84 Sepsis and Septic Shock

Trisha N. Branan, Susan E. Smith, and

Christopher M. Bland

LEARNING OBJECTIVES
Upon completion of the chapter, the reader will be able to:
1. Compare and contrast the definitions of syndromes related to sepsis.
2. Identify the pathogens associated with sepsis.
3. Discuss the pathophysiology of sepsis as it relates to systemic inflammation, coagulation, and tissue
hypoperfusion.
4. Identify patient symptoms as early or late sepsis and evaluate diagnostic and laboratory tests for patient
treatment and monitoring.
5. Assess complications of sepsis syndromes and discuss their impact on patient outcomes.
6. Design desired treatment outcomes for septic patients.
7. Formulate a treatment and monitoring plan (pharmacologic and nonpharmacologic) for septic patients.
8. Evaluate patient response and devise alternative treatment regimens for nonresponding septic patients.

INTRODUCTION intra-abdominal and genitourinary infections.2 In approximately

S
one-third of sepsis cases, a pathogen is not identified making
epsis occurs across a continuum of physiologic
deescalation from broad spectrum to a more narrowed antimi-
stages in response to infection, which manifests
crobial regimen difficult.2 A recent study demonstrated that diar-
as systemic inflammation, coagulation, and tis-
rheal illness is the most common etiology of sepsis globally, with
sue hypoperfusion, potentially leading to organ dysfunction.1
patterns of sepsis incidence varying significantly by geographical
The Third International Consensus Definitions for Sepsis and
location, highlighting the need for knowledge of local patterns to
Septic Shock define sepsis as a life-threatening organ dysfunction
determine the overall risk by an organism.11
caused by a dysregulated host response represented by an increase
Gram-positive and Gram-negative bacteria, fungal
of at least 2 points in the Sequential (Sepsis-related) Organ Fail-
species, and viruses may cause sepsis (Table 84–3). Gram-positive
ure Assessment (SOFA) score (Table 84–1). Septic shock is
infections account for 30% to 50% of sepsis and septic shock
defined as a subset of patients with further increases in mortality
cases.5,7,8 The percentages of Gram-negative, polymicrobial, and
resulting from underlying circulatory, cellular, and/or metabolic
viral sepsis cases are 25%, 25%, and 4%, respectively.5,7,8,12 A mul-
abnormalities. These patients require the use of vasopressor sup-
tinational study of 14,000 critically ill patients showed an increase
port in addition to adequate volume resuscitation to maintain a
in Gram-negative bacterial causes of infection compared to
mean arterial pressure (MAP) of at least 65 mm Hg and have an
Gram-positive bacterial and fungal causes.2 Multidrug-resistant
elevated serum lactate level greater than 2 mmol/L. Adult patients
(MDR) bacteria are responsible for approximately 25% of sepsis
with suspected sepsis may be screened using a bedside clinical
cases, are difficult to treat due to fewer antimicrobial options, and
scoring system known as the quickSOFA (qSOFA) to rapidly
increase mortality.7,8 The rate of fungal infections has significantly
identify patients who may need further diagnostic workup and
increased with Candida albicans as the most common fungal spe-
intervention (Table 84–2).2,3
cies identified; however, nonalbicans species (Candida glabrata,
Candida krusei, and Candida tropicalis) have increased from 24%
EPIDEMIOLOGY AND ETIOLOGY to 46%.5,13,14 Other fungi identified as causes of sepsis include spe-
Sepsis is the leading cause of morbidity and mortality for critically cies of Cryptococcus, Coccidioides, Fusarium, and Aspergillus.
ill patients and the 10th leading cause of death overall.4,5 Mortality
rates remain high for patients with sepsis and septic shock, with
septic shock and multiorgan failure as the most common causes of PATHOPHYSIOLOGY
death.1 Annually, there are approximately 1.7 million adults who The development of sepsis is complex and multifactorial. The nor-
develop sepsis in the United States with nearly 270,000 deaths.6 mal host response to infection is designed to localize and control
Risk factors for developing sepsis include increased age, cancer, microbial invasion and initiate repair of injured tissue through
immunodeficiency, chronic organ failure, genetic factors (male phagocytic cells and inflammatory mediators.4 Sepsis
gender and non-White ethnic origin in North America), and the results when the interplay between the host’s immune, inflamma-
presence of bacteremia.5,7-10 Pulmonary infections cause approxi- tory, and coagulant responses becomes exaggerated, extending to
mately half of all sepsis cases within the United States, followed by normal tissue distant from the initial tissue site.
1321

