 2007;34:2070-2075.

35. Hall MW, Gavrilin MA, Knatz NL, et al. Monocyte mRNA phenotype and adverse outcomes from pediatric multiple organ dysfunction syndrome. Pediatr Res. 2007;62:597-603.

36. Li P, Allen H, Banerjee S, et al. Mice deficient in IL-1b-converting enzyme are defective in production of mature IL-1b and resistant to endotoxic shock. Cell. 1995;80:401-411.

37. Dolinay T, Kim YS, Howrylak J, et al. Inflammasome-regulated cytokines are critical mediators of acute lung injury. Am J Respir Crit Care Med. 2012;185:1225-1234.

38. Lamkanfi M, Sarkar A, Vande Walle L, et al. Inflammasome-dependent release of the alarmin HMGB1 in endotoxemia. J Immunol. 2010;185:4385-4392.

39. Munford RS, Pugin J. Normal responses to injury prevent systemic inflammation and can be immunosuppressive. Am J Respir Crit Care Med. 2001;163:316-321.

40. Liu T, Gao Y-J, Ji R-R. Emerging role of Toll-like receptors in the control of pain and itch. Neurosci Bull. 2012;28:131-144.

41. Acosta C, Davies A. Bacterial lipopolysaccharide regulates nociceptin expression in sensory neurons. J Neurosci Res. 2008;86:1077-1086.

42. Barajon I, Serrao G, Arnaboldi F, et al. Toll-like receptors 3, 4, and 7 are expressed in the enteric nervous system and dorsal root ganglia. J Histochem Cytochem. 2009;57:1013-1023.

43. Wadachi R, Hargreaves KM. Trigeminal nociceptors express TLR-4 and CD14: a mechanism for pain due to infection. J Dent Res. 2006;85:49-53.

44. Netea MG, Kullberg B-J, Van der Meer JWM. Circulating cytokines as mediators of fever. Clin Infect Dis. 2000;31:S178-S184.

45. Cavaillon JM. Good and bad fever. Crit Care. 2012;16:119.46. Moore FA, Moore EE, Poggetti R, et al. Gut bacterial trans-

location via the portal vein: a clinical perspective with major torso trauma. J Trauma. 1991;31:629-636.

47. Shao B, Lu M, Katz SC, et al. A host lipase detoxifies bacte-rial lipopolysaccharides in the liver and spleen. J Biol Chem. 2007;282:13726-13735.

48. Leon LR, Kozak W, Peschon J, et al. Altered acute phase responses to inflammation in IL-1 and TNF receptor knockout mice. Ann N Y Acad Sci. 1997;813:244-254.

49. Belge KU, Dayyani F, Horelt A, et al. The proinflammatory CD14+CD16+DR++ monocytes are a major source of TNF. J Immunol. 2002;168:3536-3542.

50. Fong Y, Marano MA, Moldawer LL, et al. The acute splanchnic and peripheral tissue metabolic response to endotoxin in humans. J Clin Invest. 1990;85:1896-1904.

51. Romanovsky AA, Ivanov AI, Székely M. Neural route of pyrogen signaling to the brain. Clin Infect Dis. 2000;31:S162-S167.

52. Borovikova LV, Ivanova S, Zhang M, et al. Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin. Nature. 2000;405:458-462.

53. Huston JM, Ochani M, Rosas-Ballina M, et al. Splenectomy inactivates the cholinergic antiinflammatory pathway during lethal endotoxemia and polymicrobial sepsis. J Exp Med. 2006;203:1623-1628.

54. Reardon C, Duncan GS, Brustle A, et al. Lymphocyte-derived ACh regulates local innate but not adaptive immu-nity. Proc Natl Acad Sci U S A. 2013;110:1410-1415.

55. Pena G, Cai B, Ramos L, et al. Cholinergic regulatory lym-phocytes re-establish neuromodulation of innate immune responses in sepsis. J Immunol. 2011;187:718-725.

56. Munford RS, Pugin J. The crucial role of systemic responses in the innate (non-adaptive) host defense. J Endotoxin Res. 2001;7:327-332.

57. Davis HM, Carpenter DC, Stahl JM, et al. Human granu-locyte CD11b expression as a pharmacodynamic bio-marker of inflammation. J Immunol Methods. 2000;240:125-132.

58. Drakesmith H, Prentice AM. Hepcidin and the iron-infection axis. Science. 2012;338:768-772.

59. Didierlaurent A, Goulding J, Patel S, et al. Sustained desen-sitization to bacterial Toll-like receptor ligands after resolu-tion of respiratory influenza infection. J Exp Med. 2008;205:323-329.

60. Cross AS. Endotoxin tolerance—current concepts in his-torical perspective. J Endotoxin Res. 2002;8:83-98.

61. McCall CE, Yoza BK. Gene silencing in severe systemic inflammation. Am J Respir Crit Care Med. 2007;175:763-767.

62. Foster SL, Hargreaves DC, Medzhitov R. Gene-specific control of inflammation by TLR-induced chromatin modi-fications. Nature. 2007;447:972-978.

63. El Gazzar M, McCall CE. MicroRNAs distinguish transla-tional from transcriptional silencing during endotoxin tolerance. J Biol Chem. 2010;285:20940-20951.

64. Carson WF, Cavassani KA, Dou Y, Kunkel SL. Epigenetic regulation of immune cell functions during post-septic immunosuppression. Epigenetics. 2011;6:273-283.

65. Lu M, Varley AW, Ohta S, et al. Host inactivation of bacte-rial lipopolysaccharide prevents prolonged tolerance fol-lowing gram-negative bacterial infection. Cell Host Microbe. 2008;4:293-302.

66. Munford R, Lu M, Varley A. Chapter 2: kill the bacteria … and also their messengers? Adv Immunol. 2009;103:29-48.

67. Brudecki L, Ferguson DA, McCall CE, El Gazzar M. Myeloid-derived suppressor cells evolve during sepsis and can enhance or attenuate the systemic inflammatory response. Infect Immun. 2012;80:2026-2034.

68. van der Poll T, Coyle SM, Barbosa K, et al. Epinephrine inhibits tumor necrosis factor-a and potentiates interleukin 10 production during human endotoxemia. J Clin Invest. 1996;97:713-719.

69. Barber AE, Coyle SM, Marano MA, et al. Glucocorticoid therapy alters hormonal and cytokine responses to endo-toxin in man. J Immunol. 1993;150:1999-2006.

70. Lemaire LC, van der Poll T, Van Lanschot JB, et al. Mini-mally invasive surgery induces endotoxin-tolerance in the absence of detectable endotoxemia. J Clin Immunol. 1998;18:414-420.

71. Dugernier TL, Laterre PF, Wittebole X, et al. Compartmen-talization of the inflammatory response during acute pan-creatitis: correlation with local and systemic complications. Am J Respir Crit Care Med. 2003;168:148-157.

72. Rivera-Chavez FA, Wheeler H, Lindberg G, et al. Regional and systemic cytokine responses to acute inflammation of the vermiform appendix. Ann Surg. 2003;237:408-416.

73. Kurahashi K, Kajikawa O, Sawa T, et al. Pathogenesis of septic shock in Pseudomonas aeruginosa pneumonia. J Clin Invest. 1999;104:743-750.

74. Matute-Bello G, Frevert CW, Kajikawa O, et al. Septic shock and acute lung injury in rabbits with peritonitis: failure of the neutrophil response to localized infection. Am J Respir Crit Care Med. 2001;163:234-243.

75. Xiao W, Mindrinos MN, Seok J, et al. A genomic storm in critically injured humans. J Exp Med. 2011;208:2581-2590.

76. Bone RC. Immunologic dissonance: a continuing evolution in our understanding of the systemic inflammatory response syndrome (SIRS) and the multiple organ dysfunc-tion syndrome (MODS). Ann Intern Med. 1996;125:680-687.

77. Dhar A, Castillo L. Insulin resistance in critical illness. Curr Opin Pediatr. 2011;23:269-274.

78. Foster DW, McGarry JD. Glucose, lipid and protein metab-olism. In: Griffin JE, Ojeda SR, eds. Textbook of Endocrine Physiology. 4th ed. New York: Oxford University Press; 2000:393-420.

79. Sapolsky RM, Romero LM, Munck AU. How do glucocor-ticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocr Rev. 2000;21:55-89.

80. Annane D, Bellisant E, Sebille V, et al. Impaired pressor sensitivity to noradrenaline in septic shock patients with and without impaired adrenal function reserve. Br J Clin Pharmacol. 1998;46:589-597.

81. Bernardin G, Strosberg AD, Bernard A, et al. Beta-adrenergic receptor-dependent and -independent stimula-tion of adenylate cyclase is impaired during severe sepsis in humans. Intensive Care Med. 1998;24:1315-1322.

82. Levi M, van der Poll T, Schultz M. Systemic versus localized coagulation activation contributing to organ failure in criti-cally ill patients. Semin Immunopathol. 2012;34:167-179.

83. van der Poll T, Levi M. Crosstalk between inflammation and coagulation: the lessons of sepsis. Curr Vasc Pharma-col. 2012;10:632-638.

84. Gando S, Kameue T, Nanzaki S, et al. Participation of tissue factor and thrombin in posttraumatic systemic inflamma-tory syndrome. Crit Care Med. 1997;25:1820-1826.

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

934.e2

85. Levi M, Ten Cate H. Disseminated intravascular coagula-tion. N Engl J Med. 1999;341:586-592.

86. Faust SN, Levin M, Harrison OB, et al. Dysfunction of endothelial protein C activation in severe meningococcal sepsis. N Engl J Med. 2001;345:408-416.

87. Asakura H, Ontachi Y, Mizutani T, et al. Decreased plasma activity of antithrombin or protein C is not due to con-sumption coagulopathy in septic patients with dissemi-nated intravascular coagulation. Eur J Haematol. 2001;67:170-175.

88. Mackowiak PA, Bartlett JG, Borden EC, et al. Concepts of fever: recent advances and lingering dogma. Clin Infect Dis. 1997;25:119-138.

89. Greisman SE, Hornick RB. On the demonstration of circu-lating human endogenous pyrogen. Proc Soc Exp Biol Med. 1972;139:690-697.

