YALE JOURNAL OF BIOLOGY AND MEDICINE 92 (2019), pp.629-640.

Review

Organ Dysfunction in Sepsis: An Ominous

Trajectory From Infection To Death
César Caraballoa and Fabián Jaimesb,c,d,*
a
Center for Outcomes Research and Evaluation, Yale New Haven Health, New Haven, CT, USA; bAcademic Group of Clinical
Epidemiology, School of Medicine, University of Antioquia, Medellín, Colombia; cDepartment of Internal Medicine, School
of Medicine, University of Antioquia, Medellín, Colombia; dResearch Direction, San Vicente Foundation University Hospital,
Medellín, Colombia

Sepsis is a highly complex and lethal syndrome with highly heterogeneous clinical manifestations that
makes it difficult to detect and treat. It is also one of the major and most urgent global public health
challenges. More than 30 million people are diagnosed with sepsis each year, with 5 million attributable
deaths and long-term sequalae among survivors. The current international consensus defines sepsis as a
life-threatening organ dysfunction caused by a dysregulated host response to an infection. Over the past
decades substantial research has increased the understanding of its pathophysiology. The immune response
induces a severe macro and microcirculatory dysfunction that leads to a profound global hypoperfusion,
injuring multiple organs. Consequently, patients with sepsis might present dysfunction of virtually any
system, regardless of the site of infection. The organs more frequently affected are kidneys, liver, lungs,
heart, central nervous system, and hematologic system. This multiple organ failure is the hallmark of sepsis
and determines patients’ course from infection to recovery or death. There are tools to assess the severity
of the disease that can also help to guide treatment, like the Sequential Organ Failure Assessment (SOFA†)
score. However, sepsis disease process is vastly heterogeneous, which could explain why interventions
targeted to directly intervene its mechanisms have shown unsuccessful results and predicting outcomes
with accuracy is still elusive. Thus, it is required to implement strong public health strategies and leverage
novel technologies in research to improve outcomes and mitigate the burden of sepsis and septic shock
worldwide.

*To whom all correspondence should be addressed: Dr. Fabián Jaimes, Hospital San Vicente Fundación, Calle 64 # 51 D-154,
Medellín, Antioquia, Colombia; Tel: +57 (4) 2192433, Email: [email protected].

†Abbreviations: Ang-Tie, angiopoietin-tyrosine kinase with immunoglobulin-like loop epidermal growth factor domain ligand-receptor;
iNOS, inducible nitric oxide synthase; NO, nitric oxide; IL-1β, interleukin-1β; IL-6, interleukin-6; VCAM-1, vascular cell adhesion
molecule-1; ATP, adenosine triphosphate; ARDS, acute respiratory distress syndrome; TNF-α, tumor necrosis factor alpha; PaO2,
partial pressure of arterial oxygen; FIO2, fraction of inspired oxygen; AKI, acute kidney injury; ICAM-1, intercellular adhesion
molecule-1; DIC, disseminated intravascular coagulation; APPs, acute-phase proteins; SOFA, Sequential (sepsis-related) Organ
Failure Assessment.

Keywords: sepsis, septic shock, infection, organ dysfunction, organ failure, mortality

Author Contributions: CC and FJ designed the manuscript. CC did the literature review and wrote the first draft, with FJ overseeing
and contributing. Both authors contributed and approved the final version of the manuscript. Neither CC nor FJ received funding for
the preparation or submission of this work.

Copyright © 2019
629
630 Caraballo and Jaimes: Sepsis and death

INTRODUCTION fore systemic perfusion. This paracellular leakage seems

to be caused by a diffuse endothelial injury and dys-
Sepsis is an intricate, heterogeneous, and highly le- function mediated by proinflammatory molecules [19].
thal syndrome that can be hard to identify and treat [1]. Particularly, recent research underscores the important
Defined as a life-threatening organ dysfunction caused by role in the imbalance of the angiopoietin-tyrosine kinase
a dysregulated host response to an infection [2], sepsis with immunoglobulin-like loop epidermal growth factor
is one of the major and most urgent public health chal- domain ligand-receptor system (Ang-Tie) in patients
lenges worldwide [3,4]. It is estimated that more than 30 with sepsis. The augmented expression of Ang-2 and the
million people globally are diagnosed with sepsis each inhibition of Ang-1 blocks Tie-2 receptor and increases
year, leading to 5 million deaths [5], with high economic vascular permeability, causing tissue edema [20]. Its
burden and long-term morbidity among survivors [6]. prognostic value has been demonstrated in clinical stud-
Particularly, annually in the United States sepsis is pres- ies where high serum Ang-2/Ang-1 ratio was associated
ent in 1.7 million hospitalized patients and contributes to with increased severity of organ dysfunction and higher
270,00 deaths [7]. mortality, even in early sepsis [21-23].
Prognosis in sepsis is influenced by characteristics of Although in most cases these volume distribution
the patient (e.g. age, immunologic status, comorbidities, abnormalities can be countered by a successful resusci-
among others) [8-10] and characteristics of the infection tation with an adequate and rational vascular volume ex-
(e.g. pathogen type, virulence, site of infection, inoculum, pansion [24], some patients have a concomitant persistent
among others) [8,11,12]. Although combinations of such vasodilatory state that impedes adequate perfusion even
characteristics influence the clinical presentation and after achieving an euvolemic state. This clinical scenario
risk, sepsis is a common pathway from infection to death, is known as septic shock, the most severe manifestation
in which progressive organ dysfunction is the mean. In of sepsis. Vascular smooth muscle fails to contract with
this review, we present a comprehensive overview of the neurohormonal stimulus, resulting in a profound system-
features found in patients with sepsis that lead to multiple ic arterial and venous vasodilation [25] that reduces the
organ failure and death. pressure gradient required for venous return and, sub-
sequently, decreases cardiac output [26]. Although the
FROM INFECTION TO ORGAN mechanisms of such a dramatic vascular dysfunction are
DYSFUNCTION not well understood, inflammation-induced endothelial
dysfunction seems to be associated with an over-expres-
Sepsis definition has changed over the last few de- sion of an inducible nitric oxide synthase (iNOS) [27].
cades as our understanding of it has increased [2,13,14], The subsequent excessive production of nitric oxide (NO)
and its current definition emphasizes the presence of or- directly induces vascular smooth muscle cells relaxation
gan dysfunction (Table 1). The cornerstone of sepsis-in- and hyperpolarization, preventing their response to vaso-
duced organ damage is the instauration and perpetuation constrictors and thus perpetuating hypotension [25,28]. A
of a mismatch between perfusion and tissues metabolic deficiency of vasopressin with paradoxical simultaneous
requirements. Inflammation-induced cardiac dysfunction downregulation of vasoconstrictive receptors has also
and systemic blood volume redistribution have pivotal been described during septic shock, but its mechanism
roles on this, but are exacerbated by an impaired tissue in humans is yet to be fully understood and therapies tar-
oxygen utilization [15]. This sepsis-induced global hy- geted directly to reverse these maladaptive mechanisms
poperfusion state has common clinical manifestations have been unsuccessful [29-32].
such as hypotension, decreased capillary refill time, mot-
tled skin, and cold extremities. Besides the early initia- Cardiac Dysfunction: the Failure of the Pump
tion of antibiotic therapy and source control—which are
essential for sepsis treatment and significantly reduces After volume resuscitation or vasopressors initiation,
the risk of death [16,17]—the recommended early resus- venous return augments and patients enter in a hyperdy-
citation strategies for patients with sepsis or septic shock namic profile characterized by high cardiac output and
intend to reestablish an adequate organ perfusion [16]. low systemic vascular resistance [33]. This response,
however, is often accompanied by a depressed myocardi-
Vascular Dysfunction: the Failure of the Circuit al function. Pro-inflammatory cytokines, such as interleu-
kin-1β (IL-1β) and interleukin-6 (IL-6), depress cardio-
Several changes occur simultaneously in the system- myocyte contractility and induces expression of vascular
ic vascular bed in patients with sepsis, with an increasing cell adhesion molecule-1 (VCAM-1) in the coronary
interest in the importance of microcirculatory injury and endothelium, which mediates infiltration of neutrophils
dysfunction [18]. Capillary permeability is increased, to the myocardium [34,35]. Importantly, NO exacerbates
compromising the effective vascular volume and there-
 Caraballo and Jaimes: Sepsis and death 631

Table 1. Sepsis and Septic Shock Clinical Criteria Over Time

Consensus Clinical criteria
Sepsis-1, 1991 [13] Sepsis:
Systemic response to an infection, manifested by two or more of the following components of the systemic
inflammatory response syndrome (SIRS): a) temperature >38°C or <36°C; b) heart rate >90 beats per
minute; c) respiratory rate >20 breaths per minute or PaCO2 <32mmHg; and d) white cell blood count
>12,000 cells per mL, <4,000 cells per mL, or >10% immature forms.