Chisholm_Ch084_p1321-1332.indd 1321 11/10/21 4:37 PM

 1322  SECTION 15 | DISEASES OF INFECTIOUS ORIGIN

Table 84–1
The SOFA Score

SOFA Score 1 2 3 4
Respiration
PaO2/FiO2, mm Hg < 400 < 300 < 200 < 100
___with respiratory support___
Coagulation
Platelets × 103/mm3 < 150 < 100 < 50 < 20
Liver
Bilirubin, mg/dL (μmol/L) 1.2–1.9 (20–32) 2.0–5.9 (33–101) 6.0–11.9 (102–204) ≥ 12.0 (> 204)
Cardiovascular
Hypotension MAPa < 70 mm Hg Dopamine ≤ 5 Dopamine > 5–15, Dopamine > 15,
or dobutamine epinephrine ≤ 0.1, or epinephrine > 0.1, or
(any dose)b norepinephrine ≤ 0.1 norepinephrine > 0.1
Central nervous system
Glasgow Coma Score 13–14 10–12 6–9 <6
Renal
Creatinine, mg/dL (μmol/L) or 1.2–1.9 (110–170) 2.0–3.4 (171–299) 3.5–4.9 (300–400) ≥ 5.0 (> 440)
urine output or < 500 mL/day or < 200 mL/day
a
MAP (mean arterial pressure) = 1/3 (SBP − DBP) + DBP.
b
Adrenergic agents administered for at least 1 hour (doses given are in mcg/kg/min).
Reproduced, with permission, from Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe
organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine.
Intensive Care Med. 1996;22(7):707–710.

Pro- and Anti-inflammatory Mediators neutrophil presence is associated with more severe disease and
The key factor in the development of sepsis is inflammation, is more likely to result in the formation of neutrophil extracel-
which is intended to be a local and contained response to infec- lular traps leading to decreased functional capacity. Ultimately,
tion or injury. Infection or injury is controlled through proin- this increased number of band neutrophils over time can lead to
flammatory and anti-inflammatory mediators. Proinflammatory endothelial tissue damage and hypercoagulation.18
mediators facilitate clearance of the injuring stimulus, promote
resolution of injury, and are involved in processing of damaged CLINICAL PRESENTATION AND DIAGNOSIS
tissue.4,14-17 To control the intensity and duration of the inflamma- The clinical presentation of sepsis varies, and the rate of develop-
tory response, anti-inflammatory mediators are released that act ment of clinical manifestations may differ from patient to patient.
to regulate proinflammatory mediators.16,17 The balance between A physical examination should be performed rapidly and effi-
pro- and anti-inflammatory mediators localizes infection/injury ciently when sepsis is suspected, with efforts directed toward
of host tissue.14-17 However, systemic responses ensue when equi- uncovering the most likely cause of sepsis. The patient may not
librium in the inflammatory process is lost. provide any medical history; therefore, historical data may be
The inflammatory process in sepsis is linked to the coagula- obtained from medical records and/or family. The patient’s medi-
tion system. Proinflammatory mediators may have procoagulant cal condition, recent illnesses, infections, or activities may pro-
and antifibrinolytic effects, whereas anti-inflammatory mediators vide valuable information about the cause of sepsis.
may have fibrinolytic effects. A key factor in the inflammation of
sepsis is activated protein C that enhances fibrinolysis and inhib-
its inflammation. Protein C levels are decreased in many septic
patients.
Table 84–3
The initial immune response to a severe infectious syndrome Pathogens in Sepsis
like sepsis is associated with increased neutrophil production
and the release of both immature (eg, bands) and mature forms Organism Frequency (%)
of neutrophils into the bloodstream. This increased immature Gram-positive bacteria 50–55
Gram-negative bacteria 35–40
Fungi 5
Table 84–2 Polymicrobial 5
Anaerobes (alone) 1
Quick SOFA (qSOFA) Criteria Parasites Variable
•• Respiratory rate ≥ 22 breaths/min Viruses Variable
•• Altered mentation
•• Systolic blood pressure ≤ 100 mm Hg Data from Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology
•• Positive qSOFA score indicated by meeting ≥ 2 criteria of sepsis in the United States from 1979 through 2000. N Engl J Med.
2003;348(16):1546–1554.

