Sepsis associated with organ dysfunction, perfusion abnormalities, or hypotension: Organ system dysfunction can be described by organ failure scoring systems (12,13).
|TABLE 46.2 Revised Diagnostic Criteria for Sepsis (11)|
Sepsis with hypotension despite adequate fluid resuscitation, in conjunction with perfusion abnormalities: Standard abnormalities in an adult include mean arterial pressure (MAP) below 60 mmHg, systolic blood pressure (SBP) below 90 mmHg, or a drop in SBP greater than 40 mmHg from baseline.
Multi-Organ Dysfunction Syndrome
Presence of altered organ function in an acutely ill patient, such that homeostasis cannot be maintained without intervention: Primary multi-organ dysfunction syndrome (MODS) is the direct result of a well-defined insult in which organ dysfunction occurs early and can be directly attributable to the insult itself. Secondary MODS develops as a consequence of a host response and is identified within the context of SIRS.
The relationship of many of these conditions to each other is demonstrated in Figure 46.1. An understanding of sepsis definitions has become increasingly important since most clinical trials in the last two decades have utilized modified version of the 1991 sepsis definitions, usually requiring three rather than two SIRS criteria, in their entry criteria. The concept of a compensatory anti-inflammatory response (CARS) has also been introduced after the demonstration that traditional anti-inflammatory mediators were also elevated during sepsis (14). Over a decade after the second consensus conference, a consensus has emerged that further updating is required in order to incorporate advances in the understanding of sepsis pathophysiology and the fundamental role of the host response in producing self-harm (15). A simple example of the problem with current definitions is provided in one study that demonstrated that “the need for two or more SIRS criteria to define severe sepsis excluded one in eight otherwise similar patients with infection, organ failure, and substantial mortality and failed to define a transition point in the risk of death” (16).
Although the sepsis syndromes—from sepsis to septic shock—have been a major burden on human health in both the developed and undeveloped world, epidemiologic information has been surprisingly rare. In North America, this has been caused by the earlier lack of consensus definitions of these syndromes and, more recently, the absence of syndrome-specific diagnostic codes for sepsis within the International Classification of Disease (ICD) coding system. In the last 20 years, the development of consensus definitions and application of computerized hospital and government administrative databases has allowed substantial insight into the problem.
According to the Centers for Disease Control and Prevention (CDC), sepsis rates doubled between 2000 and 2008 (17). Martin et al. (5) have estimated 660,000 annual cases of sepsis in the United States during 2000 (adjusted rate 240/100,000 population) using an analysis of ICD-9 codes associated with National Hospital Discharge Survey data. With the exception of a single major study with much higher values (1), estimates for severe sepsis from sites across North America and Europe have been fairly consistent at 50 to 80/100,000 population (18–22); these cases account for approximately 10% to 15% of all ICU admissions (19,20,22–24). Approximately 25% of cases of sepsis (25) and 50% to 75% of cases of severe sepsis progress to septic shock (23). Septic shock represents between 5% and 8% of all ICU admissions (24,26). In the United States, the cost of sepsis and severe sepsis ranges from $22,000 to $60,000 per episode at a total cost of approximately $20 billion annually in 2011 (1,27,28); sepsis and related conditions represent the 11th leading cause of death in the United States in 2010 according to the CDC (29).
The incidence of sepsis increased approximately 9% per year between 1979 and 2001, with the greatest relative increase in fungal infections. In addition, since late 1980s, gram-positive pathogens have been numerically dominant over gram-negative organisms. However, coding rates for certain specific infections, including pneumonia, have dropped (30). Reasons for this increase include:
- An aging population with increased predisposition to illness
- An increased proportion of the subpopulation has conditions that predispose to systemic infection including chronic organ failure (e.g., cirrhosis, renal failure, cardiomyopathy, chronic obstructive pulmonary disease [COPD]) and other conditions (e.g., diabetes, cancer, HIV, etc.)
- Extensive utilization of invasive diagnostic and therapeutic modalities (indwelling catheters and devices), which lead to breakdown of native resistance to infection
- Widespread use of immunosuppressive chemotherapies for a wide range of diseases (asthma, inflammatory bowel disease, rheumatoid arthritis, systemic lupus erythematosus, and other autoimmune diseases, as well as transplants)
- Increased coding due to greater clinical awareness, which could be a confounder of the increased incidence of sepsis in the last decade (30).
Recently, reported trends detailing severe sepsis incidence and outcomes show important discrepancies. Gaieski et al. (31) showed a 3.5-fold variation in annual incidence of septic shock (from 300 to 1,031 cases/100,000) using the Nationwide Inpatient Sample US hospital population database with four different validated methods for identification of sepsis. These methods were also associated with a twofold difference in mortality (ranging from 14.7% to 29.9%), although all methods estimated a similar annual increased incidence of approximately 13% from 2004 to 2009. Other studies, comparing septic shock in different countries, show large discrepancies in incidence and mortality rates (e.g., Australia and New Zealand 22% mortality vs. Italy 61%) (38). This highlights the need for uniform and consistent method for definitions across national registries to facilitate assessment and outcome comparisons between hospitals across the world.