Severe sepsis:
Sepsis associated with organ dysfunction, hypoperfusion, or hypotension.

Septic shock:
Sepsis-induced hypotension (SBP <90 mmHg or an SBP reduction ≥40 mmHg from baseline) despite
adequate fluid resuscitation, or requiring vasopressor agents, along with the presence of perfusion
abnormalities.
Sepsis-2, 2001 [14] Sepsis:
Documented or suspected infection with some signs of systemic inflammation, which were expanded
from the SIRS criteria to include abnormalities from 5 major categories (general variables, inflammatory
variables, hemodynamic variables, organ dysfunction variables, and tissue perfusion variables).

Severe sepsis:
Sepsis associated with organ dysfunction, which can be estimated with the SOFA score.

Septic shock:
Persistent arterial hypotension (SBP <90 mmHg, MAP <60 mmHg, or reduction in SBP >40 mmHg from
baseline) despite adequate fluid resuscitation and unexplained by other causes.
Sepsis-3, 2015 [2] Sepsis:
Suspected or documented infection and acute organ dysfunction (defined as an increase of ≥ 2 points in
SOFA points).

Septic shock:
Sepsis and vasopressor therapy needed to elevate MAP ≥65 mmHg and lactate >2 mmol/L despite
adequate fluid resuscitation.
PaCO2: partial pressure of carbon dioxide; SBP: systolic blood pressure; MAP: mean arterial blood pressure; SOFA: Sequential
Organ Failure Assessment.

mitochondrial dysfunction diminishing myocardial oxy- et al. found that 13 percent of patients hospitalized due to
gen utilization, perpetuates release of pro-inflammatory sepsis experienced at least one incidental cardiac event
cytokines, and downregulates β-adrenergic receptors and had 30 percent higher risk of death than those who
[35,36]. did not [46].
Consequently, almost 1 out of 3 patients with sepsis
presents reversible left ventricular systolic impairment, Microcirculation and Cellular Dysfunction: the
driven by hypokinesia and reduced ejection fraction, with Failure in Final Oxygen Delivery and Utilization
unclear implication on survival [37]. On the other hand, While most therapeutic efforts are directed to solve
left diastolic dysfunction is present in 1 out of 2 patients the overt hemodynamic dysfunction, changes in the
and is associated with an 80 percent increased risk of microcirculation have an important role in perpetuating
death [38]. Similarly, nearly 1 out of 2 patients with sep- the organ injury even after restoration of hemodynamic
sis have right ventricular dysfunction, with an associated abnormalities. Various mechanisms can explain this
60 percent increased risk of death [39]. microcirculatory failure. The endothelial dysfunction
Furthermore, chronotropic response, the ability to and injury over-expression of iNOS is not homogeneous
modify the heart rate according to systemic requirements, thorough all organ beds, causing shunting of the flow and
is also often impaired in sepsis [40]. A recent study found hypoperfusion on the underexpresed tissues [47]. This
that those with low heart rate variability had nearly six situation is aggravated by occlusion of terminal circula-
times higher hazard of death [41]. On the other hand, tion vessels due to sepsis-induced erythrocyte decreased
sepsis is also associated with incidental clinical cardiac deformability, greater platelet aggregability, and micro-
events like acute heart failure, life-threatening arrhyth- thrombi formation [27,48]. Moreover, NO has a pivotal
mias, myocardial infarction, and non-ischemic myocar- role in the impairment of cellular oxygen utilization.
dial injury, among others [42-45]. In a recent study, Patel Regardless of the restoration of adequate tissue perfusion
632 Caraballo and Jaimes: Sepsis and death

or oxygen delivery, NO inhibits mitochondrial respira- quate volume resuscitation, hemodynamic drug support is
tion by disrupting the respiratory chain, which depletes recommended with the goal of achieving and maintaining
ATP and causes cellular dysfunction and organ injury a mean arterial blood pressure target of ≥ 65 mmHg [16].
[27,49,50]. Norepinephrine is the recommended first-choice vaso-
pressor due to its effectiveness and lower rate of adverse
Indicators of Perfusion Status events when compared to other options like dopamine
The overall effect of such an inadequate systemic [16,65], and its adoption as such is consistent among
oxygen delivery and its impaired cellular utilization has intensive care specialists worldwide [66]. However, a
major implications for tissues metabolism, increasing proportion of patients do not achieve the mean arterial
anaerobic glycolysis. This results in a higher production pressure target despite high doses of this catecholamine,
of lactate as a byproduct of pyruvate metabolism, as less reflecting the high underlying heterogeneity in the patho-
of it enters Krebs aerobic cycle. Hyperlactatemia, defined physiology of this syndrome. These non-responders have
as a serum concentration >2 mmol/L, is associated with higher mortality risk, their optimal treatment is still not
higher risk of death in patients with sepsis, independently well known, and are the focus of recent research in criti-
of hemodynamic status [51,52]. Its prognostic relevance cal care [67,68]. Recently, Chawla et al. proposed that in
is underscored by the fact that those with hyperlactatemia order to avoid prolonged hypotension, every patient with
alone (i.e. no hypotension or need for vasopressor thera- septic shock should be started on multiple vasopressors
py) have a higher risk of death than those with hypoten- of different mechanism of action and de-escalated after-
sion and normal serum lactate levels [53]. This phenotype wards according to their response, similar to the “broad
of normotension with hyperlactatemia have led to the spectrum antibiotics” approach [69].
term of “cryptic shock” [15]. Thus, lactate is commonly
used as an indicator of patients’ perfusion status and its BEYOND CIRCULATORY FAILURE: SEPSIS
sequential measurement is included in the recommended IMPLICATIONS ON OTHER ORGANS
approach to patients with sepsis as its clearance seems
indicative of an effective resuscitation [16,54]. Recent Given that sepsis is a continuous process of con-
studies have assessed the association between hyperlac- comitant insults occurring thorough the body, its damage
tatemia and clinical signs to assess the perfusion status should not be understood as isolated events on different
and guide resuscitation, aiming to identify a bed-side systems. However, for conceptualization we here describe
option. However, an observational study found no as- how sepsis affects specific organs beyond the circulatory
sociation between lactate levels and capillary refill time system and their prognostic implications.
[55] and a clinical trial found no statistically significant
benefit in survival by using the same clinical perfusion Lungs
indicator versus lactate [56]. Sepsis is the most common cause of acute respiratory
distress syndrome (ARDS) [70] and 40 percent of pa-
Strategies Aimed to Restore Tissue Perfusion tients with sepsis or septic shock develop it [71]. ARDS is
The cornerstone of sepsis and septic shock initial characterized by an acute respiratory failure with diffuse
treatment is to overcome such systemic hypoperfusion pulmonary infiltrates caused by alveolar injury and an in-
[16,24]. As mortality risk increases with the duration of creased pulmonary vascular permeability to protein-rich
hypotension [57], current guidelines recommend that at fluid. Although its etiology is yet to be fully understood,
least 30 mL/Kg of crystalloids should be given during the studies have shown that this alveolar barrier injury is
first 3 hours of treatment, with additional fluids adminis- mediated by proinflammatory cytokines—such as tumor
tration guided by a comprehensive and frequent hemody- necrosis factor alpha (TNF-α) or IL-1β—the widespread
namic status reassessment to avoid volume overload [16]. endothelial barrier dysfunction, platelet activation with
However, the strength of the recommendation is weak, microthrombi formation, and neutrophils extracellular
and some studies suggest that such an aggressive early traps formation [72-74]. This edema and alveolar damage
goal-directed therapy is not beneficial [58] and might ac- increase physiological dead space impairing gas exchange
tually increase the risk of adverse outcomes—mainly re- and causing severe hypoxemia and hypercapnia [75]. The
spiratory failure and death—in resource-limited settings severity of the condition is evaluated using the ratio of
[59-61]. As evidence suggests that a more conservative the partial pressure of arterial oxygen to the fraction of
approach is effective and safe [62], there has been an in- inspired oxygen (PaO2/FiO2), as well as the mechanical
creased interest in a more personalized fluid management ventilatory parameters required by the patient. Mortality
[63,64]. among those with ARDS is high, ranging from 35 percent
For those with persistent hypotension despite ade- to 46 percent [76]. Furthermore, those with sepsis-relat-
ed ARDS have higher 60-day mortality than those with
 Caraballo and Jaimes: Sepsis and death 633