Chisholm_Ch084_p1321-1332.indd 1322 11/10/21 4:37 PM

 CHAPTER 84 | SEPSIS AND SEPTIC SHOCK  1323

However, not all healthcare systems have this technology avail-

Possible Physical Examination Results able. Additionally, if these technologies are used, they must be
in Sepsis in conjunction with antimicrobial stewardship personnel to be
cost-effective.19
HEENT: Scleral icterus, dry mucous membranes, pinpoint At least two sets of blood cultures should be obtained to rule
pupils, dilated and fixed pupils, nystagmus out contamination, with at least one set drawn percutaneously
Neck: Jugular venous distention, carotid bruits and one set drawn through each vascular access device present
Lungs: Crackles (rales), consolidation, egophony, absent for greater than 48 hours.
breath sounds Cultures of urine (with urinalysis), respiratory secretions, cere-
brospinal fluid, and wounds should be obtained if the clinical
CV: Irregular rhythm, S3 gallop, murmurs presentation suggests infection of these specific fluids, tissues, or
Abd: Tense, distended, tender, rebound, guarding, organs. Laboratory tests should be performed to evaluate infection
hepatosplenomegaly or complications of sepsis, including complete blood count (CBC)
Rectal: Decreased tone with differential, coagulation parameters, comprehensive meta-
bolic panel (CMP), serum lactate concentration, arterial blood gas
Exts: Swollen calf, disparity of blood pressure between upper (ABG), and appropriate diagnostic radiographic imaging studies.
extremities The use of biomarkers of sepsis, such as endotoxin and pro-
Neurologic: Agitation, confusion, delirium, obtundation, calcitonin, has been controversial. Measurement of endotoxin,
coma procalcitonin, or other markers in blood or serum is not rou-
Skin: Cold, clammy, or warm; hyperemic skin; rashes tinely recommended. Concentrations of procalcitonin in serum
are usually increased in sepsis but fail to differentiate between
infection and inflammation. However, procalcitonin has a high
negative predictive value and may allow for the discontinuation
Diagnostic and Laboratory Tests of antibiotics. It is important to note that the measurement of any
Microbiologic cultures should be obtained before antimicrobial biomarker, including procalcitonin, should be used in conjunc-
therapy is initiated as long as this does not significantly delay the tion with the patient’s overall clinical assessment and never be
start of therapy. However, cultures take 6 to 72 hours for results used as the sole indicator for altering microbial therapy.
to be completed and often are negative (no growth of bacterial
organisms). Negative cultures do not rule out the presence of Complications of Sepsis
infection. Administering antimicrobials before obtaining cultures Recognition and treatment of sepsis complications,
may lead to a false-negative culture. Rapid diagnostics of blood particularly organ failure, is essential to improve outcomes. The
cultures allows for identification of specific pathogens within as cumulative burden of sepsis complications is the leading factor of
little as 20 minutes to 3 hours after initial growth is identified. mortality. The risk of death increases 20% with the failure of each

Clinical Presentation and Diagnosis of Sepsis

The signs and symptoms of septic patients are referred to as and may lead to respiratory alkalosis. Disorientation and
early and late sepsis. confusion may develop early in septic patients, particularly
in the elderly and patients with preexisting neurologic
Signs and Symptoms
impairment. Disorientation and confusion may be related
The initial clinical signs and symptoms represent early sepsis, to the infection or due to sepsis signs and symptoms (eg,
and they include fever, chills, and change in mental status. hypoxia).
Other signs and symptoms include the following:
Late sepsis represents a slow process that develops over
•• Tachycardia several hours of hypoperfusion. Signs and symptoms of late
•• Tachypnea sepsis include the following:
•• Nausea and vomiting •• Lactic acidosis
•• Hyperglycemia •• Oliguria
•• Myalgias •• Leukopenia
•• Lethargy and malaise •• Thrombocytopenia
•• Proteinuria •• Myocardial depression
•• Leukocytosis •• Pulmonary edema
•• Hypoxia •• Hypotension
•• Hyperbilirubinemia •• Hypoglycemia
Septic patients may have an elevated, low, or normal •• Gastrointestinal hemorrhage
temperature. The absence of fever is common in neonates Oliguria often follows hypotension because of decreased renal
and elderly patients. Hypothermia is associated with a poor perfusion. Metabolic acidosis ensues because of diminished
prognosis. Hyperventilation may occur before fever and chills clearance of lactate by the kidneys and liver.