90. Ross G, Roth J, Störr B, et al. Afferent nerves are involved in the febrile response to injection of LPS into artificial subcutaneous chambers in guinea pigs. Physiol Behav. 2000;71:305-313.

91. Rosenspire AJ, Kindzelskii AL, Petty HR. Cutting edge: fever-associated temperatures enhance neutrophil responses to lipopolysaccharide: a potential mechanism involving cell metabolism. J Immunol. 2002;169:5396-5400.

92. Majetschak M, Flach R, Kreuzfelder E, et al. The extent of traumatic damage determines a graded depression of the endotoxin responsiveness of peripheral blood mononu-clear cells from patients with blunt injuries. Crit Care Med. 1999;27:313-318.

93. Brandtzaeg P, Osnes L, Ovstebo R, et al. Net inflammatory capacity of human septic shock plasma evaluated by a monocyte-based target cell assay: identification of interleukin-10 as a major functional deactivator of human monocytes. J Exp Med. 1996;184:51-60.

94. Ertel W, Kremer J-P, Kenney J, et al. Downregulation of proinflammatory cytokine release in whole blood from septic patients. Blood. 1995;85:1341-1347.

95. Randow F, Syrbe U, Meisel C, et al. Mechanism of endotoxin desensitization: involvement of interleukin 10 and trans-forming growth factor b. J Exp Med. 1995;181:1887-1892.

96. Munoz C, Carlet J, Fitting C, et al. Dysregulation of in vitro cytokine production by monocytes during sepsis. J Clin Invest. 1991;88:1747-1754.

97. Sfeir T, Saha DC, Astiz M, Rackow EC. Role of interleukin-10 in monocyte hyporesponsiveness associated with septic shock. Crit Care Med. 2001;29:129-133.

98. Spolarics Z, Siddiqi M, Siegel JH, et al. Depressed interleukin-12-producing activity by monocytes correlates with adverse clinical course and a shift toward Th2-type lymphocyte pattern in severely injured male trauma patients. Crit Care Med. 2003;31:1722-1729.

99. Ertel W, Keel M, Neidhardt R, et al. Inhibition of the defense system stimulating interleukin-12 interferon-gamma pathway during critical illness. Blood. 1997;89:1612-1620.

100. Tilg H, Dinarello CA, Mier JW. IL-6 and APPs: anti-inflammatory and immunosuppressive mediators. Immunol Today. 1997;9:428-432.

101. Steensberg A, Fischer CP, Keller C, et al. IL-6 enhances plasma IL-1ra, IL-10, and cortisol in humans. Am J Physiol Endocrinol Metab. 2003;285:E433-E437.

102. Bethin KE, Vogt SK, Muglia LJ. Interleukin-6 is an essen-tial, corticotropin-releasing hormone-independent stimu-lator of the adrenal axis during immune system activation. Proc Natl Acad Sci USA. 2000;97:9317-9322.

103. van der Poll T, Keogh CV, Guirao X, et al. Interleukin-6 gene-deficient mice show impaired defense against pneu-mococcal pneumonia. J Infect Dis. 1997;176:439-444.

104. Xing Z, Gauldie J, Cox G, et al. IL-6 is an antiinflammatory cytokine required for controlling local or systemic acute inflammatory responses. J Clin Invest. 1998;101:311-320.

105. Kopf M, Baumann H, Freer G, et al. Impaired immune and acute-phase responses in interleukin-6-deficient mice. Nature. 1994;368:339-342.

106. Bucala R. MIF rediscovered: cytokine, pituitary hormone, and glucocorticoid-induced regulator of the immune response. FASEB J. 1996;10:1607-1613.

107. Calandra T, Echtenacher B, Le Roy D, et al. Protection from septic shock by neutralization of macrophage migration inhibitory factor. Nature Med. 2000;6:164-170.

108. Calandra T, Froidevaux W, Martin C, Roger T. Macrophage migration inhibitory factor and host innate immune defenses against bacterial sepsis. J Infect Dis. 2003;187:S385-S390.

109. Honma N, Koseki H, Akasaka T, et al. Deficiency of the macrophage migration inhibitory factor gene has no signifi-cant effect on endotoxaemia. Immunology. 2000;100:84-90.

110. Wang H, Bloom O, Zhang M, et al. HMG-1 as a late media-tor of endotoxin lethality in mice. Science. 1999;285:248-251.

111. Rouhiainen A, Tumova S, Valmu L, et al. Analysis of proinflammatory activity of highly purified eukaryotic recombinant HMGB1 (amphoterin). J Leukoc Biol. 2007;81:49-58.

112. Chan JK, Roth J, Oppenheim JJ, et al. Alarmins: awaiting a clinical response. J Clin Invest. 2012;122:2711-2719.

113. Angus DC, Yang L, Kong L, et al. Circulating high-mobility group box 1 (HMGB1) concentrations are elevated in both uncomplicated pneumonia and pneumonia with severe sepsis. Crit Care Med. 2007;35:1061-1067.

114. Sunden-Cullberg J, Norrby-Teglund A, Rouhiainen A, et al. Persistent elevation of high mobility group box-1 protein (HMGB1) in patients with severe sepsis and septic shock. Crit Care Med. 2005;33:564-573.

115. Suffredini AF, Reda D, Banks SM, et al. Effects of recombi-nant dimeric TNF receptor on human inflammatory responses following intravenous endotoxin administration. J Immunol. 1995;155:5038-5045.

116. Van Dissel JT, van Langevelde P, Westendorp RG, et al. Anti-inflammatory cytokine profile and mortality in febrile patients. Lancet. 1998;351:950-953.

117. Rose WE, Eickhoff JC, Shukla SK, et al. Elevated serum interleukin-10 at time of hospital admission is predictive of mortality in patients with Staphylococcus aureus bacte-remia. J Infect Dis. 2012;206:1604-1611.

118. Reid VL, Webster NR. Role of microparticles in sepsis. Br J Anaesth. 2012;109:503-513.

119. Zhang Q, Itagaki K, Hauser CJ. Mitochondrial DNA is released by shock and activates neutrophils via p38 map kinase. Shock. 2010;34:55-59.

120. Zhang Q, Raoof M, Chen Y, et al. Circulating mitochon-drial DAMPs cause inflammatory responses to injury. Nature. 2010;464:104-107.

121. Crouser ED, Shao G, Julian MW, et al. Monocyte activation by necrotic cells is promoted by mitochondrial proteins and formyl peptide receptors. Crit Care Med. 2009;37:2000-2009.

122. Sursal T, Stearns-Kurosawa DJ, Itagaki K, et al. Plasma bacterial and mitochondrial DNA distinguish bacterial sepsis from sterile systemic inflammatory response syn-drome and quantify inflammatory tissue injury in nonhu-man primates. Shock. 2013;39:55-62.

123. Wolbink GJ, Bossink AW, Groeneveld AB, et al. Com-plement activation in patients with sepsis is in part mediated by C-reactive protein. J Infect Dis. 1998;177:81-87.

124. Brandtzaeg P, Hogasen K, Kierulf P, Mollnes TE. The exces-sive complement activation in fulminant meningococcal septicemia is predominantly caused by alternative pathway activation. J Infect Dis. 1996;173:647-655.

125. Nuijens JH, Eerenberg-Belmer AJ, Huijbregts CC, et al. Proteolytic inactivation of plasma C1 inhibitor in sepsis. J Clin Invest. 1989;84:443-450.

126. Jansen PM, Eisele B, De Jong IW, et al. Effect of Cl inhibitor on inflammatory and physiologic response patterns in pri-mates suffering from lethal septic shock. J Immunol. 1998;160:475-484.

127. Liu D, Lu F, Qin G, et al. C1 inhibitor-mediated protection from sepsis. J Immunol. 2007;179:3966-3972.

128. Riedemann NC, Guo RF, Ward PA. Novel strategies for the treatment of sepsis. Nat Med. 2003;9:517-524.

129. Schouten M, Wiersinga WJ, Levi M, van der Poll T. Inflam-mation, endothelium, and coagulation in sepsis. J Leukoc Biol. 2008;83:536-545.

130. Aras O, Shet A, Bach RR, et al. Induction of microparticle- and cell-associated intravascular tissue factor in human endotoxemia. Blood. 2004;103:4545-4553.

131. Fourrier F, Chopin C, Goudemand J, et al. Septic shock, multiple organ failure, and disseminated intravascular coagulation. Compared patterns of antithrombin III, protein C, and protein S deficiencies. Chest. 1992;101:816-823.

132. McGovern VJ. Shock revisited. Pathol Annu. 1984;19(pt 1):15-36.

133. Robboy SJ, Major MC, Colman RW, Minna JD. Pathology of disseminated intravascular coagulation (DIC). Analysis of 26 cases. Hum Pathol. 1972;3:327-343.

134. Hotchkiss RS, Karl IE. Medical progress: the pathophysiol-ogy and treatment of sepsis. N Engl J Med. 2003;348:138-150.

135. Tinsley KW, Grayson MH, Swanson PE, et al. Sepsis induces apoptosis and profound depletion of splenic inter-digitating and follicular dendritic cells. J Immunol. 2003;171:909-914.

136. Takasu O, Gaut JP, Watanabe E, et al. Mechanisms of cardiac and renal dysfunction in patients dying of sepsis. Am J Respir Crit Care Med. 2013;187:509-517.

137. Abraham E, Reinhart K, Opal S, et al. Efficacy and safety of tifacogin (recombinant tissue factor pathway inhibitor) in severe sepsis: a randomized controlled trial. JAMA. 2003;290:238-247.

138. Marti-Carvajal AJ, Sola I, Gluud C, et al. Human recombi-nant protein C for severe sepsis and septic shock in adult and paediatric patients. Cochrane Database Syst Rev. 2012;12:CD004388.

139. Willis TM, Hopp RJ, Romero JR, Larsen PD. The protective effect of brachial plexus palsy in purpura fulminans. Pediatr Neurol. 2001;24:379-381.