ARDS caused by any other reason [77]. Whereas these be propagated by neutrophils extracellular traps, which
patients benefit from lung-protective mechanical ventila- induce platelet aggregation, thrombin production, and fi-
tion strategies to aid respiratory muscles and maintain ad- brin clots formation [98]. Then, microthrombi formation
equate gas exchange [78], pharmacological interventions in small vessels further impairs perfusion and oxygen
to prevent the occurrence or mitigate the impact of ARDS delivery, causing organ injury and dysfunction [23].
on survival have been unsuccessful [79,80]. Overall, this procoagulant up-regulation causes
platelet consumption and coagulation factors depletion,
Kidneys leading to the classical sepsis-associated thrombocyto-
The renal system is another common target of this penia and overt disseminated intravascular coagulation
progressive organ dysfunction. Sepsis is the most com- (DIC), especially with expression of tissue factor and
mon contributing factor for acute kidney injury (AKI) secretion of von Willebrand factor when monocytes and
in critically-ill patients, [81] and more than half of pa- endothelial cells are activated to the point of cytokine
tients with sepsis or septic shock develop it [82,83]. AKI release following injury [99]. Among patients with sepsis
is defined as a serum creatinine increase of ≥ 0.3 mg/dl and septic shock, up to 55 percent and 61 percent have
in 48 hours, 50 percent increase from baseline in 7 days thrombocytopenia and/or DIC, respectively [100,101].
or urine output < 0.5ml/kg/h for more than 6 hours [84]. Both of these conditions are associated with worse out-
Patients with sepsis-associated AKI have 62 percent and comes such as higher risk of major bleeding events and
36 percent higher risk of in-hospital mortality compared death [97,101-105]. The current treatment of these coag-
to those with sepsis without AKI [85] and to those with ulation abnormalities consist on prevention and treatment
non-sepsis associated AKI, respectively [86]. Despite of major bleeding events [106], whereas therapies aimed
its high frequency, the underlying mechanisms of sep- to intervene the pathophysiology of this condition have
sis-associated AKI are not completely understood. Renal been unsuccessful [107-109].
hypoperfusion leading to acute tubular necrosis has been
the paradigm, but current evidence suggests an even more Liver
important role of the local microcirculation and inflam- The liver is far from a bystander in sepsis: it is a reg-
matory signals, including ischemia-reperfusion injury, ulator of the inflammatory process and a target of host
oxidative stress, and tubular apoptosis [87,88]. Moreover, response. When exposed to lipopolysaccharides, Kupffer
sepsis treatment can also contribute to AKI by the usage cells increase the release of IL-1β, IL-6, and TNF-α
of nephrotoxic drugs and excessive or less-physiological [110,111]. In response to the proinflammatory cytokines,
fluid resuscitation. Volume overload increases central hepatocytes release acute-phase proteins (APPs) into
venous pressure, which also increases renal vascular systemic circulation, with widespread proinflammatory
pressure, causing subsequent organ edema, increased and anti-inflammatory effects [112]. Thus, it has been
intracapsular pressure and decreased glomerular filtration hypothesized that hepatocytes, via APPs, have a pivotal
rate [89,90]. There is a recent interest is the role of resus- role in balancing the immune response in sepsis, prevent-
citation fluid selection in the development of sepsis-as- ing an excessive inflammatory or immunosuppressed
sociated AKI. When compared to balanced crystalloids state [111]. This regulatory role gains importance when
(e.g. lactated Ringer’s solution), evidence suggests that considering that up to 46 percent of patients with sepsis
the high concentration of chloride in normal saline (0.9% have concomitant hepatic dysfunction [113], which has
sodium chloride) might be associated with worse renal been associated with a higher 28-day mortality [114].
outcomes and survival [91-95]. Two major mechanisms seem to explain the liver injury
and subsequent dysfunction in sepsis: hypoxic hepatitis
Coagulation System and sepsis-induced cholestasis. Hypoxic hepatitis is com-
Pro-inflammatory cytokines also increase the endo- monly defined as a clinical setting that leads to reduced
thelial luminal expression and serum circulation of inter- oxygen delivery or utilization by the liver (e.g. cardiac,
cellular adhesion molecule-1 (ICAM-1) and VCAM-1, respiratory, or circulatory failure), with an increase of at
contributing to platelet adhesion and coagulation cascade least 20-fold the upper limit of normal serum aminotrans-
activation. Additionally, the anticoagulant mechanisms ferase levels, and without other potential causes of liver
are downregulated by the same proinflammatory cyto- injury [115,116]. In sepsis, the profound hemodynamic
kines. Endothelial production of thrombomodulin—a alterations, microthrombi formation, sinusoidal obstruc-
glycoprotein that inhibits conversion of fibrinogen to tion, and endothelium dysfunction impairs liver perfusion
fibrin by binding thrombin—is severely impaired, reduc- leading to subsequent injury and hypoxic hepatitis [112].
ing activation of Protein C, a strong anticoagulant with In a recent study that included 1116 critically ill patients
fibrinolytic properties [96,97]. Interestingly, this seems to with this condition [117], sepsis was the second leading
predisposing factor of hypoxic hepatitis, with an in-hos-
634 Caraballo and Jaimes: Sepsis and death