Chisholm_Ch084_p1321-1332.indd 1323 11/10/21 4:37 PM

 1324  SECTION 15 | DISEASES OF INFECTIOUS ORIGIN

Treatment for sepsis focuses on infection, inflammation, hypo-

Patient Encounter Part 1 perfusion, and widespread tissue injury. Septic patients may
require multiple simultaneous treatment regimens to achieve
A 58-year-old man with a history of hypertension, obesity,
desired outcomes of decreased morbidity and mortality.
and COPD presents to the emergency department with
complaints of fever, cough, and increased sputum production Initial Resuscitation
for the last 2 days. The patient is becoming increasingly short
A landmark study of early goal–directed therapy (EGDT) using
of breath and confused.
a standardized protocol that required the use of a special cath-
What information is suggestive of infection and/or sepsis? eter for central venous oxygen saturation monitoring decreased
What information do we need in order to confirm or diagnose 28-day mortality in septic patients by approximately 16%.25 Three
sepsis in this patient? subsequent randomized controlled studies comparing EGDT to
groups of patients receiving contemporary care (with or without
the use of protocols) found no differences in mortality.27 These
results demonstrate that continued focus on early recognition and
additional organ. The most common complications are respira- treatment of these patients may play a more important role than
tory and cardiovascular compromise, usually seen as acute respi- protocol-based therapy; however, the use of EGDT was not asso-
ratory distress syndrome (ARDS), hemodynamic compromise, ciated with harm.
and elevated serum lactate levels. Other complications include Upon recognition of sepsis, resuscitation should begin imme-
altered mentation, acute kidney injury (AKI) that may require diately. Within the first 3 hours, at least 30 mL/kg of crystalloid
renal replacement therapy, paralytic ileus, disseminated intravas- fluid should be administered intravenously. Subsequent fluid
cular coagulation (DIC), and adrenal insufficiency.2 administration should be guided by frequent reassessment of
hemodynamic status and dynamic variables to predict fluid
responsiveness and prevent volume overload. The use of static
TREATMENT AND OUTCOME EVALUATION variables, such as central venous pressure (CVP), to assess the
Desired Outcomes need for additional fluid administration should not be used
The primary treatment goal of sepsis is to prevent mor- alone. Rather, these variables in conjunction with dynamic mea-
bidity and mortality through rapid recognition and intervention. sures, such as the passive leg raise technique, fluid challenges, and
Treatment is aimed at early implementation of therapies, such as stroke volume or pulse pressure variation, may be better predic-
fluid resuscitation and antimicrobials, reducing or eliminating tors of fluid responsiveness.22,25,26,28
organ dysfunction, eliminating the source of infection, avoid- Emerging noninvasive techniques, such as the use of cardiac
ing adverse reactions of treatment, and providing cost-effective ultrasound, have recently shown reliability in assessing intra-
therapy.20-26 vascular volume status through measuring inferior vena cava
diameter changes and collapsibility.29,30 Resuscitation should also
General Approach to Treatment target the normalization of blood lactate levels in patients with
Once a diagnosis of sepsis has been made, rapid intervention an initially elevated blood lactate as a marker of improved tissue
with appropriate therapies is crucial to decrease morbidity and perfusion.22
mortality.20,22
In addition to appropriately timed diagnostic tests, pertinent Fluid Therapy
approaches in the management of septic patients are as follows Crystalloid fluids (such as 0.9% sodium chloride or lactated
(Figure 84–1)26: Ringer solutions) or colloids (albumin products) are used for
1. Prompt recognition of the septic patient and early resuscitation, and clinical studies comparing these two fluids
implementation of therapies. types have found them to be equivalent.28,31,32 Crystalloids require
more fluid volume, which may lead to more edema (utilize cau-
2. Fluid therapy, using crystalloids primarily, to restore tion in patients at risk of fluid overload, eg, congestive heart fail-
intravascular volume depletion. ure, renal failure, and ARDS); however, albumin is significantly
3. Early administration of broad-spectrum antimicrobial more expensive. Hydroxyethyl starch (HES), another type of col-
therapy. loid, should not be used due to studies demonstrating increased
4. Vasopressor therapy, using norepinephrine initially, to morbidity and mortality.33,34 For these reasons, crystalloids are
maintain hemodynamic stability (on average an MAP preferred versus colloids for initial resuscitation, except in cases
of 65 mm Hg [8.6 kPa]) in patients with septic shock where large amounts of crystalloids are needed, and hypervol-
refractory to fluid resuscitation. emia may be harmful to the patient.22
After initial fluid administration, there remains clinical con-
5. Intravenous (IV) hydrocortisone may be considered for
troversy surrounding the appropriate type of fluid for contin-
patients who remain hemodynamically unstable despite
ued resuscitation. There are data that suggest potential benefit
adequate fluid resuscitation and vasopressor support.
of using balanced crystalloid solutions or a chloride-restrictive
6. Glycemic control via infusion of regular insulin to maintain strategy. Large (n > 29,000), prospective, multiple-crossover trials
glucose levels between 140 and 180 mg/dL (7.8 and suggest the administration of large volumes of isotonic saline may
10.0 mmol/L). be associated with increased incidences of AKI and renal replace-
7. Adjunctive therapies: blood product administration, ment therapy secondary to hyperchloremic metabolic acidosis.35,36
analgesia, sedation, neuromuscular blockade, renal
replacement therapy, sodium bicarbonate therapy, venous Anti-infective Therapy
thromboembolism (VTE) prophylaxis, stress ulcer Appropriate empiric antimicrobial therapy
prophylaxis, and nutrition. decreases 28-day mortality compared with inappropriate