140. Rao AK, Schapira M, Clements ML, et al. A prospective study of platelets and plasma proteolytic systems during the early stages of Rocky Mountain spotted fever. N Engl J Med. 1988;318:1021-1028.

141. Fisher-Hoch SP, Platt GS, Neild GH, et al. Pathophysiology of shock and hemorrhage in a fulminating viral infection (Ebola). J Infect Dis. 1985;152:887-894.

142. Bannerman DD, Goldblum SE. Mechanisms of bacterial lipopolysaccharide-induced endothelial apoptosis. Am J Physiol Lung Cell Mol Physiol. 2003;284:L899-L914.

143. Aird WC. The role of the endothelium in severe sepsis and multiple organ dysfunction syndrome. Blood. 2003;101:3765-3777.

144. Mutunga M, Fulton B, Bullock R, et al. Circulating endo-thelial cells in patients with septic shock. Am J Respir Crit Care Med. 2001;163:195-200.

145. McGill SN, Ahmed NA, Christou NV. Increased plasma von Willebrand factor in the systemic inflammatory response syndrome is derived from generalized endothelial cell activation. Crit Care Med. 1998;26:296-300.

146. Yano K, Liaw PC, Mullington JM, et al. Vascular endothe-lial growth factor is an important determinant of sepsis morbidity and mortality. J Exp Med. 2006;203:1447-1458.

147. Calfee CS, Gallagher D, Abbott J, et al. Plasma angiopoietin-2 in clinical acute lung injury: prognostic and pathogenetic significance. Crit Care Med. 2012;40:1731-1737.

148. Helliwell TR, Wilkinson A, Griffiths RD, et al. Microvas-cular endothelial activation in the skeletal muscles of patients with multiple organ failure. J Neurol Sci. 1998;154:26-34.

149. Kubli S, Boegli Y, Ave AD, et al. Endothelium-dependent vasodilation in the skin microcirculation of patients with septic shock. Shock. 2003;19:274-280.

150. Leone M, Boutiere B, Camoin-Jau L, et al. Systemic endo-thelial activation is greater in septic than in traumatic-hemorrhagic shock but does not correlate with endothelial activation in skin biopsies. Crit Care Med. 2002;30:808-814.

151. Aird WC. Endothelial cell heterogeneity. Crit Care Med. 2003;31:S221-S230.

152. Hotchkiss RS, Swanson PE, Freeman BD, et al. Apoptotic cell death in patients with sepsis, shock, and multiple organ dysfunction. Crit Care Med. 1999;27:1230-1251.

153. Brealey D, Karyampudi S, Jacques TS, et al. Mitochondrial dysfunction in a long-term rodent model of sepsis and organ failure. Am J Physiol Regul Integr Comp Physiol. 2004;286:R491-R497.

154. Trzeciak S, Cinel I, Phillip Dellinger R, et al. Resuscitating the microcirculation in sepsis: the central role of nitric oxide, emerging concepts for novel therapies, and chal-lenges for clinical trials. Acad Emerg Med. 2008;15:399-413.

155. De Backer D, Creteur J, Preiser JC, et al. Microvascular blood flow is altered in patients with sepsis. Am J Respir Crit Care Med. 2002;166:98-104.

156. Spronk PE, Ince C, Gardien MJ, et al. Nitroglycerin in septic shock after intravascular volume resuscitation. Lancet. 2002;360:1395-1396.

157. Sakr Y, Dubois MJ, De Backer D, et al. Persistent microcir-culatory alterations are associated with organ failure and death in patients with septic shock. Crit Care Med. 2004;32:1825-1831.

158. Robin ED. Special report: dysoxia. Abnormal tissue oxygen utilization. Arch Intern Med. 1977;137:905-910.

159. Lehr HA, Bittinger F, Kirkpatrick CJ. Microcirculatory dysfunction in sepsis: a pathogenetic basis for therapy? J Pathol. 2000;190:373-386.

160. Cree MG, Fram RY, Herndon DN, et al. Human mitochon-drial oxidative capacity is acutely impaired after burn trauma. Am J Surg. 2008;196:234-239.

161. O’Neill LA, Hardie DG. Metabolism of inflammation limited by AMPK and pseudo-starvation. Nature. 2013;493:346-355.

162. Krawczyk CM, Holowka T, Sun J, et al. Toll-like receptor-induced changes in glycolytic metabolism regulate den-dritic cell activation. Blood. 2010;115:4742-4749.

163. Ward P, Thompson C. Metabolic reprogramming: a cancer hallmark even Warburg did not anticipate. Cancer Cell. 2012;21:297-308.

164. Garedew A, Henderson SO, Moncada S. Activated macro-phages utilize glycolytic ATP to maintain mitochondrial membrane potential and prevent apoptotic cell death. Cell Death Differ. 2010;17:1540-1550.

165. Fink M. Cytopathic hypoxia in sepsis. Acta Anaesthesiol Scand. 1997;110:87-95.

166. Singer M, De Santis V, Vitale D, Jeffcoate W. Multiorgan failure is an adaptive, endocrine-mediated, metabolic response to overwhelming systemic inflammation. Lancet. 2004;364:545-548.

167. Liaudet L. Poly(adenosine 5′-diphosphate) ribose poly-merase activation as a cause of metabolic dysfunction in critical illness. Curr Opin Clin Nutr Metab Care. 2002;5:175-184.

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

934.e3

168. Jones SA, Mills KH, Harris J. Autophagy and inflammatory diseases. Immunol Cell Biol. 2013;91:250-258.

169. Fullerton JN, Singer M. Organ failure in the ICU: cellular alterations. Semin Respir Crit Care Med. 2011;32:581-586.

170. Brealey D, Brand M, Hargreaves I, et al. Association between mitochondrial dysfunction and severity and outcome of septic shock. Lancet. 2002;360:219-223.

171. Adrie C, Bachelet M, Vayssier-Taussat M, et al. Mitochon-drial membrane potential and apoptosis peripheral blood monocytes in severe human sepsis. Am J Respir Crit Care Med. 2001;164:389-395.

172. Carre JE, Orban JC, Re L, et al. Survival in critical illness is associated with early activation of mitochondrial biogen-esis. Am J Respir Crit Care Med. 2010;182:745-751.

173. Jeger V, Djafarzadeh S, Jakob SM, Takala J. Mitochondrial function in sepsis. Eur J Clin Invest. 2013;43:532-542.

174. Christou NV, Meakins JL, Gordon J, et al. The delayed hypersensitivity response and host resistance in surgical patients. 20 years later. Ann Surg. 1995;222:534-548.

175. Manjuck J, Saha DC, Astiz M, et al. Decreased response to recall antigens is associated with depressed costimulatory receptor expression in septic critically ill patients. J Lab Clin Med. 2000;135:153-160.

176. Fumeaux T, Pugin J. Role of interleukin-10 in the intracel-lular sequestration of human leukocyte antigen-DR in monocytes during septic shock. Am J Respir Crit Care Med. 2002;166:1475-1482.

177. McCall CE, Grosso-Wilmoth LM, LaRue K, et al. Tolerance to endotoxin-induced expression of the interleukin-1 beta gene in blood neutrophils of humans with the sepsis syn-drome. J Clin Invest. 1993;91:853-861.

178. Marie C, Muret J, Fitting C, et al. Reduced ex vivo interleukin-8 production by neutrophils in septic and non-septic systemic inflammatory response syndrome. Blood. 1998;91:3439-3446.

179. Cavaillon JM, Adrie C, Fitting C, Adib-Conquy M. Endo-toxin tolerance: is there a clinical relevance? J Endotoxin Res. 2003;9:101-107.

180. Hotchkiss RS, Nicholson DW. Apoptosis and caspases regulate death and inflammation in sepsis. Nat Rev Immunol. 2006;6:813-822.

181. Boomer JS, To K, Chang KC, et al. Immunosuppression in patients who die of sepsis and multiple organ failure. JAMA. 2011;306:2594-2605.

182. Oberholzer C, Oberholzer A, Clare-Salzler M, Moldawer LL. Apoptosis in sepsis: a new target for therapeutic explo-ration. FASEB J. 2001;15:879-892.

183. Hotchkiss RS, Tinsley KW, Swanson PE, et al. Prevention of lymphocyte cell death in sepsis improves survival in mice. Proc Natl Acad Sci U S A. 1999;96:14541-14546.

184. Iwata A, Stevenson VM, Minard A, et al. Over-expression of Bcl-2 provides protection in septic mice by a trans effect. J Immunol. 2003;171:3136-3141.

185. Keel M, Ungethüm U, Steckholzer U, et al. Interleukin-10 counterregulates proinflammatory cytokine-induced inhi-bition of neutrophil apoptosis during severe sepsis. Blood. 1997;90:3356-3363.

186. Hallin GW, Simpson SQ, Crowell RE, et al. Cardiopulmo-nary manifestations of hantavirus pulmonary syndrome. Crit Care Med. 1996;24:252-258.

187. Wright WF, Pritt BS. Update: the diagnosis and manage-ment of dengue virus infection in North America. Diagn Microbiol Infect Dis. 2012;73:215-220.

188. Findling JW, Waters VO, Raff H. The dissociation of renin and aldosterone during critical illness. J Clin Endocrinol Metab. 1987;64:592-595.

189. Zipser RD, Davenport MW, Martin KL, et al. Hyperrenin-emic hypoaldosteronism in the critically ill: a new entity. J Clin Endocrinol Metab. 1981;53:867-873.

190. Nishio K, Akai Y, Murao Y, et al. Increased plasma concen-trations of adrenomedullin correlate with relaxation of vascular tone in patients with septic shock. Crit Care Med. 1997;25:953-957.

191. Annane D, Sanquer S, Sébille V, et al. Compartmentalised inducible nitric-oxide synthase activity in septic shock. Lancet. 2000;355:1143-1148.

192. Holmes CL, Patel BM, Russell JA, Walley KR. Physiology of vasopressin relevant to management of septic shock. Chest. 2001;120:989-1002.

193. Landry DW, Oliver JA. The pathogenesis of vasodilatory shock. N Engl J Med. 2001;345:588-595.

194. Fein AM, Bernard GR, Criner GJ, et al. Treatment of severe systemic inflammatory response syndrome and sepsis with a novel bradykinin antagonist, Deltibant (CP-0127). JAMA. 1997;277:482-487.