pital mortality of 53 percent, only behind cardiac failure. cians to systematically follow patients’ progress through-
On the other hand, the definition of sepsis-induced out the hospitalization and adjust treatment accordingly.
cholestasis is not as well standardized, its etiology is still Since the different potential organ dysfunction we have
to be elucidated, and its prognostic relevance is not clear. mentioned so far does not occur on a strictly linear or
Sepsis-induced cholestasis is understood as an impaired isolated manner but are part of a highly complex and in-
bile formation and defective flow caused by a non-ob- tegrated process—and not all occur in every patient—the
structive intrahepatic insult [112], and its diagnosis is challenge for the clinicians and researchers has been to
commonly made by an elevation of total serum bilirubin objectively assess the true magnitude or “amount” of
greater than 2 mg/dl and aminotransferases greater of at organ failure for each patient. Accordingly, in 1996 Vin-
least 2-fold the upper normal limit [118]. Animal mod- cent et al. [132] presented the Sequential (sepsis-related)
els have suggested that proinflammatory cytokines alter Organ Failure Assessment (SOFA) score, with the goal
the hepatocytes expression of bile acids transporters, of objectively estimating the degree of organ dysfunction
reverting the normal bile acid transport into the blood. over time in patients with sepsis. SOFA evaluates the
Furthermore, pro-inflammatory cytokines and NO lead to respiratory, hematologic, cardiovascular, hepatic, renal,
ductular cholestasis by inhibiting cholangiocytes secre- and central nervous system of each patient, assigning
tion [119,120]. each system a value from 0 (normal organ function) to 4
(most abnormal organ function). Therefore, SOFA score
Central Nervous System ranges from 0 to 24 (Table 2).
Up to 70 percent of critically-ill patients with sep- Although it was not originally intended as a predic-
sis have any degree of sepsis-associated encephalopathy tive model, the association between organ dysfunction
[121]. Beyond the direct infections of the brain and its and death has inspired research about the usage of SOFA
surrounding tissues (e.g. encephalitis or meningitis), sep- to predict mortality in patients with sepsis, showing good
sis injures the central nervous system by a wide range predictive performance [133-137]. Notably, the latest
of mechanisms, with the mismatch of systemic perfusion consensus complemented the conceptual definition of
over metabolic requirements having an essential role. sepsis by defining life-threatening organ dysfunction as
The severe systemic hemodynamic instability can over- an acute change in total SOFA score ≥ 2 points conse-
come the central nervous system finely tuned perfusion quent to the infection [2], since such change was associ-
regulation mechanisms, leading to critical brain ischemic ated with approximately 10 percent increased mortality
lesions [122]. Additionally, the onset of cardiac arrhyth- risk [137]. A pending issue, however, is the improvement
mias and sepsis-induced coagulopathy may further ex- of the cardiovascular component of SOFA, as it does not
plain the increased the risk of ischemic and hemorrhagic directly measure that organ dysfunction, but the require-
stroke among patients with sepsis [123-126]. On the other ment of specific interventions that have changed in the
hand, the marked inflammatory response contributes to last years [138].
microcirculatory failure and disruption of the blood-brain The components of SOFA require tests and resourc-
barrier, allowing inflammatory mediators and neurotox- es that might not be readily available at bedside outside
ins into brain tissue [127]. Importantly, the increased intensive care units (ICU), which limits its application on
NO diffuses even through the intact blood-brain barrier other settings. Given that nearly half of patients with sep-
causing oxidative stress, which can lead to neuronal sis present in the emergency department [139], alternative
dysfunction and apoptosis [128]. The disruption of cho- tools have been developed for early sepsis detection out-
linergic and dopaminergic neurotransmission also play a side the ICU. Commonly used scores that use bedside-on-
key role in this acute brain dysfunction [129,130], which ly measures with this intention are the Modified Early
can range from delirium to seizures and comma [127]. Warning Score (MEWS) [140], the National Early Warn-
Moreover, when critical areas—like the brainstem—are ing Score (NEWS) [141], and the quick SOFA (qSOFA)
compromised by these insults, the autonomic dysfunction [137]. MEWS considers patients systolic blood pressure,
is exacerbated, perpetuating the hemodynamic instability heart rate, respiratory rate, temperature, and level of con-
and increasing the risk of death [129,131]. sciousness, whereas NEWS also considers the SpO2. On
the other hand, qSOFA was introduced in the latest sepsis
consensus and assess for abnormalities in respiratory
ESTIMATING THE MAGNITUDE rate, systolic blood pressure, and mental status. However,
AND IMPORTANCE OF THE ORGAN it has performed worse than MEWS and NEWS identify-
DYSFUNCTION ing critically-ill infected patients [142,143], despite the
Standard and accurate criteria for organs dysfunction fact that these were not developed to screen for sepsis
are of great importance in critical care since it helps clini- but to identify patients with high-risk of major in-hospital
complications or ICU admission. The leverage of novel
 Caraballo and Jaimes: Sepsis and death 635

Table 2. Sequential Organ Failure Assessment Scorea

Score
System 0 1 2 3 4
Respiratory
PaO2/FIO2, mm Hg ≥400 <400 <300 <200 with respiratory <100 with respiratory
support support
Coagulation
Platelets, ×103/μl ≥150 <150 <100 <50 <20
Liver
Bilirubin, mg/dl <1.2 1.2-1.9 2.0-5.9 6.0-11.9 >12.0
Cardiovascular
Mean arterial pressure ≥70 mm Hg <70 mm Hg Dopamine <5 or Dopamine 5.1-15 Dopamine >15
or adrenergic agent dobutamine (any or epinephrine ≤0.1 or epinephrine >0.1
administered for at least dose)b or norepinephrine ≤0.1b or norepinephrine >0.1b
1 hour
Central nervous system
Glasgow Coma score 15 13-14 10-12 6-9 <6
Renal
Creatinine or urine <1.2 mg/dl 1.2-1.9 mg/dl 2.0-3.4 mg/dl 3.5-4.9 mg/dl or >5.0 mg/dl or
output <500 ml/day <200 ml/day
PaO
a 2
: partial pressure of arterial oxygen; FIO2: fraction of inspired oxygen.
Adapted from Vincent JL et al. [132]
b
Doses are presented as μg/kg/min

research methods and digital resources, such as artificial regions will help design strategies that improve care and
intelligence and electronic health records, have shown survival.
promise in improving the accuracy of personalized risk
estimation [144]. In a remarkable example of this, Delah- CONCLUSION
anty et al. [145] used machine learning methods to iden-
tify patients at high risk of sepsis in more than 2 million Sepsis is a highly complex and lethal syndrome with
medical encounters, developing a new sepsis screening a convoluted pathway from infection to death consisting
tool that outperformed the rest. of multiple organ dysfunction. Each organ injury contrib-
utes to the patient’s risk of death, with an intricate cross-
FUTURE PERSPECTIVES TO REDUCE THE talk among the whole system. Despite its high prevalence
GLOBAL BURDEN OF SEPSIS and intensive research, the vast underlying heterogeneity
of sepsis might be the reason for the failure of interven-
Even though sepsis is recognized as an urgent health tions beyond supportive measures—including infection
challenge worldwide [4], its current global burden may control—in improving outcomes. A more personalized
be underestimated due to scarcity of information avail- approach is needed, and the recent advances using novel
able from lower and middle-income countries, where research methodologies have provided promising results
most cases of sepsis might occur [146,147]. Thus, to ad- in this regard. This research and subsequent interventions
dress this disparity in representation, the improvements will need the support from strong public health initiatives
in acute and individual treatment of sepsis need to be worldwide. All these efforts will continue to help patients
backed-up by strong public health strategies that im- with sepsis to change their trajectory away from death
proves its understanding worldwide. The African Sepsis and towards recovery.
Alliance signed the Kampala Declaration in 2017 and the
Latin-American Institute of Sepsis signed the São Paulo REFERENCES
Declaration in 2018, both calling for urgent national and
1. Angus DC, van der Poll T. Severe sepsis and septic shock. N
international actions to improve the prevention, diagno-
Engl J Med. 2013;369(9):840–51.
sis, and treatment of sepsis and to dedicate human and 2. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M,
financial resources to these goals [148,149]. Hopefully, Annane D, Bauer M, et al. The Third International Con-
these calls for action will resonate and the increased un- sensus Definitions for Sepsis and Septic Shock (Sepsis-3).
derstanding of sepsis burden in lower and middle-income JAMA. 2016;315(8):801–10.
636 Caraballo and Jaimes: Sepsis and death