Chisholm_Ch084_p1321-1332.indd 1324 11/10/21 4:37 PM

 CHAPTER 84 | SEPSIS AND SEPTIC SHOCK  1325

Sepsis

Early recognition and

Administer broad-
intervention—initial
spectrum anti-
hemodynamic Source control
infective
resuscitation
Determine likely site of
infection and causative
pathogens, and patient
Fluids parameters that affect
(Crystalloid) anti-infective empiric
choice Insulin to maintain
Refractory glucose
Continued 140–180 mg/dL
Nonrefractory hypotension— (7.8–10.0 mmol/L)
vasopressors

Nonrefractory Refractory

Continue current Consider Continually monitor Potential adjunctive therapy

management—wean low-dose anti-infectives • Analgesia and sedation
off medications steroids Review: C&S, anti- • VTE prophylaxis
when appropriate infective dose, and • Stress ulcer prophylaxis
via clinical response frequency • Nutrition

Wean off steroid therapy

once vasopressors are
no longer required

Initiate change when:

• Resistant organisms
• Patient not improving
• Step-down therapy
appropriate

Bold = improved morbidity and mortality data

FIGURE 84–1. Therapeutic approach to sepsis. (C&S, cultures and susceptibilities.)

empiric therapy (24% vs 39%).20,21,37 Additionally, appropriate cognizant of growing prevalence of bacterial resistance in
therapy administered within 1 hour of sepsis recognition also community and healthcare settings.
decreases complications and mortality.20,21,37 Empiric antimi- •• Patient history (underlying disease, previous cultures or
crobial therapy should include multiple agents for most cases, infections, including any recent antibiotic therapy, and drug
depending on the likely site of infection and causative patho- allergy/intolerance).
gens. Anti-infective clinical trials in sepsis and septic shock
•• Adverse reactions.
patients are scarce and have not demonstrated differences
among agents; therefore, factors that determine selection are •• Cost.
as follows: Anti-infective regimens should be broad spectrum since delays
•• Site/source of infection. in the appropriate therapy result in increased mortality. In the sub-
set of patients with septic shock, empiric therapy should include
•• Causative pathogens.
combination regimens with multiple antibiotics from different
•• Community- or nosocomial-acquired infection. mechanistic classes (eg, two antipseudomonal antimicrobials in
•• Immune status of patient. patients with known risk factors to ensure appropriate coverage of
•• Antibiotic susceptibility and resistance profile for the MDR organisms). Please reference Chapter 69 for more antimi-
institution and local community. Clinicians should be crobial information and other chapters based on specific disease
state leading to sepsis.