With respect to individual characteristics, age is a substantial risk factor for sepsis, severe sepsis, and septic shock (1,5,32). Patients over the age of 65 are approximately 13-fold more likely to develop sepsis compared to others (5). Similarly, septic shock is 18-fold more likely in those over 80 years of age compared to those in the 20- to 29-year-old age group (26). Given that the average age of the North American population is increasing and that the incidence of all the sepsis-related syndromes is markedly elevated in the elderly (26), the fact that the average age of patients with sepsis has climbed over the last few decades should be no surprise (1,5). The fact that septic shock is substantially a geriatric illness is reflected in the median age of 67 years (32). The persistent 60:40 male:female preponderance in sepsis, severe sepsis, and septic shock may have its origins in men’s increased predisposition to smoking-associated cases of pneumonia and peptic ulcer disease or gastrointestinal malignancy–associated gastric and bowel perforation (1,5,20,23,25,26). Nonwhite ethnic groups are also at substantially increased risk, particularly blacks (5). However, low socioeconomic status is a substantial risk factor for septic shock, with a fourfold increased risk in the lowest quintile of income compared to any other quintile (26). In this context, it is unclear whether ethnicity may be relevant only as a marker of socioeconomic status. Comorbidities are common in patients with sepsis, as might be expected given an average age of 55 to 65 for sepsis and perhaps higher for septic shock (5,22,32–36). Diabetes, COPD, renal failure, congestive heart failure, and malignancy can each be found in 10% to 20% of patients with sepsis or septic shock. Approximately one-half of patients with severe sepsis have at least one major medical comorbidity (5); patients with septic shock have an even higher incidence (over 90%) of major comorbidities. Alcoholism and substance abuse substantially increases risk of sepsis and death from sepsis and septic shock (37).
As might be expected, mortality increases with the severity of the septic syndrome. Mortality for sepsis is below 15%, for severe sepsis 25% to 50%, and for septic shock over 50% (1,5,18–20,23–25,32,38). This mortality rate for septic shock, while staggering, nevertheless represents an improvement in survival from that of 35 years ago when mortality rates frequently exceeded 80% (39,40). Early septic mortality (less than 3 days) appears to be associated most closely with shock, with other deaths within the first week due to multiple organ failure; later deaths tend to be most closely associated with pre-existing comorbidities (41). Of those succumbing to septic shock, approximately 75% are early deaths—within 1 week of shock—primarily due to hyperdynamic circulatory failure (42).
Throughout recorded history, there has been evolution of the organisms that cause infectious diseases and of the associated clinical syndromes. This phenomenon has become particularly pronounced since the advent of antibiotics in the last half of the previous century. By the 1960s and 1970s, gram-negative organisms had become the dominant pathogens over Staphylococcus aureus and Streptococci. During the 1980s, resistant gram-positive organisms (i.e., methicillin-resistant S. aureus [MRSA], coagulase-negative staphylococci, penicillin-resistant Streptococcus pneumoniae, and enterococci) re-emerged as major pathogens. Gram-positive cocci account for approximately 40% to 50% of single isolates (excluding fungi) in sepsis and septic shock (23,32,38,43–45). In a recent study involving 14,000 ICU patients in 75 countries, gram-negative bacteria were isolated in 62% of patients with severe sepsis who had positive cultures, gram-positive bacteria in 47%, and fungi in 19% (46).
Most recently, yeast and other fungi have demonstrated a remarkable increase in contribution to sepsis (5% of total) and septic shock (8.2% of total) with an increase of about 10% per year (5,32,44,45). Candida albicans remain numerically dominant (about 60% of total fungal infections) but fluconazole-resistant yeasts are the most rapidly increasing species (47–49).
Other major concerns in recent years include the emergence of extended spectrum β-lactamase (ESBL; reliably sensitive only to carbapenems) (50) and carbapenemase-resistant (51) gram-negative bacilli, vancomycin-resistant enterococci (52), and an endemic strain of virulent MRSA in the community (53). In addition, concerns regarding sporadic cases of vancomycin-resistant S. aureus (VRSA) are growing (54).
In terms of the clinical infections associated with sepsis syndromes, lower respiratory tract infections dominate at 25% to 50% of the total in most studies (5,20,23,25,33,34,38,43,55). Intra-abdominal infections account for a disproportionate fraction of cases of severe sepsis (10% to 32%) relative to their contribution to ICU infections (5% to 7%) (44,56). Intra-abdominal infections may have an even greater role in septic shock where they accounted for 29% of cases of septic shock in a recent study (32). Conversely, the urinary system accounts for as much as 16% to 31% of ICU infections, but only 8% to 11% of cases of severe sepsis and septic shock (20,25,32,43,44,56). Positive blood cultures are found typically in only one-third of all cases of sepsis, and one-third are reported to have negative cultures from all sites (55,57).