3. World Health Organization. Seventieth World Health As- shock from the first hour: results from a guideline-based
sembly. Improving the prevention, diagnosis and clinical performance improvement program. Crit Care Med.
management of sepsis. 2017. Accessed on May 5, 2019. 2014;42(8):1749–55.
Available at: http://apps.who.int/gb/ebwha/pdf_files/ 18. Hawiger J, Veach RA, Zienkiewicz J. New paradigms in
WHA70/A70_R7-en.pdf?ua=1&ua=1 sepsis: from prevention to protection of failing microcircu-
4. Reinhart K, Daniels R, Kissoon N, Machado FR, Schachter lation. J Thromb Haemost. 2015;13(10):1743–56.
RD, Finfer S. Recognizing Sepsis as a Global Health Prior- 19. Rudiger A, Singer M. Mechanisms of sepsis-induced cardi-
ity - A WHO Resolution. N Engl J Med. 2017;377(5):414– ac dysfunction. Crit Care Med. 2007;35(6):1599–608.
7. 20. Fiedler U, Augustin HG. Angiopoietins: a link between
5. Fleischmann C, Scherag A, Adhikari NK, Hartog CS, angiogenesis and inflammation. Trends Immunol.
Tsaganos T, Schlattmann P, et al. Assessment of Global 2006;27(12):552–8.
Incidence and Mortality of Hospital-treated Sepsis. Current 21. Ricciuto DR, dos Santos CC, Hawkes M, Toltl LJ, Conroy
Estimates and Limitations. Am J Respir Crit Care Med. AL, Rajwans N, et al. Angiopoietin-1 and angiopoie-
2016;193(3):259–72. tin-2 as clinically informative prognostic biomarkers of
6. Tiru B, DiNino EK, Orenstein A, Mailloux PT, Pesaturo A, morbidity and mortality in severe sepsis. Crit Care Med.
Gupta A, et al. The Economic and Humanistic Burden of 2011;39(4):702–10.
Severe Sepsis. Pharmacoeconomics. 2015;33(9):925–37. 22. Fang Y, Li C, Shao R, Yu H, Zhang Q, Zhao L. Prognostic
7. Rhee C, Dantes R, Epstein L, Murphy DJ, Seymour CW, significance of the angiopoietin-2/angiopoietin-1 and an-
Iwashyna TJ, et al. Incidence and Trends of Sepsis in US giopoietin-1/Tie-2 ratios for early sepsis in an emergency
Hospitals Using Clinical vs Claims Data, 2009-2014Inci- department. Crit Care. 2015;19:367.
dence and Trends of Sepsis in US Hospitals, 2009-2014In- 23. Fang Y, Li C, Shao R, Yu H, Zhang Q. The role of bio-
cidence and Trends of Sepsis in US Hospitals, 2009-2014. markers of endothelial activation in predicting morbidity
JAMA. 2017;318(13):1241–9. and mortality in patients with severe sepsis and septic
8. Mayr FB, Yende S, Angus DC. Epidemiology of severe shock in intensive care: A prospective observational study.
sepsis. Virulence. 2014;5(1):4–11. Thromb Res. 2018;171:149–54.
9. Rowe TA, McKoy JM. Sepsis in Older Adults. Infect Dis 24. Guarracino F, Bertini P, Pinsky MR. Cardiovascular deter-
Clin North Am. 2017;31(4):731–42. minants of resuscitation from sepsis and septic shock. Crit
10. Sinapidis D, Kosmas V, Vittoros V, Koutelidakis IM, Care. 2019;23(1):118.
Pantazi A, Stefos A, et al. Progression into sepsis: an in- 25. Burgdorff AM, Bucher M, Schumann J. Vasoplegia in
dividualized process varying by the interaction of comor- patients with sepsis and septic shock: pathways and mecha-
bidities with the underlying infection. BMC Infect Dis. nisms. J Int Med Res. 2018;46(4):1303–10.
2018;18(1):242. 26. Bloch A, Berger D, Takala J. Understanding circulatory
11. Caraballo C, Ascuntar J, Hincapie C, Restrepo C, Bernal E, failure in sepsis. Intensive Care Med. 2016;42(12):2077–9.
Jaimes F. Association between site of infection and in-hos- 27. Morelli A, Passariello M. Hemodynamic coherence in sep-
pital mortality in patients with sepsis admitted to emergen- sis. Best Pract Res Clin Anaesthesiol. 2016;30(4):453–63.
cy departments of tertiary hospitals in Medellin, Colombia. 28. Barrett LK, Singer M, Clapp LH. Vasopressin: mechanisms
Rev Bras Ter Intensiva. 2019;31(1):47–56. of action on the vasculature in health and in septic shock.
12. Klastrup V, Hvass AM, Mackenhauer J, Fuursted K, Crit Care Med. 2007;35(1):33–40.
Schonheyder HC, Kirkegaard H. Site of infection and 29. Sharawy N, Lehmann C. New directions for sepsis and
mortality in patients with severe sepsis or septic shock. A septic shock research. J Surg Res. 2015;194(2):520–7.
cohort study of patients admitted to a Danish general inten- 30. Sharawy N. Vasoplegia in septic shock: do we really fight
sive care unit. Infect Dis (Lond). 2016;48(10):726–31. the right enemy? J Crit Care. 2014;29(1):83–7.
13. American College of Chest Physicians/Society of Critical 31. Levy B, Collin S, Sennoun N, Ducrocq N, Kimmoun A,
Care Medicine Consensus Conference: definitions for sep- Asfar P, et al. Vascular hyporesponsiveness to vasopressors
sis and organ failure and guidelines for the use of innova- in septic shock: from bench to bedside. Intensive Care
tive therapies in sepsis. Crit Care Med. 1992;20(6):864–74. Med. 2010;36(12):2019–29.
14. Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, 32. Lopez A, Lorente JA, Steingrub J, Bakker J, McLuckie A,
Cook D, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS In- Willatts S, et al. Multiple-center, randomized, placebo-con-
ternational Sepsis Definitions Conference. Crit Care Med. trolled, double-blind study of the nitric oxide synthase
2003;31(4):1250–6. inhibitor 546C88: effect on survival in patients with septic
15. Howell MD, Donnino M, Clardy P, Talmor D, Shapiro NI. shock. Crit Care Med. 2004;32(1):21–30.
Occult hypoperfusion and mortality in patients with sus- 33. Rabuel C, Mebazaa A. Septic shock: a heart story since the
pected infection. Intensive Care Med. 2007;33(11):1892–9. 1960s. Intensive Care Med. 2006;32(6):799–807.
16. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli 34. Sergi C, Shen F, Lim DW, Liu W, Zhang M, Chiu B, et al.
M, Ferrer R, et al. Surviving Sepsis Campaign: Internation- Cardiovascular dysfunction in sepsis at the dawn of emerg-
al Guidelines for Management of Sepsis and Septic Shock: ing mediators. Biomed Pharmacother. 2017;95:153–60.
2016. Intensive Care Med. 2017;43(3):304–77. 35. Liu YC, Yu MM, Shou ST, Chai YF. Sepsis-Induced Car-
17. Ferrer R, Martin-Loeches I, Phillips G, Osborn TM, diomyopathy: mechanisms and Treatments. Front Immu-
Townsend S, Dellinger RP, et al. Empiric antibiotic nol. 2017;8:1021.
treatment reduces mortality in severe sepsis and septic 36. Dal-Secco D, DalBo S, Lautherbach NE, Gava FN, Celes
 Caraballo and Jaimes: Sepsis and death 637