Chisholm_Ch084_p1321-1332.indd 1325 11/10/21 4:37 PM

 1326  SECTION 15 | DISEASES OF INFECTIOUS ORIGIN

Patient Encounter Part 2: Medical History, Physical Examination, and Diagnostic Tests
PMH: Hypertension, obesity, COPD VS: BP 86/48 mm Hg, P 132 beats/min, RR 34 breaths/min,
FH: Father has a history of hypertension, hyperlipidemia, and T 39.2°C (102.6°F)
chronic kidney disease; mother had a history of hypertension, Neurologic: Obtunded, oriented to person only
diabetes mellitus, hyperlipidemia, coronary artery disease, died Skin: Cool, clammy
from myocardial infarction at age 68
Labs: Serum creatinine 2.2 mg/dL (194.5 μmol/L); glucose 201
SH: High school teacher; previous smoker (one pack per day, mg/dL (11.2 mmol/L); white blood cells: leukocytosis (18,300/
quit 1 year ago); rare alcohol intake mm3 [18.3 × 109/L]) with left shift; lactic acid 4.2 mEq/L (4.2
Allergies: NKDA mmol/L)
Meds: Lisinopril 20 mg by mouth once daily; Cultures: Blood and sputum cultures pending
hydrochlorothiazide 25 mg by mouth once daily; tiotropium Radiology: Chest X-ray consistent with left lower lobe infiltrate
respimat two inhalations once daily; albuterol inhaler two puffs
every 6 hours as needed; multivitamin one tablet by mouth According to the patient’s parameters, what is his most likely
once daily, aspirin 81 mg by mouth once daily diagnosis (ie, systemic inflammatory response syndrome,
sepsis, or septic shock)?
ROS: Unable to obtain; patient has become more confused
What are the goals of treatment?
PE: Formulate an initial plan for therapy.
Within normal limits except as noted below

Monitoring and Treatment Strategies to •• Initiate step-down therapy based on microbiologic cultures
Maximize Efficacy and Minimize Toxicity for to prevent resistance, reduce toxicity, including Clostridioides
Antimicrobials difficile infection, and minimize cost.
•• Administer broad-spectrum antimicrobials for initial therapy •• Monotherapy is equivalent to combination therapy once a
as early as possible and within the first hour of recognition of causative pathogen has been identified in the vast majority of
sepsis. cases.
•• Appropriate cultures should be obtained before initiating Duration of Anti-infective Therapy
antibiotic therapy but should not prevent prompt
administration. Average duration of antimicrobial therapy for septic patients is
7 to 10 days. However, durations vary depending on the site of
•• Administer antibiotics that concentrate at the site of infection. infection and response to therapy. Step-down therapy from IV to
•• Evaluate and integrate rapid diagnostic findings if available to oral antimicrobials is recommended in certain targeted patients
ensure optimal therapy. who are hemodynamically stable, afebrile for 48 to 72 hours, have
•• Monitor patient parameters to ensure adequate dosing. a normalized WBC, have an organism susceptible to oral thera-
•• Abnormal renal and hepatic function will increase drug pies, and are able to take oral medications. Antimicrobials should
concentration and predispose the patient to toxicity. be discontinued if it is determined the cause of shock was not
infectious.
•• Ensure antibiotic dosing is changed to normal doses once
renal dysfunction has resolved to limit the development of Source Control
treatment failure, antimicrobial resistance, or both. Evaluate septic patients for the presence of a localized infection
•• Septic patients may have increased volume of distribution due amenable to source control measures. Common source control
to initial large-volume resuscitation. measures include drainage and debridement, device removal, and
•• Reevaluate the initial dosing regimen daily to optimize amputation.22,23 Implementation of source control methods should
activity, prevent the development of resistance, reduce toxicity, be instituted as soon as possible following initial fluid resuscita-
and decrease costs. tion. The selection of optimal source control methods must weigh
benefits and risks of the intervention. Source control measures may
cause complications (bleeding, fistulas, and organ injury); there-
fore, the method with the least risk should be employed.22
Patient Encounter Part 3: Treatment and
Outcome Evaluation HEMODYNAMIC SUPPORT
The patient has been intubated for acute hypoxic respiratory Vasopressors and Inotropic Therapy
failure secondary to pneumonia. He remains hypotensive When fluid resuscitation does not provide adequate
despite previous interventions. His serum creatinine remains arterial pressure and organ perfusion, vasopressors and/or ino-
elevated at 3.1 mg/dL (274.1 μmol/L), and he has no urinary tropic agents should be initiated. Vasopressors are recommended
output. in patients with a systolic blood pressure less than 90 mm Hg
What additional therapies will you consider adding to this or MAP lower than 60 to 65 mm Hg (8.0–8.6 kPa), after failed
patient? treatment with crystalloids.22,25,26 Vasopressors and inotropes are
effective in treating life-threatening hypotension and improving