195. Carre JE, Singer M. Cellular energetic metabolism in sepsis: the need for a systems approach. Biochim Biophys Acta. 2008;1777:763-771.

196. Orrenius S, Zhivotovsky B, Nicotera P. Regulation of cell death: the calcium-apoptosis link. Nat Rev Mol Cell Biol. 2003;4:552-565.

197. Exline MC, Crouser ED. Mitochondrial mechanisms of sepsis-induced organ failure. Front Biosci. 2008;13:5030-5041.

198. Burgner D, Jamieson SE, Blackwell JM. Genetic susceptibil-ity to infectious diseases: big is beautiful, but will bigger be even better? Lancet Infect Dis. 2006;6:653-663.

199. Dale AP, Read RC. Genetic susceptibility to meningococcal infection. Expert Rev Anti Infect Ther. 2013;11:187-199.

200. Brouwer MC, de Gans J, Heckenberg SG, et al. Host genetic susceptibility to pneumococcal and meningococcal disease: a systematic review and meta-analysis. Lancet Infect Dis. 2009;9:31-44.

201. Clark MF, Baudouin SV. A systematic review of the quality of genetic association studies in human sepsis. Intensive Care Med. 2006;32:1706-1712.

202. Teuffel O, Ethier MC, Beyene J, Sung L. Association between tumor necrosis factor-alpha promoter -308 A/G polymorphism and susceptibility to sepsis and sepsis mor-tality: a systematic review and meta-analysis. Crit Care Med. 2010;38:276-282.

203. Zhu L, Li X, Miao C. Lack of association between TLR4 Asp299Gly and Thr399Ile polymorphisms and sepsis sus-ceptibility: a meta-analysis. Gene. 2012;501:213-218.

204. Tiancha H, Huiqin W, Jiyong J, et al. Association between lymphotoxin-alpha intron +252 polymorphism and sepsis: a meta-analysis. Scand J Infect Dis. 2011;43:436-447.

205. Li L, Nie W, Zhou H, et al. Association between plasmino-gen activator inhibitor-1 -675 4G/5G polymorphism and sepsis: a meta-analysis. PLoS One. 2013;8:e54883.

206. Man M, Close SL, Shaw AD, et al. Beyond single-marker analyses: mining whole genome scans for insights into treat-ment responses in severe sepsis. Pharmacogenomics J. 2012;13:218-226.

207. Bernard GR, Vincent JL, Laterre P, et al. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med. 2001;344:699-709.

208. Munford RS. Severe sepsis and septic shock: the role of gram-negative bacteremia. Annu Rev Pathol. 2006;1:467-496.

209. Feezor RJ, Oberholzer C, Baker HV, et al. Molecular char-acterization of the acute inflammatory response to infec-tions with gram-negative versus gram-positive bacteria. Infect Immun. 2003;71:5803-5813.

210. Cohen J, Abraham E. Microbiologic findings and correla-tions with serum tumor necrosis factor-a in patients with severe sepsis and septic shock. J Infect Dis. 1999;180:116-121.

211. Abe R, Oda S, Sadahiro T, et al. Gram-negative bacteremia induces greater magnitude of inflammatory response than Gram-positive bacteremia. Crit Care. 2010;14:R27.

212. Nau GJ, Richmond JF, Schlesinger A, et al. Human macro-phage activation programs induced by bacterial pathogens. Proc Natl Acad Sci U S A. 2002;99:1503-1508.

213. Wang ZM, Liu C, Dziarski R. Chemokines are the main proinflammatory mediators in human monocytes activated by Staphylococcus aureus, peptidoglycan, and endotoxin. J Biol Chem. 2000;275:20260-20267.

214. Boldrick JC, Alizadeh AA, Diehn M, et al. Stereotyped and specific gene expression programs in human innate immune responses to bacteria. Proc Natl Acad Sci U S A. 2002;99:972-977.

215. Tang BM, McLean AS, Dawes IW, et al. Gene-expression profiling of peripheral blood mononuclear cells in sepsis. Crit Care Med. 2009;37:882-888.

216. Ramilo O, Allman W, Chung W, et al. Gene expression patterns in blood leukocytes discriminate patients with acute infections. Blood. 2007;109:2066-2077.

217. Shenep JL, Flynn PM, Barrett FF, et al. Serial quantitation of endotoxemia and bacteremia during therapy for gram-negative bacterial sepsis. J Infect Dis. 1988;157:565-568.

218. Butler T, Levin J, Linh NN, et al. Yersinia pestis infection in Vietnam. II. Quantitative blood cultures and detection of endotoxin in the cerebrospinal fluid of patients with meningitis. J Infect Dis. 1976;133:493-499.

219. Walsh AL, Smith MD, Wuthiekanun V, et al. Prognostic significance of quantitative bacteremia in septicemic meli-oidosis. Clin Infect Dis. 1995;21:1498-1500.

220. Gradel KO, Dethlefsen C, Schonheyder HC, Nielsen H. Magnitude of bacteraemia is associated with increased mortality in non-typhoid salmonellosis: a one-year follow-up study. APMIS. 2008;116:147-153.

221. Murono K, Hirano Y, Koyano S, et al. Molecular compari-son of bacterial isolates from blood with strains colonizing pharynx and intestine in immunocompromised patients with sepsis. J Med Microbiol. 2003;52:527-530.

222. Sannes MR, Kuskowski MA, Owens K, et al. Virulence factor profiles and phylogenetic background of Escherichia coli isolates from veterans with bacteremia and uninfected control subjects. J Infect Dis. 2004;190:2121-2128.

223. Venier AG, Talon D, Patry I, et al. Patient and bacterial determinants involved in symptomatic urinary tract infec-tion caused by Escherichia coli with and without bacterae-mia. Clin Microbiol Infect. 2007;13:205-208.

224. Wang MC, Tseng CC, Chen CY, et al. The role of bacterial virulence and host factors in patients with Escherichia coli bacteremia who have acute cholangitis or upper urinary tract infection. Clin Infect Dis. 2002;35:1161-1166.

225. Yu VL, Hansen DS, Ko WC, et al. Virulence characteristics of Klebsiella and clinical manifestations of K. pneumoniae bloodstream infections. Emerg Infect Dis. 2007;13:986-993.

226. Wenzel RP. Nosocomial candidemia: risk factors and attributable mortality. Clin Infect Dis. 1995;20:1531-1534.

227. Blot SI, Vandewoude KH, Hoste EA, Colardyn FA. Outcome and attributable mortality in critically ill patients with bacteremia involving methicillin-susceptible and methicillin-resistant Staphylococcus aureus. Arch Intern Med. 2002;162:2229-2235.

228. Linden PK, Pasculle AW, Manez R, et al. Differences in outcomes for patients with bacteremia due to vancomycin-resistant Enterococcus faecium or vancomycin-susceptible E. faecium. Clin Infect Dis. 1996;22:663-670.

229. Sreeramoju PV, Tolentino J, Garcia-Houchins S, Weber SG. Predictive factors for the development of central line-associated bloodstream infection due to gram-negative bacteria in intensive care unit patients after surgery. Infect Control Hosp Epidemiol. 2008;29:51-56.

230. Schwenzer KJ, Gist A, Durbin CG. Can bacteremia be pre-dicted in surgical intensive care unit patients? Intensive Care Med. 1994;20:425-430.

231. Bates DW, Sands K, Miller E, et al. Predicting bacteremia in patients with sepsis syndrome. J Infect Dis. 1997;176:1538-1551.

232. Metersky ML, Ma A, Bratzler DW, Houck PM. Predicting bacteremia in patients with community-acquired pneumo-nia. Am J Respir Crit Care Med. 2004;169:342-347.

233. Kumar A, Ellis P, Arabi Y, et al. Initiation of inappropriate antimicrobial therapy results in a fivefold reduction of sur-vival in human septic shock. Chest. 2009;136:1237-1248.

234. DiGiovine B, Chenoweth C, Watts C, Higgins M. The attributable mortality and costs of primary nosocomial bloodstream infections in the intensive care unit. Am J Respir Crit Care Med. 1999;160:976-981.

235. Rello J, Ochagavia A, Sabanes E, et al. Evaluation of outcome of intravenous catheter-related infections in criti-cally ill patients. Am J Respir Crit Care Med. 2000;162:1027-1030.

236. Renaud B, Brun-Buisson C. Outcomes of primary and catheter-related bacteremia. A cohort and case-control study in critically ill patients. Am J Respir Crit Care Med. 2001;163:1584-1590.

237. Blot S, Vandewoude K, De Bacquer D, Colardyn F. Noso-comial bacteremia caused by antibiotic-resistant gram-negative bacteria in critically ill patients: clinical outcome and length of hospitalization. Clin Infect Dis. 2002;34:1600-1606.

238. Roth VR, Arduino MJ, Nobiletti J, et al. Transfusion-related sepsis due to Serratia liquefaciens in the United States. Transfusion (Paris). 2000;40:931-935.

239. Blot SI, Vandewoude KH, Colardyn FA. Evaluation of outcome in critically ill patients with nosocomial Entero-bacter bacteremia: results of a matched cohort study. Chest. 2003;123:1208-1213.

240. Bates DW, Pruess KE, Lee TH. How bad are bacteremia and sepsis? Outcomes in a cohort with suspected bacteremia. Arch Intern Med. 1995;155:593-598.

241. Felty AR, Keefer CS. Bacillus coli sepsis. JAMA. 1924;82:1430-1433.

242. Waage A, Halstensen A, Espevik T. Association between tumour necrosis factor in serum and fatal outcome in patients with meningococcal disease. Lancet. 1987;1:355-357.

243. Brandtzaeg P, Kierulf P, Gaustad P, et al. Plasma endotoxin as a predictor of multiple organ failure and death in sys-temic meningococcal disease. J Infect Dis. 1989;159:195-204.

244. Ovstebo R, Brandtzaeg P, Brusletto B, et al. Use of robotized DNA isolation and real-time PCR to quantify and identify close correlation between levels of Neisseria meningitidis DNA and lipopolysaccharides in plasma and cerebrospinal fluid from patients with systemic meningococcal disease. J Clin Microbiol. 2004;42:2980-2987.