MR, Benedet PO, et al. Cardiac hyporesponsiveness in shock. Crit Care Med. 2009;37(5):1670–7.
severe sepsis is associated with nitric oxide-dependent 52. Liu Z, Meng Z, Li Y, Zhao J, Wu S, Gou S, et al. Prognos-
activation of G protein receptor kinase. Am J Physiol Heart tic accuracy of the serum lactate level, the SOFA score and
Circ Physiol. 2017;313(1):H149–63. the qSOFA score for mortality among adults with Sepsis.
37. Sevilla Berrios RA, O’Horo JC, Velagapudi V, Pulido JN. Scand J Trauma Resusc Emerg Med. 2019;27(1):51.
Correlation of left ventricular systolic dysfunction deter- 53. Shankar-Hari M, Phillips GS, Levy ML, Seymour CW, Liu
mined by low ejection fraction and 30-day mortality in VX, Deutschman CS, et al. Developing a New Definition
patients with severe sepsis and septic shock: a systematic and Assessing New Clinical Criteria for Septic Shock: For
review and meta-analysis. J Crit Care. 2014;29(4):495–9. the Third International Consensus Definitions for Sepsis
38. Sanfilippo F, Corredor C, Fletcher N, Landesberg G, Ben- and Septic Shock (Sepsis-3). JAMA. 2016;315(8):775–87.
edetto U, Foex P, et al. Diastolic dysfunction and mortality 54. Londono J, Nino C, Archila A, Valencia M, Cardenas D,
in septic patients: a systematic review and meta-analysis. Perdomo M, et al. Antibiotics has more impact on mortal-
Intensive Care Med. 2015;41(6):1004–13. ity than other early goal-directed therapy components in
39. Vallabhajosyula S, Kumar M, Pandompatam G, Sakhuja A, patients with sepsis: an instrumental variable analysis. J
Kashyap R, Kashani K, et al. Prognostic impact of isolated Crit Care. 2018;48:191–7.
right ventricular dysfunction in sepsis and septic shock: 55. Londono J, Nino C, Diaz J, Morales C, Leon J, Bernal
an 8-year historical cohort study. Ann Intensive Care. E, et al. Association of Clinical Hypoperfusion Variables
2017;7(1):94. With Lactate Clearance and Hospital Mortality. Shock.
40. de Castilho FM, Ribeiro AL, Nobre V, Barros G, de Sousa 2018;50(3):286–92.
MR. Heart rate variability as predictor of mortality in sep- 56. Hernández G, Ospina-Tascón GA, Damiani LP, Estensso-
sis: A systematic review. PLoS One. 2018;13(9):e0203487. ro E, Dubin A, Hurtado J, et al. Effect of a Resuscitation
41. de Castilho FM, Ribeiro AL, da Silva JL, Nobre V, de Strategy Targeting Peripheral Perfusion Status vs Serum
Sousa MR. Heart rate variability as predictor of mor- Lactate Levels on 28-Day Mortality Among Patients With
tality in sepsis: A prospective cohort study. PLoS One. Septic Shock: The ANDROMEDA-SHOCK Randomized
2017;12(6):e0180060. Clinical TrialEffect on Septic Shock Mortality of Resus-
42. Frencken JF, Donker DW, Spitoni C, Koster-Brouwer ME, citation Targeting Peripheral Perfusion vs Serum Lactate
Soliman IW, Ong DS, et al. Myocardial Injury in Patients LevelsEffect on Septic Shock Mortality of Resuscitation
With Sepsis and Its Association With Long-Term Outcome. Targeting Peripheral Perfusion vs Serum Lactate Levels.
Circ Cardiovasc Qual Outcomes. 2018;11(2):e004040. JAMA. 2019;321(7):654–64.
43. Shahreyar M, Fahhoum R, Akinseye O, Bhandari S, Dang 57. Kumar A, Roberts D, Wood KE, Light B, Parrillo JE,
G, Khouzam RN. Severe sepsis and cardiac arrhythmias. Sharma S, et al. Duration of hypotension before initiation
Ann Transl Med. 2018;6(1):6. of effective antimicrobial therapy is the critical determi-
44. Wang HE, Moore JX, Donnelly JP, Levitan EB, Safford nant of survival in human septic shock. Crit Care Med.
MM. Risk of Acute Coronary Heart Disease After Sepsis 2006;34(6):1589–96.
Hospitalization in the REasons for Geographic and Racial 58. Early, Goal-Directed Therapy for Septic Shock—
Differences in Stroke (REGARDS) Cohort. Clin Infect Dis. A Patient-Level Meta-Analysis. N Engl J Med.
2017;65(1):29–36. 2017;376(23):2223–34.
45. Jafarzadeh SR, Thomas BS, Warren DK, Gill J, Fraser VJ. 59. Andrews B, Semler MW, Muchemwa L, Kelly P, Lakhi S,
Longitudinal Study of the Effects of Bacteremia and Sepsis Heimburger DC, et al. Effect of an Early Resuscitation Pro-
on 5-year Risk of Cardiovascular Events. Clin Infect Dis. tocol on In-hospital Mortality Among Adults With Sepsis
2016;63(4):495–500. and Hypotension: A Randomized Clinical Trial. JAMA.
46. Patel N, Bajaj NS, Doshi R, Gupta A, Kalra R, Singh 2017;318(13):1233–40.
A, et al. Cardiovascular Events and Hospital Deaths 60. Andrews B, Muchemwa L, Kelly P, Lakhi S, Heimburger
Among Patients With Severe Sepsis. Am J Cardiol. DC, Bernard GR. Simplified severe sepsis protocol: a
2019;123(9):1406–13. randomized controlled trial of modified early goal-directed
47. Ince C. The microcirculation is the motor of sepsis. Crit therapy in Zambia. Crit Care Med. 2014;42(11):2315–24.
Care. 2005;9 Suppl 4:S13–9. 61. Maitland K, Kiguli S, Opoka RO, Engoru C, Olupot-
48. Miranda M, Balarini M, Caixeta D, Bouskela E. Microcir- Olupot P, Akech SO, et al. Mortality after fluid bolus in
culatory dysfunction in sepsis: pathophysiology, clinical African children with severe infection. N Engl J Med.
monitoring, and potential therapies. Am J Physiol Heart 2011;364(26):2483–95.
Circ Physiol. 2016;311(1):H24–35. 62. Silversides JA, Major E, Ferguson AJ, Mann EE, McAuley
49. Fink MP. Bench-to-bedside review: cytopathic hypoxia. DF, Marshall JC, et al. Conservative fluid management or
Crit Care. 2002;6(6):491. deresuscitation for patients with sepsis or acute respira-
50. Brealey D, Brand M, Hargreaves I, Heales S, Land J, tory distress syndrome following the resuscitation phase
Smolenski R, et al. Association between mitochondrial of critical illness: a systematic review and meta-analysis.
dysfunction and severity and outcome of septic shock. Intensive Care Med. 2017;43(2):155–70.
Lancet. 2002;360(9328):219–23. 63. Marik P, Bellomo R. A rational approach to fluid therapy in
51. Mikkelsen ME, Miltiades AN, Gaieski DF, Goyal M, sepsis. Br J Anaesth. 2015;116(3):339–49.
Fuchs BD, Shah CV, et al. Serum lactate is associated with 64. Brown RM, Semler MW. Fluid Management in Sepsis. J
mortality in severe sepsis independent of organ failure and Intensive Care Med. 2019;34(5):364–73.
638 Caraballo and Jaimes: Sepsis and death