Chisholm_Ch084_p1321-1332.indd 1326 11/10/21 4:37 PM

 CHAPTER 84 | SEPSIS AND SEPTIC SHOCK  1327

cardiac index, but complications such as tachycardia and myo-

cardial ischemia require slow titration of the adrenergic agents Patient Encounter Part 4
to restore MAP without impairing stroke volume. Vasopressor
Three days later, the patient’s blood pressure begins to
therapy may also be required transiently to sustain life and main-
respond to therapy with an MAP remaining greater than
tain perfusion in the face of life-threatening hypotension, even
65 mm Hg (8.6 kPa) on minimal vasopressor support.
when fluid resuscitation is in progress and hypovolemia has not
yet been corrected. Agents commonly considered for vasopressor What changes to the patient’s medication regimen do you
or inotropic support include norepinephrine, epinephrine, dopa- recommend?
mine, phenylephrine, vasopressin, and dobutamine. Norepineph-
rine is the first-line vasopressor to correct hypotension in septic
shock.22
Norepinephrine is a potent α-adrenergic agent with less pro- Angiotensin II is the newest drug approved for the treatment
nounced β-adrenergic activity. Doses of 0.01 to 3 mcg/kg/min of hypotension in adults with septic or other distributive shock
can reliably increase blood pressure through vasoconstriction states. A naturally occurring hormone of the renin–angiotensin–
with small changes in heart rate or cardiac index. Norepineph- aldosterone system, angiotensin causes vasoconstriction and
rine is a more potent agent than dopamine in refractory septic increases the release of aldosterone, which increases blood pres-
shock.22,25,26 Norepinephrine induces less arrhythmias compared sure. A recent study evaluated the use of angiotensin II in patients
with dopamine.38 with vasodilatory shock states, like septic shock, who were requir-
Dopamine is an α- and β-adrenergic agent with dopaminer- ing high-dose vasopressors and found that administration of
gic activity. Low doses of dopamine (1–5 mcg/kg/min) maintain angiotensin II was associated with a significant increase in the
renal perfusion; higher doses (> 5 mcg/kg/min) exhibit α- and number of patients who reached goal MAP at 3 hours and allowed
β-adrenergic activity and are frequently utilized to support blood for decreased catecholamine doses of other vasopressors.40 The
pressure and improve cardiac function, mainly through increas- exact place in therapy for angiotensin II has not been established,
ing stroke volume and heart rate. Because of the effects on heart but multimodal utilization of different vasoactive medications
rate, dopamine causes more tachycardia and thus increases may allow for decreased individual vasopressor doses, potentially
potential for arrhythmias versus norepinephrine. Based on these decreasing side effects.
data, dopamine should not be used routinely in the management
of septic shock.22,38 Low doses of dopamine should not be used for Corticosteroids
renal protection as part of the treatment of sepsis.22,25,26 Stress-induced adrenal insufficiency complicates 9% to 24%
Epinephrine is a nonspecific α- and β-adrenergic agonist of septic patients and is associated with increased mortality.
that can increase cardiac index and produce significant periph- The role of steroid use in septic shock remains unclear. Previ-
eral vasoconstriction. Some human and animal studies suggest ous studies have demonstrated a mortality benefit or quicker
it can also increase lactate levels and impair blood flow to the reversal of shock in patients with sepsis-induced adrenal insuf-
splanchnic system; however, studies comparing norepinephrine ficiency treated with hydrocortisone.41 However, a large, mul-
to epinephrine show no difference in mortality rates. Epineph- ticenter, randomized controlled trial (CORTICUS) showed no
rine may be added to norepinephrine in patients with persistent difference in mortality rates in septic shock patients treated
hypotension.22,25,26 with hydrocortisone.42 For these reasons, the use of
Phenylephrine is a fast-acting, short-acting pure α1-agonist. IV hydrocortisone at a dose of 200 mg/day should be reserved
Phenylephrine is the least likely vasopressor to cause tachycar- for patients who remain hemodynamically unstable despite
dia but may decrease stroke volume. Phenylephrine should be fluid and vasopressor therapy.22 Patients should be tapered from
reserved for use in patients with high cardiac output in whom steroid therapy when vasopressors are no longer required. The
tachycardia or ischemia limits the use of other vasopressors or as decision of whether to taper and how to taper hydrocortisone
salvage therapy.22,25,26 doses is dependent on individual patient characteristics or pro-
Vasopressin levels are increased during hypotension to main- vider preference. There is no standard recommendation; how-
tain blood pressure by vasoconstriction. However, there is a vaso- ever, an example of a steroid tapering regimen may be seen in
pressin deficiency in septic shock. Low, fixed doses of exogenous the CORTICUS trial.42
vasopressin increase MAP, leading to the discontinuation of cate-
cholamines. However, routine use of vasopressin with norepineph- Glucose Control
rine is not recommended because of no difference in mortality The optimal blood glucose range in septic patients is unknown.
compared with norepinephrine monotherapy.39 Vasopressin is a Tight glycemic control (80–110 mg/dL [4.4–6.1 mmol/L])
direct vasoconstrictor without inotropic or chronotropic effects improved survival in postoperative surgical patients but did not
and may result in decreased cardiac output and hepatosplanchnic show a benefit in medical critically ill patients.43,44 Subsequent
flow. The addition of vasopressin to norepinephrine may be con- studies, such as the large NICE-SUGAR trial, do not demonstrate
sidered in patients with refractory shock despite adequate fluid a mortality benefit in favor of tight glycemic control, but actu-
resuscitation and high-dose catecholamines.22,25,26 ally a higher incidence of severe hypoglycemia and increased
Dobutamine is recommended as the first-line inotropic agent. mortality rates.45 Following initial stabilization of sep-
Dobutamine is a β-adrenergic inotropic agent that can be utilized tic patients, current guidelines recommend initiating IV insulin
for the improvement of cardiac output and oxygen delivery. Doses therapy when two consecutive blood glucose measurements are
of 2 to 20 mcg/kg/min increase cardiac index; however, heart rate greater than 180 mg/dL (10.0 mmol/L) and then maintaining a
increases significantly. Dobutamine should be considered in sep- blood glucose of less than or equal to 180 mg/dL (10.0 mmol/L).
tic patients with adequate filling pressure and blood pressure, but Blood glucose levels should be monitored frequently, every 1 to
low cardiac index. If used in hypotensive patients, dobutamine 2 hours, until glucose values and insulin infusion rates are stable,
should be combined with vasopressor therapy.22,25,26 every 4 hours thereafter.22