245. Suffredini AF, Fromm RE, Parker MM, et al. The cardio-vascular response of normal humans to the administration of endotoxin. N Engl J Med. 1989;321:280-287.

246. Bahador M, Cross AS. From therapy to experimental model: a hundred years of endotoxin administration to human subjects. J Endotoxin Res. 2007;13:251-279.

247. Romaschin AD, Harris DM, Ribeiro MB, et al. A rapid assay of endotoxin in whole blood using autologous neutrophil dependent chemiluminescence. J Immunol Methods. 1998;212:169-185.

248. Lu M, Munford RS. The transport and inactivation kinetics of bacterial lipopolysaccharide influence its immunological potency in vivo. J Immunol. 2011;187:3314-3320.

249. Kitchens RL, Thompson PA, Viriyakosol S, et al. Plasma CD14 decreases monocyte responses to LPS by promoting the transfer of cell-bound LPS to plasma lipoproteins. J Clin Invest. 2001;108:485-493.

250. Levels JH, Lemaire LC, van den Ende AE, et al. Lipid com-position and lipopolysaccharide binding capacity of lipo-proteins in plasma and lymph of patients with systemic

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

934.e4

inflammatory response syndrome and multiple organ failure. Crit Care Med. 2003;31:1647-1653.

251. Kitchens RL, Thompson PA, Munford RS, O’Keefe GE. Acute inflammation and infection maintain circulating phospholipid levels and enhance lipopolysaccharide binding to plasma lipoproteins. J Lipid Res. 2003;44:2339-2348.

252. Munford RS. Detoxifying endotoxin: time, place, and person. J Endotoxin Res. 2005;11:69-84.

253. Froon AH, Dentener MA, Greve JW, et al. Lipopolysac-charide toxicity-regulating proteins in bacteremia. J Infect Dis. 1995;171:1250-1257.

254. Zweigner J, Gramm HJ, Singer OC, et al. High concentra-tions of lipopolysaccharide-binding protein in serum of patients with severe sepsis or septic shock inhibit the lipopolysaccharide response in human monocytes. Blood. 2001;98:3800-3808.

255. Thompson PA, Tobias PS, Viriyakosol S, et al. Lipopolysac-charide (LPS)-binding protein inhibits responses to cell-bound LPS. J Biol Chem. 2003;278:28367-28371.

256. van Langevelde P, Joop K, Van Loon J, et al. Endotoxin, cytokines, and procalcitonin in febrile patients admitted to the hospital: identification of subjects at high risk of mor-tality. Clin Infect Dis. 2000;31:1343-1348.

257. Hurley JC. Concordance of endotoxemia with gram-negative bacteremia: a meta-analysis using receiver operat-ing characteristic curves. Arch Pathol Lab Med. 2000;124:1157-1164.

258. Hurley JC. Reappraisal with meta-analysis of bacteremia, endotoxemia, and mortality in gram-negative sepsis. J Clin Microbiol. 1995;33:1278-1282.

259. Munford RS, Varley AW. Shield as signal: lipopolysaccha-rides and the evolution of immunity to gram-negative bacteria. PLoS Pathog. 2006;2:e67.

260. Unnikrishnan M, Altmann DM, Proft T, et al. The bacterial superantigen streptococcal mitogenic exotoxin Z is the major immunoactive agent of Streptococcus pyogenes. J Immunol. 2002;169:2561-2569.

261. Kotb M. Bacterial pyrogenic exotoxins as superantigens. Clin Microbiol Rev. 1995;8:411-426.

262. Sriskandan S, Altmann DM. The immunology of sepsis. J Pathol. 2008;214:211-223.

263. Xu SX, McCormick JK. Staphylococcal superantigens in colonization and disease. Front Cell Infect Microbiol. 2012;2:52.

264. Hellman J, Roberts JD Jr, Tehan MM, et al. Bacterial peptidoglycan-associated lipoprotein is released into the bloodstream in gram-negative sepsis and causes inflamma-tion and death in mice. J Biol Chem. 2002;277:14274-14280.

265. Munford RS. Sensing gram-negative bacterial lipopolysac-charides: a human disease determinant? Infect Immun. 2008;76:454-465.

266. Leroy EM, Baize S, Volchkov VE, et al. Human asymptom-atic Ebola infection and strong inflammatory response. Lancet. 2000;355:2210-2215.

267. Alcaïs A, Abel L, Casanova JL. Human genetics of infec-tious diseases: between proof of principle and paradigm. J Clin Invest. 2009;119:2506-2514.

268. Serhan CN, Brain SD, Buckley CD, et al. Resolution of inflammation: state of the art, definitions and terms. FASEB J. 2007;21:325-332.

269. Nathan C, Ding A. Nonresolving inflammation. Cell. 2010;140:871-882.

270. Chen X, El Gazzar M, Yoza BK, McCall CE. The NF- kappaB factor RelB and histone H3 lysine methyl-transferase G9a directly interact to generate epigenetic silencing in endotoxin tolerance. J Biol Chem. 2009;284:27857-27865.

271. Quartin AA, Schein RM, Kett DH, Peduzzi PN. Magnitude and duration of the effect of sepsis on survival. JAMA. 1997;277:1058-1063.

272. Sharshar T, Hopkinson NS, Orlikowski D, Annane D. Science review: the brain in sepsis—culprit and victim. Crit Care. 2005;9:37-44.

273. Gofton TE, Young GB. Sepsis-associated encephalopathy. Nat Rev Neurol. 2012;8:557-566.

274. Sharshar T, Carlier R, Bernard F, et al. Brain lesions in septic shock: a magnetic resonance imaging study. Intensive Care Med. 2007;33:798-806.

275. Gunther ML, Morandi A, Krauskopf E, et al. The associa-tion between brain volumes, delirium duration, and cogni-tive outcomes in intensive care unit survivors: the VISIONS cohort magnetic resonance imaging study. Crit Care Med. 2012;40:2022-2032.

276. Iwashyna TJ, Ely EW, Smith DM, Langa KM. Long-term cognitive impairment and functional disability among sur-vivors of severe sepsis. JAMA. 2010;304:1787-1794.

277. Sharshar T, Blanchard A, Paillard M, et al. Circulating vasopressin levels in septic shock. Crit Care Med. 2003;31:1752-1758.

278. Sharshar T, Carlier R, Blanchard A, et al. Depletion of neurohypophyseal content of vasopressin in septic shock. Crit Care Med. 2002;30:497-500.

279. Lamberts SW, Bruining HA, de Jong FH. Corticosteroid therapy in severe illness. N Engl J Med. 1997;337:1285-1292.

280. Shenker Y, Skatrud JB. Adrenal insufficiency in critically ill patients. Am J Respir Crit Care Med. 2001;163:1520-1523.

281. Briegel J, Schelling G, Haller M, et al. A comparison of the adrenocortical response during septic shock and after com-plete recovery. Intensive Care Med. 1996;22:894-899.

282. Bollaert PE, Fieux F, Charpentier C, Levy B. Baseline cor-tisol levels, cortisol response to corticotropin, and progno-sis in late septic shock. Shock. 2003;19:13-15.

283. Rothwell PM, Udwadia ZF, Lawler PG. Cortisol response to corticotropin and survival in septic shock. Lancet. 1991;337:582-583.

284. Manglik S, Flores E, Lubarsky L, et al. Glucocorticoid insuf-ficiency in patients who present to the hospital with severe sepsis: a prospective clinical trial. Crit Care Med. 2003;31:1668-1675.

285. Marik PE, Pastores SM, Annane D, et al. Recommenda-tions for the diagnosis and management of corticosteroid insufficiency in critically ill adult patients: consensus state-ments from an international task force by the American College of Critical Care Medicine. Crit Care Med. 2008;36:1937-1949.

286. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurement, physio-logical interpretation and clinical use. Circulation. 1996;93:1043-1065.

287. Lake DE, Richman JS, Griffin MP, Moorman JR. Sample entropy analysis of neonatal heart rate variability. Am J Physiol Regul Integr Comp Physiol. 2002;283:R789-R797.

288. Griffin MP, O’Shea TM, Bissonette EA, et al. Abnormal heart rate characteristics preceding neonatal sepsis and sepsis-like illness. Pediatr Res. 2003;53:920-926.

289. Annane D, Trabold F, Sharshar T, et al. Inappropriate sym-pathetic activation at onset of septic shock. Am J Respir Crit Care Med. 1999;160:458-465.

290. Chen WL, Chen JH, Huang CC, et al. Heart rate variability measures as predictors of in-hospital mortality in ED patients with sepsis. Am J Emerg Med. 2008;26:395-401.

291. Seely AJE, Christou NV. Multiple organ dysfunction syn-drome: exploring the paradigm of complex nonlinear systems. Crit Care Med. 2000;28:2193-2200.

292. Godin PJ, Buchman TG. Uncoupling of biological oscilla-tors: a complementary hypothesis concerning the patho-genesis of multiple organ dysfunction syndrome. Crit Care Med. 1996;24:1107-1116.

293. Bolton CF, Young GB, Zochodne DW. The neurological complications of sepsis. Ann Neurol. 1993;33:94-100.

294. Lacomis D. Electrophysiology of neuromuscular disorders in critical illness. Muscle Nerve. 2013;47:452-463.

295. Diaz NL, Finol HJ, Torres SH, et al. Histochemical and ultrastructural study of skeletal muscle in patients with sepsis and multiple organ failure syndrome (MOFS). Histol Histopathol. 1998;13:121-128.

296. De Jonghe B, Lacherade JC, Durand MC, Sharshar T. Criti-cal illness neuromuscular syndromes. Neurol Clin. 2008;26:507-520, ix.

297. Venet F, Davin F, Guignant C, et al. Early assessment of leukocyte alterations at diagnosis of septic shock. Shock. 2010;34:358-363.

298. Hotchkiss RS, Monneret G, Payen D. Immunosuppression in sepsis: a novel understanding of the disorder and a new therapeutic approach. Lancet Infect Dis. 2013;13:260-268.