65. De Backer D, Biston P, Devriendt J, Madl C, Chochrad critically ill patients: clinical characteristics and outcomes.
D, Aldecoa C, et al. Comparison of dopamine and nor- Clin J Am Soc Nephrol. 2007;2(3):431–9.
epinephrine in the treatment of shock. N Engl J Med. 82. Poukkanen M, Vaara ST, Pettila V, Kaukonen KM,
2010;362(9):779–89. Korhonen AM, Hovilehto S, et al. Acute kidney injury in
66. Scheeren TW, Bakker J, De Backer D, Annane D, Asfar patients with severe sepsis in Finnish Intensive Care Units.
P, Boerma EC, et al. Current use of vasopressors in septic Acta Anaesthesiol Scand. 2013;57(7):863–72.
shock. Ann Intensive Care. 2019;9(1):20. 83. Vincent JL, Sakr Y, Sprung CL, Ranieri VM, Reinhart K,
67. Sacha GL, Lam SW, Duggal A, Torbic H, Bass SN, Welch Gerlach H, et al. Sepsis in European intensive care units:
SC, et al. Predictors of response to fixed-dose vasopressin results of the SOAP study. Crit Care Med. 2006;34(2):344–
in adult patients with septic shock. Ann Intensive Care. 53.
2018;8(1):35. 84. Khwaja A. KDIGO clinical practice guidelines for acute
68. Kasugai D, Nishikimi M, Nishida K, Higashi M, Yama- kidney injury. Nephron Clin Pract. 2012;120(4):c179–84.
moto T, Numaguchi A, et al. Timing of administration of 85. Bagshaw SM, Lapinsky S, Dial S, Arabi Y, Dodek P,
epinephrine predicts the responsiveness to epinephrine in Wood G, et al. Acute kidney injury in septic shock: clinical
norepinephrine-refractory septic shock: a retrospective outcomes and impact of duration of hypotension prior to
study. J Intensive Care. 2019;7:20. initiation of antimicrobial therapy. Intensive Care Med.
69. Chawla LS, Ostermann M, Forni L, Tidmarsh GF. Broad 2009;35(5):871–81.
spectrum vasopressors: a new approach to the initial man- 86. Uchino S, Kellum JA, Bellomo R, Doig GS, Morimatsu
agement of septic shock? Crit Care. 2019;23(1):124. H, Morgera S, et al. Acute Renal Failure in Critically
70. Rubenfeld GD, Caldwell E, Peabody E, Weaver J, Martin Ill PatientsA Multinational, Multicenter Study. JAMA.
DP, Neff M, et al. Incidence and outcomes of acute lung 2005;294(7):813–8.
injury. N Engl J Med. 2005;353(16):1685–93. 87. Lerolle N, Nochy D, Guerot E, Bruneval P, Fagon JY,
71. Fujishima S, Gando S, Daizoh S, Kushimoto S, Ogura H, Diehl JL, et al. Histopathology of septic shock induced
Mayumi T, et al. Infection site is predictive of outcome in acute kidney injury: apoptosis and leukocytic infiltration.
acute lung injury associated with severe sepsis and septic Intensive Care Med. 2010;36(3):471–8.
shock. Respirology. 2016;21(5):898–904. 88. Ma S, Evans RG, Iguchi N, Tare M, Parkington HC,
72. Matthay MA, Ware LB, Zimmerman GA. The acute respi- Bellomo R, et al. Sepsis-induced acute kidney injury:
ratory distress syndrome. J Clin Invest. 2012;122(8):2731– A disease of the microcirculation. Microcirculation.
40. 2019;26(2):e12483.
73. Park I, Kim M, Choe K, Song E, Seo H, Hwang Y, et al. 89. Poston JT, Koyner JL. Sepsis associated acute kidney
Neutrophils disturb pulmonary microcirculation in sep- injury. BMJ. 2019;364:k4891.
sis-induced acute lung injury. Eur Respir J. 2019;53(3). 90. Bellomo R, Kellum JA, Ronco C, Wald R, Martensson J,
74. Evans CE, Zhao YY. Impact of thrombosis on pulmonary Maiden M, et al. Acute kidney injury in sepsis. Intensive
endothelial injury and repair following sepsis. Am J Physi- Care Med. 2017;43(6):816–28.
ol Lung Cell Mol Physiol. 2017;312(4):L441–l51. 91. Semler MW, Wanderer JP, Ehrenfeld JM, Stollings JL, Self
75. Ranieri VM, Rubenfeld GD, Thompson BT, Ferguson ND, WH, Siew ED, et al. Balanced Crystalloids versus Saline in
Caldwell E, Fan E, et al. Acute respiratory distress syn- the Intensive Care Unit. The SALT Randomized Trial. Am
drome: the Berlin Definition. JAMA. 2012;307(23):2526– J Respir Crit Care Med. 2017;195(10):1362–72.
33. 92. Rochwerg B, Alhazzani W, Sindi A, Heels-Ansdell D,
76. Bellani G, Laffey JG, Pham T, Fan E, Brochard L, Esteban Thabane L, Fox-Robichaud A, et al. Fluid resuscitation
A, et al. Epidemiology, Patterns of Care, and Mortal- in sepsis: a systematic review and network meta-analysis.
ity for Patients With Acute Respiratory Distress Syn- Ann Intern Med. 2014;161(5):347–55.
drome in Intensive Care Units in 50 Countries. JAMA. 93. Self WH, Semler MW, Wanderer JP, Wang L, Byrne DW,
2016;315(8):788–800. Collins SP, et al. Balanced Crystalloids versus Saline in
77. Sheu CC, Gong MN, Zhai R, Chen F, Bajwa EK, Noncritically Ill Adults. N Engl J Med. 2018;378(9):819–
Clardy PF, et al. Clinical characteristics and outcomes 28.
of sepsis-related vs non-sepsis-related ARDS. Chest. 94. Shaw AD, Raghunathan K, Peyerl FW, Munson SH,
2010;138(3):559–67. Paluszkiewicz SM, Schermer CR. Association between in-
78. Zampieri FG, Mazza B. Mechanical Ventilation in Sepsis: travenous chloride load during resuscitation and in-hospital
A Reappraisal. Shock. 2017;47(1S Suppl 1):41-6. mortality among patients with SIRS. Intensive Care Med.
79. Fan E, Brodie D, Slutsky AS. Acute Respiratory Distress 2014;40(12):1897–905.
Syndrome: Advances in Diagnosis and Treatment. JAMA. 95. Semler MW, Self WH, Wanderer JP, Ehrenfeld JM, Wang
2018;319(7):698–710. L, Byrne DW, et al. Balanced Crystalloids versus Saline in
80. Tongyoo S, Permpikul C, Mongkolpun W, Vattanavanit V, Critically Ill Adults. N Engl J Med. 2018;378(9):829–39.
Udompanturak S, Kocak M, et al. Hydrocortisone treat- 96. Levi M, Van Der Poll T. Thrombomodulin in sepsis. Min-
ment in early sepsis-associated acute respiratory distress erva Anestesiol. 2013;79(3):294–8.
syndrome: results of a randomized controlled trial. Crit 97. Simmons J, Pittet JF. The coagulopathy of acute sepsis.
Care. 2016;20(1):329. Curr Opin Anaesthesiol. 2015;28(2):227–36.
81. Bagshaw SM, Uchino S, Bellomo R, Morimatsu H, 98. McDonald B, Davis RP, Kim SJ, Tse M, Esmon CT,
Morgera S, Schetz M, et al. Septic acute kidney injury in Kolaczkowska E, et al. Platelets and neutrophil extracellu-
 Caraballo and Jaimes: Sepsis and death 639

lar traps collaborate to promote intravascular coagulation 114. Vincent JL, Angus DC, Artigas A, Kalil A, Basson BR,
during sepsis in mice. Blood. 2017;129(10):1357–67. Jamal HH, et al. Effects of drotrecogin alfa (activated) on
99. Aird WC, editor. The hematologic system as a marker of organ dysfunction in the PROWESS trial. Crit Care Med.
organ dysfunction in sepsis. Mayo Clinic Proceedings. 2003;31(3):834–40.
Elsevier; 2003. 115. Waseem N, Chen PH. Hypoxic Hepatitis: A Review and
100. Saito S, Uchino S, Hayakawa M, Yamakawa K, Kudo D, Clinical Update. J Clin Transl Hepatol. 2016;4(3):263–8.
Iizuka Y, et al. Epidemiology of disseminated intravascular 116. Nesseler N, Launey Y, Aninat C, Morel F, Malledant Y,
coagulation in sepsis and validation of scoring systems. J Seguin P. Clinical review: the liver in sepsis. Crit Care.
Crit Care. 2019;50:23–30. 2012;16(5):235.
101. Sharma B, Sharma M, Majumder M, Steier W, Sangal A, 117. Van den Broecke A, Van Coile L, Decruyenaere A, Col-
Kalawar M. Thrombocytopenia in septic shock patients—a paert K, Benoit D, Van Vlierberghe H, et al. Epidemiology,
prospective observational study of incidence, risk factors causes, evolution and outcome in a single-center cohort
and correlation with clinical outcome. Anaesth Intensive of 1116 critically ill patients with hypoxic hepatitis. Ann
Care. 2007;35(6):874–80. Intensive Care. 2018;8(1):15.
102. Azkárate I, Choperena G, Salas E, Sebastián R, Lara G, 118. Jenniskens M, Langouche L, Vanwijngaerden YM,
Elósegui I, et al. Epidemiology and prognostic factors in Mesotten D, Van den Berghe G. Cholestatic liver (dys)
severe sepsis/septic shock. Evolution over six years [En- function during sepsis and other critical illnesses. Intensive
glish Edition]. Med Intensiva. 2016;40(1):18–25. Care Med. 2016;42(1):16–27.
103. Burunsuzoglu B, Salturk C, Karakurt Z, Ongel EA, Takir 119. Spirli C, Fabris L, Duner E, Fiorotto R, Ballardini
HB, Kargin F, et al. Thrombocytopenia: A Risk Factor of G, Roskams T, et al. Cytokine-stimulated nitric oxide
Mortality for Patients with Sepsis in the Intensive Care production inhibits adenylyl cyclase and cAMP-de-
Unit. Turk Thorac J. 2016;17(1):7–14. pendent secretion in cholangiocytes. Gastroenterology.
104. Venkata C, Kashyap R, Farmer JC, Afessa B. Thrombocy- 2003;124(3):737–53.
topenia in adult patients with sepsis: incidence, risk factors, 120. Spirli C, Nathanson MH, Fiorotto R, Duner E, Denson
and its association with clinical outcome. J Intensive Care. LA, Sanz JM, et al. Proinflammatory cytokines inhibit
2013;1(1):9. secretion in rat bile duct epithelium. Gastroenterology.
105. Seki Y, Wada H, Kawasugi K, Okamoto K, Uchiyama T, 2001;121(1):156–69.
Kushimoto S, et al. A prospective analysis of disseminated 121. Gofton TE, Young GB. Sepsis-associated encephalopathy.
intravascular coagulation in patients with infections. Intern Nat Rev Neurol. 2012;8(10):557–66.
Med. 2013;52(17):1893–8. 122. Sharshar T, Annane D, de la Grandmaison GL, Brouland
106. Scully M, Levi M. How we manage haemostasis during JP, Hopkinson NS, Francoise G. The neuropathology of
sepsis. Br J Haematol. 2019;185(2):209–18. septic shock. Brain Pathol. 2004;14(1):21–33.
107. Vincent J-L, Francois B, Zabolotskikh I, Daga MK, 123. Walkey AJ, Hammill BG, Curtis LH, Benjamin EJ. Long-
Lascarrou J-B, Kirov MY, et al. Effect of a Recombinant term outcomes following development of new-onset atrial
Human Soluble Thrombomodulin on Mortality in Patients fibrillation during sepsis. Chest. 2014;146(5):1187–95.
With Sepsis-Associated Coagulopathy: The SCARLET 124. Boehme AK, Ranawat P, Luna J, Kamel H, Elkind MS.
Randomized Clinical TrialEffect of Thrombomodulin on Risk of Acute Stroke After Hospitalization for Sepsis.
Mortality in Patients With Sepsis-Associated Coagulop- Stroke. 2017;48(3):574–80.
athyEffect of Thrombomodulin on Mortality in Patients 125. Shao IY, Elkind MS, Boehme AK. Risk Factors for Stroke
With Sepsis-Associated Coagulopathy. JAMA. 2019. in Patients With Sepsis and Bloodstream Infections. Stroke.
108. Ranieri VM, Thompson BT, Barie PS, Dhainaut JF, 2019;50(5):1046–51.
Douglas IS, Finfer S, et al. Drotrecogin Alfa (Acti- 126. Polito A, Eischwald F, Maho AL, Polito A, Azabou E,
vated) in Adults with Septic Shock. N Engl J Med. Annane D, et al. Pattern of brain injury in the acute setting
2012;366(22):2055–64. of human septic shock. Crit Care. 2013;17(5):R204.
109. Jaimes F, De La Rosa G, Morales C, Fortich F, Arango 127. Sweis R, Ortiz J, Biller J. Neurology of Sepsis. Curr
C, Aguirre D, et al. Unfractioned heparin for treatment of Neurol Neurosci Rep. 2016;16(3):21.
sepsis: A randomized clinical trial (The HETRASE Study). 128. Siami S, Annane D, Sharshar T. The encephalopathy in
Crit Care Med. 2009;37(4):1185–96. sepsis [viii.]. Crit Care Clin. 2008;24(1):67–82.
110. Su GL. Lipopolysaccharides in liver injury: molecular 129. Annane D, Sharshar T. Cognitive decline after sepsis.
mechanisms of Kupffer cell activation. Am J Physiol Gas- Lancet Respir Med. 2015;3(1):61–9.
trointest Liver Physiol. 2002;283(2):G256–65. 130. van Gool WA, van de Beek D, Eikelenboom P. Systemic
111. Yan J, Li S, Li S. The role of the liver in sepsis. Int Rev infection and delirium: when cytokines and acetylcholine
Immunol. 2014;33(6):498–510. collide. Lancet. 2010;375(9716):773–5.
112. Strnad P, Tacke F, Koch A, Trautwein C. Liver - guardian, 131. Sharshar T, Gray F, Lorin de la Grandmaison G, Hopkin-
modifier and target of sepsis. Nat Rev Gastroenterol Hepa- son NS, Ross E, Dorandeu A, et al. Apoptosis of neurons
tol. 2017;14(1):55–66. in cardiovascular autonomic centres triggered by inducible
113. Brun-Buisson C, Meshaka P, Pinton P, Vallet B. EPISEP- nitric oxide synthase after death from septic shock. Lancet.
SIS: a reappraisal of the epidemiology and outcome of 2003;362(9398):1799–805.
severe sepsis in French intensive care units. Intensive Care 132. Vincent JL, Moreno R, Takala J, Willatts S, De Mendon-
Med. 2004;30(4):580–8. ca A, Bruining H, et al. The SOFA (Sepsis-related Organ
640 Caraballo and Jaimes: Sepsis and death