PATHOGENESIS OF SEPSIS, SEVERE SEPSIS, AND SEPTIC SHOCK
Sepsis and septic shock or sepsis-associated multiple organ failure typically begin with a nidus of infection (e.g., pneumonia, peritonitis, urinary tract infection, abscess); within that nidus, the organism replicates. Eventually, the infection at the inciting focus releases sufficient microbial antigens to elicit a systemic inflammatory response designed to eliminate the invading microbes (Fig. 46.2). A large number of constitutive and/or inducible elements of invasive microorganisms are capable of inciting the systemic inflammatory responses that result in sepsis and septic shock (Table 46.3). Beyond endotoxin (lipopolysaccharide; LPS) of gram-negative bacteria, other major triggers of the systemic inflammatory response characteristic of sepsis include various exotoxins (all bacteria), peptidoglycans (streptococci), and teichoic acid (S. aureus); lipoarabinomannan of mycobacteria; and mannoproteins and β-glucan of fungi (58). Bacterial DNA may possess sufficient antigenic properties—based on unique CG repetitions and lack of deoxyribonucleic acid (DNA) methylation—to initiate a substantial inflammatory response independent of other bacterial elements (59,60,61); bacterial ribonucleic acid (RNA) may able to do the same (62). Recent investigations suggest a surprising commonality of signaling mechanisms in septic shock via toll-like receptors (TLRs) from a broad range of etiologic agents (60,63–66).
Despite the large number of potential elements of pathogenic microorganisms that can drive the septic response, endotoxin of gram-negative bacteria remains the prototype of such factors and the model for subsequent research. This antigen is thought to be central in initiation of the powerful host response during infection with these organisms (67). LPS and other antigens interact with immune cells—particularly macrophages—resulting in the induction of proinflammatory cytokines such as TNFα) and interleukin-1β (IL-1β) secreted by monocytes, macrophages and other cells (Fig. 46.2) (68). These cytokines initiate a complex signaling sequence involving release of secondary mediators—platelet-activating factor, leukotrienes, prostaglandins—monocytes and endothelial tissue factor expression, inducible nitric oxide synthetase (iNOS) induction, microvascular coagulation, cell-adhesion molecule upregulation and apoptosis (69–72). To maintain homeostasis, and likely as part of a feedback mechanism, several anti-inflammatory mediators are also released, including IL-10, transforming growth factor-β (TGFβ), and IL-1 receptor antagonist (IL-1ra). If homeostasis cannot be maintained, progressive and sequential dysfunction of various organ systems (i.e., MODS) may occur. If the inflammatory stimulus is particularly intense or if there is limited cardiovascular reserve, effects on the cardiovascular system as manifested by septic shock may dominate the clinical presentation.
MICROBIAL ANTIGEN SIGNALING
As the prototypical and best-studied microbial antigen, an understanding of signaling cascade of endotoxin is instructive. Endotoxin is an amphiphilic macromolecule located on the outer cell wall membrane of gram-negative bacteria. It is composed of lipid A, a diglucosamine-based acylated phospholipid, and a polysaccharide side chain (Fig. 46.3) (73,74). The polysaccharide chain is composed of a short, highly conserved, proximal section (core polysaccharide) and a highly variable, longer distal oligosaccharide side chain. The core polysaccharide and lipid A are sometimes referred to as the core glycolipid. The highly conserved lipid A moiety is the toxic element of endotoxin and can reproduce the manifestations of endotoxic shock when administered alone (74–79). As a circulating form in the plasma, endotoxin exists in a multimeric, aggregate form.
Lipopolysaccharide binding protein (LBP) is an acute-phase reactant protein present in plasma (73,80,81). The levels increase with inflammatory stimulation. LBP catalyzes the transfer of endotoxin from serum aggregates to either serum lipoproteins such as high-density lipoprotein (HDL) leading to endotoxin neutralization or to CD14 receptors (either membrane-bound [mCD14] or soluble [sCD14]), the putative primary LPS receptor (Fig. 46.4). The degree to which endotoxin is shunted through either pathway appears to have a significant role in the phenotypic physiologic response (58). LBP, by forming a complex with endotoxin monomers, appears to enhance the ability of endotoxin to bind CD14 and allows cellular activation at relatively low endotoxin concentrations (73,81). Although LBP appears to be a specific carrier molecule for endotoxin, available data suggests that other microorganism toxins associated with sepsis may use similar carrier proteins (82,83).
|TABLE 46.3 Elements of Microorganisms Capable of Inducing Septic Response|