299. Forel JM, Chiche L, Thomas G, et al. Phenotype and func-tions of natural killer cells in critically-ill septic patients. PLoS One. 2012;7:e50446.

300. Hotchkiss RS, Tinsley KW, Swanson PE, et al. Sepsis-induced apoptosis causes progressive profound depletion of B and CD4+ T lymphocytes in humans. J Immunol. 2001;166:6952-6963.

301. Holub M, Kluckova Z, Helcl M, et al. Lymphocyte subset numbers depend on the bacterial origin of sepsis. Clin Microbiol Infect. 2003;9:202-211.

302. Reddy RC, Chen GH, Tekchandani PK, Standiford TJ. Sepsis-induced immunosuppression: from bad to worse. Immunol Res. 2001;24:273-287.

303. Jiang LN, Yao YM, Sheng ZY. The role of regulatory T cells in the pathogenesis of sepsis and its clinical implication. J Interferon Cytokine Res. 2012;32:341-349.

304. Brunialti MK, Santos MC, Rigato O, et al. Increased per-centages of T helper cells producing IL-17 and monocytes expressing markers of alternative activation in patients with sepsis. PLoS One. 2012;7:e37393.

305. Venet F, Lepape A, Monneret G. Clinical review: flow cytometry perspectives in the ICU: from diagnosis of infec-tion to monitoring of injury-induced immune dysfunc-tions. Crit Care. 2011;15:231.

306. Aird WC. The hematologic system as a marker of organ dysfunction in sepsis. Mayo Clin Proc. 2003;78:869-881.

307. Stéphan F, Thiolière B, Verdy E, Tulliez N. Role of hemophagocytic histiocytosis in the etiology of

thrombocytopenia in patients with sepsis syndrome or septic shock. Clin Infect Dis. 1997;25:1159-1164.

308. Khovidhunkit W, Memon RA, Feingold KR, Grunfeld C. Infection and inflammation-induced proatherogenic changes of lipoproteins. J Infect Dis. 2000;181:S462-S472.

309. Harris RL, Musher DM, Bloom K, et al. Manifestations of sepsis. Arch Intern Med. 1987;147:1895-1906.

310. NICE-SUGAR Study Investigators, Finfer S, Liu B, et al. Hypoglycemia and risk of death in critically ill patients. N Engl J Med. 2012;367:1108-1118.

311. James JH, Luchette FA, McCarter FD, Fischer JE. Lactate is an unreliable indicator of tissue hypoxia in injury or sepsis. Lancet. 1999;354:505-508.

312. Levy B. Lactate and shock state: the metabolic view. Curr Opin Crit Care. 2006;12:315-321.

313. Levy B, Gibot S, Franck P, et al. Relation between muscle Na+K+ ATPase activity and raised lactate concentrations in septic shock: a prospective study. Lancet. 2005;365:871-875.

314. Jansen TC, van Bommel J, Bakker J. Blood lactate monitor-ing in critically ill patients: a systematic health technology assessment. Crit Care Med. 2009;37:2827-2839.

315. Bernard GR, Wheeler AP, Russell JA, et al. The effects of ibuprofen on the physiology and survival of patients with sepsis. N Engl J Med. 1997;336:912-918.

316. Wada H, Thachil J, Di Nisio M, et al. Guidance for diagno-sis and treatment of DIC from harmonization of the rec-ommendations from three guidelines. J Thromb Haemost. 2013;11:761-767.

317. Bick RL. Disseminated intravascular coagulation: patho-physiological mechanisms and manifestations. Semin Thromb Hemost. 1998;24:3-18.

318. Habe K, Wada H, Ito-Habe N, et al. Plasma ADAMTS13, von Willebrand factor (VWF) and VWF propeptide pro-files in patients with DIC and related diseases. Thromb Res. 2012;129:598-602.

319. Mant MJ, King EG. Severe, acute disseminated intravascu-lar coagulation. A reappraisal of its pathophysiology, clini-cal significance and therapy based on 47 patients. Am J Med. 1979;67:557-563.

320. Chandler WL. Treatment of thrombosis associated with septic shock. J Lab Clin Med. 1991;118:513-514.

321. Matthay MA, Ware LB, Zimmerman GA. The acute respiratory distress syndrome. J Clin Invest. 2012;122:2731-2740.

322. Force ADT, Ranieri VM, Rubenfeld GD, et al. Acute respi-ratory distress syndrome: the Berlin Definition. JAMA. 2012;307:2526-2533.

323. Agrawal A, Matthay MA, Kangelaris KN, et al. Plasma angiopoietin-2 predicts the onset of acute lung injury in critically ill patients. Am J Respir Crit Care Med. 2013;187:736-742.

324. Bellomo R. Acute renal failure. Semin Respir Crit Care Med. 2011;32:639-650.

325. Mehta RL, Bouchard J, Soroko SB, et al. Sepsis as a cause and consequence of acute kidney injury: program to improve care in acute renal disease. Intensive Care Med. 2011;37:241-248.

326. Puleo F, Arvanitakis M, Van Gossum A, Preiser JC. Gut failure in the ICU. Semin Respir Crit Care Med. 2011;32:626-638.

327. Doig CJ, Sutherland LR, Sandham JD, et al. Increased intes-tinal permeability is associated with the development of multiple organ dysfunction syndrome in critically ill ICU patients. Am J Respir Crit Care Med. 1998;158:444-451.

328. Franson TR, Hierholzer WJ Jr, LaBrecque DR. Frequency and characteristics of hyperbilirubinemia associated with bacteremia. Rev Infect Dis. 1985;7:1-9.

329. Henrion J. Hypoxic hepatitis. Liver Int. 2012;32:1039-1052.330. Cook CH, Martin LC, Yenchar JK, et al. Occult herpes

family viral infections are endemic in critically ill surgical patients. Crit Care Med. 2003;31:1923-1929.

331. Heininger A, Jahn G, Engel C, et al. Human cytomegalovi-rus infections in nonimmunosuppressed critically ill patients. Crit Care Med. 2001;29:541-547.

332. Limaye AP, Kirby KA, Rubenfeld GD, et al. Cytomegalovi-rus reactivation in critically ill immunocompetent patients. JAMA. 2008;300:413-422.

333. Luyt CE, Combes A, Deback C, et al. Herpes simplex virus lung infection in patients undergoing prolonged mechani-cal ventilation. Am J Respir Crit Care Med. 2007;175:935-942.

334. Kutza AS, Muhl E, Hackstein H, et al. High incidence of active cytomegalovirus infection among septic patients. Clin Infect Dis. 1998;26:1076-1082.

335. Jaber S, Chanques G, Borry J, et al. Cytomegalovirus infec-tion in critically ill patients: associated factors and conse-quences. Chest. 2005;127:233-241.

336. Landelle C, Lepape A, Français A, et al. Nosocomial infec-tion after septic shock among intensive care unit patients. Infect Control Hosp Epidemiol. 2008;29:1054-1065.

337. Torgersen C, Moser P, Luckner G, et al. Macroscopic post-mortem findings in 235 surgical intensive care patients with sepsis. Anesth Analg. 2009;108:1841-1847.

Ch

apter 75 Sepsis,SevereSepsis,andSepticShock

934.e5

338. Musher D. Cutaneous and soft-tissue manifestations of sepsis due to gram-negative enteric bacilli. Rev Infect Dis. 1980;2:854-866.

339. Betrosian AP, Berlet T, Agarwal B. Purpura fulminans in sepsis. Am J Med Sci. 2006;332:339-345.

340. Salvo I, de Cian W, Musicco M, et al. The Italian SEPSIS study: preliminary results on the incidence and evolution of SIRS, sepsis, severe sepsis, and septic shock. Intensive Care Med. 1995;21:S244-S249.

341. Kumar A, Haery C, Parrillo JE. Myocardial dysfunction in septic shock: Part I. Clinical manifestation of cardiovas-cular dysfunction. J Cardiothorac Vasc Anesth. 2001;15:364-376.

342. Flierl MA, Rittirsch D, Huber-Lang MS, et al. Molecular events in the cardiomyopathy of sepsis. Mol Med. 2008;14:327-336.

343. Pulido JN, Afessa B, Masaki M, et al. Clinical spectrum, frequency, and significance of myocardial dysfunction in severe sepsis and septic shock. Mayo Clin Proc. 2012;87:620-628.

344. Abraham E, Shoemaker WD, Bland RD, Cobo JC. Sequen-tial cardiorespiratory patterns in septic shock. Crit Care Med. 1983;11:799-803.

345. Shoemaker WC, Appel PL, Kram HB, et al. Temporal hemodynamic and oxygen transport patterns in medical patients. Septic shock. Chest. 1993;104:1529-1536.

346. Johnson SB, Lissauer M, Bochicchio GV, et al. Gene expres-sion profiles differentiate between sterile SIRS and early sepsis. Ann Surg. 2007;245:611-621.

347. Lissauer ME, Johnson SB, Bochicchio GV, et al. Differential expression of toll-like receptor genes: sepsis compared with sterile inflammation 1 day before sepsis diagnosis. Shock. 2009;31:238-244.

348. Lissauer ME, Johnson SB, Siuzdak G, et al. Coagulation and complement protein differences between septic and uninfected systemic inflammatory response syndrome patients. J Trauma. 2007;62:1082-1092; discussion 1092-1094.

349. Buchan CA, Bravi A, Seely AJ. Variability analysis and the diagnosis, management, and treatment of sepsis. Curr Infect Dis Rep. 2012;14:512-521.

350. O’Grady NP, Barie PS, Bartlett JG, et al. Guidelines for evaluation of new fever in critically ill adult patients: 2008 update from the American College of Critical Care Medi-cine and the Infectious Diseases Society of America. Crit Care Med. 2008;36:1330-1349.

351. Lamy B, Roy P, Carret G, et al. What is the relevance of obtaining multiple blood samples for culture? A compre-hensive model to optimize the strategy for diagnosing bacteremia. Clin Infect Dis. 2002;35:842-850.

352. Riedel S, Carroll KC. Blood cultures: key elements for best practices and future directions. J Infect Chemother. 2010;16:301-316.