Failure Assessment) score to describe organ dysfunction/ Model for the Early Identification of Patients at Risk for
failure. On behalf of the Working Group on Sepsis-Related Sepsis. Ann Emerg Med. 2019;73(4):334–44.
Problems of the European Society of Intensive Care Medi- 146. Jawad I, Luksic I, Rafnsson SB. Assessing available
cine. Intensive Care Med. 1996;22(7):707–10. information on the burden of sepsis: global estimates
133. de Grooth HJ, Geenen IL, Girbes AR, Vincent JL, Parienti of incidence, prevalence and mortality. J Glob Health.
JJ, Oudemans-van Straaten HM. SOFA and mortality end- 2012;2(1):010404.
points in randomized controlled trials: a systematic review 147. Rudd KE, Kissoon N, Limmathurotsakul D, Bory S,
and meta-regression analysis. Crit Care. 2017;21(1):38. Mutahunga B, Seymour CW, et al. The global burden
134. Raith EP, Udy AA, Bailey M, McGloughlin S, MacIsaac of sepsis: barriers and potential solutions. Crit Care.
C, Bellomo R, et al. Prognostic Accuracy of the SOFA 2018;22(1):232.
Score, SIRS Criteria, and qSOFA Score for In-Hospital 148. Instituto Latinoamericano de Sepsis and Global Sepsis
Mortality Among Adults With Suspected Infection Admit- Alliance. São Paulo Declaration [internet]. Cited 2019 May
ted to the Intensive Care Unit. JAMA. 2017;317(3):290– 17. Available from: https://ilas.org.br/see-declaration.php
300. 149. African Sepsis Alliance. Kampala Declaration 2017 [in-
135. Minne L, Abu-Hanna A, de Jonge E. Evaluation of ternet]. Cited 2019 May 17. Available from: https://www.
SOFA-based models for predicting mortality in the ICU: A africansepsisalliance.org/kampaladeclaration.
systematic review. Crit Care. 2008;12(6):R161.
136. Khwannimit B, Bhurayanontachai R, Vattanavanit V.
Comparison of the accuracy of three early warning scores
with SOFA score for predicting mortality in adult sepsis
and septic shock patients admitted to intensive care unit.
Heart Lung. 2019;48(3):240–4.
137. Seymour CW, Liu VX, Iwashyna TJ, Brunkhorst FM,
Rea TD, Scherag A, et al. Assessment of Clinical Criteria
for Sepsis: For the Third International Consensus Defini-
tions for Sepsis and Septic Shock (Sepsis-3)Assessment of
Clinical Criteria for SepsisAssessment of Clinical Criteria
for Sepsis. JAMA. 2016;315(8):762–74.
138. Yadav H, Harrison AM, Hanson AC, Gajic O, Kor DJ,
Cartin-Ceba R. Improving the Accuracy of Cardiovascular
Component of the Sequential Organ Failure Assessment
Score. Crit Care Med. 2015;43(7):1449–57.
139. Levy MM, Rhodes A, Phillips GS, Townsend SR, Schorr
CA, Beale R, et al. Surviving Sepsis Campaign: association
between performance metrics and outcomes in a 7.5-year
study. Crit Care Med. 2015;43(1):3–12.
140. Subbe CP, Kruger M, Rutherford P, Gemmel L. Validation
of a modified Early Warning Score in medical admissions.
QJM. 2001;94(10):521–6.
141. Smith GB, Prytherch DR, Meredith P, Schmidt PE, Feath-
erstone PI. The ability of the National Early Warning Score
(NEWS) to discriminate patients at risk of early cardiac
arrest, unanticipated intensive care unit admission, and
death. Resuscitation. 2013;84(4):465–70.
142. Goulden R, Hoyle MC, Monis J, Railton D, Riley V,
Martin P, et al. qSOFA, SIRS and NEWS for predicting in-
hospital mortality and ICU admission in emergency admis-
sions treated as sepsis. Emerg Med J. 2018;35(6):345–9.
143. Churpek MM, Snyder A, Han X, Sokol S, Pettit N, How-
ell MD, et al. Quick Sepsis-related Organ Failure Assess-
ment, Systemic Inflammatory Response Syndrome, and
Early Warning Scores for Detecting Clinical Deterioration
in Infected Patients outside the Intensive Care Unit. Am J
Respir Crit Care Med. 2017;195(7):906–11.
144. Seymour CW, Kennedy JN, Wang S, Chang CH, Elliott
CF, Xu Z, et al. Derivation, Validation, and Potential Treat-
ment Implications of Novel Clinical Phenotypes for Sepsis.
JAMA. 2019;321(20):2003–2017.
145. Delahanty RJ, Alvarez J, Flynn LM, Sherwin RL, Jones
SS. Development and Evaluation of a Machine Learning