353. Mancini N, Carletti S, Ghidoli N, et al. The era of molecular and other non-culture-based methods in diagnosis of sepsis. Clin Microbiol Rev. 2010;23:235-251.

354. Gaur AH, Flynn PM, Giannini MA, et al. Difference in time to detection: a simple method to differentiate catheter-related from non-catheter-related bloodstream infection in immunocompromised pediatric patients. Clin Infect Dis. 2003;37:469-475.

355. Malani A, Trimble K, Parekh V, et al. Review of clinical trials of skin antiseptic agents used to reduce blood culture contamination. Infect Control Hosp Epidemiol. 2007;28:892-895.

356. Suwanpimolkul G, Pongkumpai M, Suankratay C. A ran-domized trial of 2% chlorhexidine tincture compared with 10% aqueous povidone-iodine for venipuncture site dis-infection: effects on blood culture contamination rates. J Infect. 2008;56:354-359.

357. Van Deuren M, van Dijke BJ, Koopman RJ, et al. Rapid diagnosis of acute meningococcal infections by needle aspiration or biopsy of skin lesions. BMJ. 1993;306:1229-1232.

358. Pierrakos C, Vincent JL. Sepsis biomarkers: a review. Crit Care. 2010;14:R15.

359. Arafah BM. Hypothalamic pituitary adrenal function during critical illness: limitations of current assessment methods. J Clin Endocrinol Metab. 2006;91:3725-3745.

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

361. Minneci PC, Deans KJ, Eichacker PQ, Natanson C. The effects of steroids during sepsis depend on dose and sever-ity of illness: an updated meta-analysis. Clin Microbiol Infect. 2009;15:308-318.

362. Barnett JA, Sanford JP. Bacterial shock. JAMA. 1969;209:1514-1517.

363. Young LS. Sepsis syndrome. In: Mandell GL, Bennett JE, Dolin R, eds. Principles and Practice of Infectious Diseases. 5th ed. Philadelphia: Churchill Livingstone; 2000:806-819.

364. Kollef MH, Sherman G, Ward S, Fraser VJ. Inadequate antimicrobial treatment of infections: a risk factor for hospital mortality among critically ill patients. Chest. 1999;115:462-474.

365. Kreger BE, Craven DE, McCabe WR. Gram-negative bac-teremia. IV. Re-evaluation of clinical features and treat-ment in 612 patients. Am J Med. 1980;68:344-355.

366. Leibovici L, Paul M, Poznanski O, et al. Monotherapy versus beta-lactam-aminoglycoside combination treat-ment for gram-negative bacteremia: a prospective, obser-vational study. Antimicrob Agents Chemother. 1997;41:1127-1133.

367. Leibovici L, Shraga I, Drucker M, et al. The benefit of appropriate empirical antibiotic treatment in patients with bloodstream infection. J Intern Med. 1998;244:379-386.

368. Ibrahim EH, Sherman G, Ward S, et al. The influence of inadequate antimicrobial treatment of bloodstream infec-tions on patient outcomes in the ICU setting. Chest. 2000;118:146-155.

369. Labelle A, Juang P, Reichley R, et al. The determinants of hospital mortality among patients with septic shock receiv-ing appropriate initial antibiotic treatment. Crit Care Med. 2012;40:2016-2021.

370. Greisman SE, DuBuy JB, Woodward CL. Experimental gram-negative bacterial sepsis: prevention of mortality not preventable by antibiotics alone. Infect Immun. 1979;25:538-557.

371. Kumar A, Roberts D, Wood KE, et al. Duration of hypoten-sion before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med. 2006;34:1589-1596.

372. Houck PM, Bratzler DW, Nsa W, et al. Timing of antibiotic administration and outcomes for Medicare patients hospi-talized with community-acquired pneumonia. Arch Intern Med. 2004;164:637-644.

373. Kollef M, Micek S, Hampton N, et al. Septic shock attrib-uted to Candida infection: importance of empiric therapy and source control. Clin Infect Dis. 2012;54:1739-1746.

374. Harbarth S, Nobre V, Pittet D. Does antibiotic selection impact patient outcome? Clin Infect Dis. 2007;44:87-93.

375. Kumar A, Safdar N, Kethireddy S, Chateau D. A survival benefit of combination antibiotic therapy for serious infec-tions associated with sepsis and septic shock is contingent only on the risk of death: a meta-analytic/meta-regression study. Crit Care Med. 2010;38:1651-1664.

376. Eichacker PQ, Parent C, Kalil A, et al. Risk and the efficacy of antiinflammatory agents: retrospective and confirma-tory studies of sepsis. Am J Respir Crit Care Med. 2002;166:1197-1205.

377. Buijk SE, Mouton JW, Gyssens IC, et al. Experience with a once-daily dosing program of aminoglycosides in critically ill patients. Intensive Care Med. 2002;28:936-942.

378. Drusano GL, Lodise TP. Editorial commentary: saving lives with optimal antimicrobial chemotherapy. Clin Infect Dis. 2013;56:245-247.

379. Dulhunty JM, Roberts JA, Davis JS, et al. Continuous infu-sion of beta-lactam antibiotics in severe sepsis: a multi-center double-blind, randomized controlled trial. Clin Infect Dis. 2013;56:236-244.

380. Weinstein MP, Murphy JR, Reller LB, Lichtenstein KA. The clinical significance of positive blood cultures: a compre-hensive analysis of 500 episodes of bacteremia and funge-mia in adults. II. Clinical observations, with special reference to factors influencing prognosis. Rev Infect Dis. 1983;5:54-70.

381. Marshall JC, Maier RV, Jimenez M, Dellinger EP. Source control in the management of severe sepsis and septic shock: an evidence-based review. Crit Care Med. 2004;32:S513-S526.

382. Natanson C, Danner RL, Reilly JM, et al. Antibiotics versus cardiovascular support in a canine model of human septic shock. Am J Physiol. 1990;259:H1440-H1447.

383. Perel P, Roberts I, Ker K. Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Data-base Syst Rev. 2013;2:CD000567.

384. SAFE Study Investigators. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med. 2004;350:2247-2256.

385. Delaney AP, Dan A, McCaffrey J, Finfer S. The role of albumin as a resuscitation fluid for patients with sepsis: a systematic review and meta-analysis. Crit Care Med. 2011;39:386-391.

385a. Caironi P, Tognoni G, Masson S, et al. Albumin replace-ment in patients with severe sepsis or septic shock. N Engl J Med. 2014 Mar 18. [Epub ahead of print]

386. Zarychanski R, Abou-Setta AM, Turgeon AF, et al. Associa-tion of hydroxyethyl starch administration with mortality and acute kidney injury in critically ill patients requiring volume resuscitation: a systematic review and meta-analysis. JAMA. 2013;309:678-688.

387. Perner A, Haase N, Guttormsen AB, et al. Hydroxyethyl starch 130/0.42 versus Ringer’s acetate in severe sepsis. N Engl J Med. 2012;367:124-134.

388. Marik PE, Monnet X, Teboul JL. Hemodynamic parame-ters to guide fluid therapy. Ann Intensive Care. 2011;1:1.

389. Bellomo R, Reade MC, Warrillow SJ. The pursuit of a high central venous oxygen saturation in sepsis: growing con-cerns. Crit Care. 2008;12:130.

390. Perel A. Bench-to-bedside review: the initial hemody-namic resuscitation of the septic patient according to Sur-viving Sepsis Campaign guidelines—does one size fit all? Crit Care. 2008;12:223.

391. Suffredini AF, Munford RS. Novel therapies for septic shock over the past 4 decades. JAMA. 2011;306:194-199.

391a. The ProCESS Investigators. A randomized trial of proto-col-based care for early septic shock. N Engl J Med. 2014 Mar 18. [Epub ahead of print]

392. Bellomo R, Chapman M, Finfer S, et al. Low-dose dopa-mine 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.

393. Marik PE, Iglesias J, Investigators NIS. Low-dose dopamine does not prevent acute renal failure in patients with septic shock and oliguria. Am J Med. 1999;107:387-390.

394. De Backer D, Biston P, Devriendt J, et al. Comparison of dopamine and norepinephrine in the treatment of shock. N Engl J Med. 2010;362:779-789.

395. Annane D, Vignon P, Renault A, et al. Norepinephrine plus dobutamine versus epinephrine alone for manage-ment of septic shock: a randomised trial. Lancet. 2007;370:676-684.

396. Annane D, Sébille V, Charpentier C, et al. Effect of treat-ment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA. 2002;288:862-871.

397. The COIITSS Study Investigators. Corticosteroid treat-ment and intensive insulin therapy for septic shock in adults: a randomized controlled trial. JAMA. 2010;303:341-348.

398. Vincent JL, Sun QH, Dubois MJ. Clinical trials of immu-nomodulatory therapies in severe sepsis and septic shock. Clin Infect Dis. 2002;34:1084-1093.

399. Marshall JC. Such stuff as dreams are made on: mediator-directed therapy in sepsis. Nat Rev Drug Discov. 2003;2:391-405.

400. Carlet J, Cohen J, Calandra T, et al. Sepsis: time to recon-sider the concept. Crit Care Med. 2008;36:964-966.

401. Freeman BD, Natanson C. Anti-inflammatory therapies in sepsis and septic shock. Expert Opin Investig Drugs. 2000;9:1651-1663.

402. Albertson TE, Panacek EA, MacArthur RD, et al. Multi-center evaluation of a human monoclonal antibody to Enterobacteriaceae common antigen in patients with gram-negative sepsis. Crit Care Med. 2003;31:419-427.

403. Levin M, Quint PA, Goldstein B, et al. Recombinant bactericidal/permeability-increasing protein (rBPI21) as adjunctive treatment for children with severe meningococ-cal sepsis: a randomised trial. Lancet. 2000;356:961-967.

404. Opal SM, Laterre PF, Francois B, et al. Effect of eritoran, an antagonist of MD2-TLR4, on mortality in patients with severe sepsis: the ACCESS randomized trial. JAMA. 2013;309:1154-1162.

405. Dellinger RP, Tomayko JF, Angus DC, et al. Efficacy and safety of a phospholipid emulsion (GR270773) in gram-negative severe sepsis: results of a phase II multicenter, randomized, placebo-controlled, dose-finding clinical trial. Crit Care Med. 2009;37:2929-2938.

406. Warren BL, Eid A, Singer P, et al. High-dose randomized antithrombin III in severe sepsis: a randomized controlled trial. JAMA. 2001;286:1869-1878.

407. Warren HS, Suffredini AF, Eichacker PQ, Munford RS. Risks and benefits of activated protein C treatment for severe sepsis. N Engl J Med. 2002;347:1027-1030.

408. Nadel S, Goldstein B, Williams MD, et al. Drotrecogin alfa (activated) in children with severe sepsis: a multicentre phase III randomised controlled trial. Lancet. 2007;369:836-843.

409. Abraham E, Laterre PF, Garg R, et al. Drotrecogin alfa (activated) for adults with severe sepsis and a low risk of death. N Engl J Med. 2005;353:1332-1341.

410. Ranieri VM, Thompson BT, Barie PS, et al. Drotrecogin alfa (activated) in adults with septic shock. N Engl J Med. 2012;366:2055-2064.

411. Annane D, Timsit JF, Megarbane B, et al. Recombinant human activated protein C for adults with septic shock: a randomized controlled trial. Am J Respir Crit Care Med. 2013;187:1091-1097.

412. Kerschen EJ, Fernandez JA, Cooley BC, et al. Endotoxemia and sepsis mortality reduction by non-anticoagulant acti-vated protein C. J Exp Med. 2007;204:2439-2448.

413. Zarychanski R, Doucette S, Fergusson D, et al. Early intra-venous unfractionated heparin and mortality in septic shock. Crit Care Med. 2008;36:2973-2979.

414. Wasserman D, Ioannovich JD, Hinzmann RD, et al; Group fTSBS. Interferon-gamma in the prevention of severe burn-related infections: a European phase III multicenter trial. The Severe Burns Study Group. Crit Care Med. 1998;26:434-439.

415. Dries DJ. Interferon-gamma therapy for infectious compli-cations of injury: a called third strike? Crit Care Med. 1998;26:419-420.

Part

 II 

Maj

or 

Clin

ical

 Syn

dro

mes

934.e6

416. Docke WD, Randow F, Syrbe U, et al. Monocyte deactiva-tion in septic patients: restoration by IFN-gamma treat-ment. Nat Med. 1997;3:678-681.

417. Root RK, Lodato RF, Patrick W, et al. Multicenter, double-blind, placebo-controlled study of the use of filgrastim in patients hospitalized with pneumonia and severe sepsis. Crit Care Med. 2003;31:367-373.

418. Nelson S, Heyder AM, Stone J, et al. A randomized con-trolled trial of filgrastim for the treatment of hospitalized patients with multilobar pneumonia. J Infect Dis. 2000;182:970-973.

419. Presneill JJ, Harris T, Stewart AG, et al. A randomized phase II trial of granulocyte-macrophage colony-stimulating factor therapy in severe sepsis with respiratory dysfunction. Am J Respir Crit Care Med. 2002;166:138-143.

420. Bo L, Wang F, Zhu J, et al. Granulocyte-colony stimulating factor (G-CSF) and granulocyte-macrophage colony stim-ulating factor (GM-CSF) for sepsis: a meta-analysis. Crit Care. 2011;15:R58.

421. Zanetti G, Glauser MP, Baumgartner JD. Use of immuno-globulins in prevention and treatment of infection in criti-cally ill patients: review and critique. Rev Infect Dis. 1991;13:985-992.

422. Campbell WN, Hendrix E, Cryz S Jr, Cross AS. Immu-nogenicity of a 24-valent Klebsiella capsular polysaccha-ride vaccine and an eight-valent Pseudomonas O-polysaccharide conjugate vaccine administered to victims of acute trauma. Clin Infect Dis. 1996;23:179-181.

423. Turgeon AF, Hutton B, Fergusson DA, et al. Meta-analysis: intravenous immunoglobulin in critically ill adult patients with sepsis. Ann Intern Med. 2007;146:193-203.

424. Schrage B, Duan G, Yang LP, et al. Different preparations of intravenous immunoglobulin vary in their efficacy to neutralize streptococcal superantigens: implications for treatment of streptococcal toxic shock syndrome. Clin Infect Dis. 2006;43:743-746.

425. Kruger P, Bailey M, Bellomo R, et al. A multicenter ran-domized trial of atorvastatin therapy in intensive care patients with severe sepsis. Am J Respir Crit Care Med. 2013;187:743-750.

426. Patel JM, Snaith C, Thickett DR, et al. Randomized double-blind placebo-controlled trial of 40 mg/day of atorvastatin in reducing the severity of sepsis in ward patients (ASEPSIS Trial). Crit Care. 2012;16:R231.

427. Cross AS, Opal SM. A new paradigm for the treatment of sepsis: is it time to consider combination therapy? Ann Intern Med. 2003;138:502-505.

428. Cook D, Guyatt G, Marshall J, et al. A comparison of sucralfate and ranitidine for the prevention of upper

gastrointestinal bleeding in patients requiring mechanical ventilation. Canadian Critical Care Trials Group. N Engl J Med. 1998;338:791-797.

429. Kress JP, Pohlman AS, O’Connor MF, Hall JB. Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med. 2000;342:1471-1477.

430. Schortgen F, Clabault K, Katsahian S, et al. Fever control using external cooling in septic shock. Am J Respir Crit Care Med. 2012;185:1088-1095.

431. Doebbeling BN, Stanley GL, Sheetz CT, et al. Comparative efficacy of alternative hand-washing agents in reducing nosocomial infections in intensive care units. N Engl J Med. 1992;327:88-93.

432. O’Grady NP, Alexander M, Burns LA, et al. Summary of recommendations: guidelines for the prevention of intra-vascular catheter-related infections. Clin Infect Dis. 2011;52:1087-1099.

433. Drakulovic MB, Torres A, Bauer TT, et al. Supine body position as a risk factor for nosocomial pneumonia in mechanically ventilated patients: a randomised trial. Lancet. 1999;354:1851-1858.

434. Rello J, Sonora R, Jubert P, et al. Pneumonia in intubated patients: role of respiratory airway care. Am J Respir Crit Care Med. 1996;154:111-115.

435. McCabe WR, Jackson GG. Gram-negative bacteremia I. Etiology and ecology. Arch Intern Med. 1962;110:83-91.

436. Perl TM, Dvorak L, Hwang T, Wenzel RP. Long-term sur-vival and function after suspected gram-negative sepsis. JAMA. 1995;274:338-345.

437. Brun-Buisson C, Doyon F, Carlet J, et al. Bacteremia and severe sepsis in adults: a multicenter prospective survey in ICUs and wards of 24 hospitals. Am J Respir Crit Care Med. 1996;154:617-624.

438. Gogos CA, Drosou E, Bassaris HP, Skoutelis A. Pro- versus anti-inflammatory cytokine profile in patients with severe sepsis: a marker for prognosis and future therapeutic options. J Infect Dis. 2000;181:176-180.

439. Yende S, D’Angelo G, Kellum JA, et al. Inflammatory markers at hospital discharge predict subsequent mortality after pneumonia and sepsis. Am J Respir Crit Care Med. 2008;177:1242-1247.

440. Schmidt H, Hoyer D, Wilhelm J, et al. The alteration of autonomic function in multiple organ dysfunction syn-drome. Crit Care Clin. 2008;24:149-163.

441. Chen WL, Kuo CD. Characteristics of heart rate variability can predict impending septic shock in emergency depart-ment patients with sepsis. Acad Emerg Med. 2007;14:392-397.

442. Friedland JS, Porter JC, Daryanani S, et al. Plasma proin-flammatory cytokine concentrations, Acute Physiology and Chronic Health Evaluation (APACHE) III scores and survival in patients in an intensive care unit. Crit Care Med. 1996;24:1775-1781.

443. Vincent JL, de Mendonca A, Cantraine F, et al. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, pro-spective study. Working group on “sepsis-related prob-lems” of the European Society of Intensive Care Medicine. Crit Care Med. 1998;26:1793-1800.

444. Annane D, Bellissant E, Cavaillon JM. Septic shock. Lancet. 2005;365:63-78.

445. Rothwell PM, Lawler PG. Prediction of outcome in inten-sive care patients using endocrine parameters. Crit Care Med. 1995;23:78-83.

446. Marshall JC, Cook DJ, Christou NV, et al. Multiple organ dysfunction score: a reliable descriptor of a complex clini-cal outcome. Crit Care Med. 1995;23:1638-1652.

447. Ferreira FL, Bota DP, Bross A, et al. Serial evaluation of the SOFA score to predict outcome in critically ill patients. JAMA. 2001;286:1754-1758.

448. Baccala R, Gonzalez-Quintial R, Lawson BR, et al. Sensors of the innate immune system: their mode of action. Nat Rev Rheumatol. 2009;5:448-456.

449. Angus DC, Birmingham MC, Balk RA, et al. E5 murine monoclonal antiendotoxin antibody in gram-negative sepsis: a randomized controlled trial. JAMA. 2000;283:1723-1730.

450. Fisher CJ Jr, Dhainaut J-FA, Opal SM, et al. Recombinant human interleukin 1 receptor antagonist in the treatment of patients with sepsis syndrome: results from a random-ized, double-blind, placebo-controlled trial. JAMA. 1994;271:1836-1843.

451. Abraham E, Glauser MP, Butler T, et al. p55 Tumor necrosis factor receptor fusion protein in the treatment of patients with severe sepsis and septic shock: a randomized con-trolled multicenter trial. Ro 45-2081 Study Group. JAMA. 1997;277:1531-1538.

452. Kieft H, Hoepelman AIM, Zhou W, et al. The sepsis syn-drome in a Dutch university hospital: clinical observations. Arch Intern Med (Chic). 1993;153:2241-2247.

453. Fisher CJ, Agosti JM, Opal SM, et al. Treatment of septic shock with the tumor necrosis factor receptor: Fc fusion protein. N Engl J Med. 1996;334:1697-1702.