Chisholm_Ch084_p1321-1332.indd 1327 11/10/21 4:37 PM

 1328  SECTION 15 | DISEASES OF INFECTIOUS ORIGIN

Adjunctive Therapies considered in patients with pH less than 7.15 who are not respon-
▶▶ Blood Product Administration sive to vasopressors or have increasing vasopressor requirements
to maintain an MAP of at least 65 mm Hg.22
There are no trials showing the optimal hemoglobin concen-
tration in patients with sepsis. However, based on other studies ▶▶ Metabolic Resuscitation
that included subgroups of septic patients, with the exception of
There has been growing interest in vitamin deficiencies in patients
myocardial ischemia, severe hypoxemia, acute hemorrhage, or
with sepsis, specifically vitamin C and thiamine (vitamin B1).
ischemic coronary artery disease, it is recommended to target a
Several studies have evaluated the administration of high-dose
hemoglobin concentration of 7 to 9 g/dL (70–90 g/L; 4.34–5.59
vitamin C, thiamine, or the combination of vitamin C, thiamine,
mmol/L), and red blood cell transfusion should occur when
and hydrocortisone (commonly referred to as metabolic resus-
the hemoglobin concentration is less than 7 g/dL (70 g/L; 4.34
citation) on vasopressor requirements and patient outcomes.48,49
mmol/L).22,46 Erythropoietin-stimulating agents, fresh frozen
There are no robust data to support the use of these medications
plasma, and antithrombin should not be administered, unless
in patients with septic shock. Ongoing studies will help elucidate
other compelling indications exist, to treat sepsis-induced
if there is clinical value in the future use of these interventions.
abnormalities.22

▶▶ Sedation, Analgesia, and Neuromuscular Blockade

▶▶ Venous Thromboembolism Prophylaxis
It is common for sepsis patients to require mechanical ventila- Combination of pharmacologic and mechanical prophylaxis
tion during their course of therapy. These patients may require against VTE is recommended for septic patients when possible.
analgesic or sedative agents to facilitate mechanical ventilation. Low-dose unfractionated heparin, low-molecular-weight heparin
Historically, concerns for patient safety and posttraumatic stress (eg, enoxaparin or dalteparin), or pentasaccharide therapy (eg,
disorder emphasized the need for sedation in these patients. fondaparinux) may be utilized. Graduated compression stock-
However, there continues to be ample data indicating that, while ing or an intermittent compression device is recommended for
sedation may be necessary, limiting exposure to certain sedative patients with a contraindication to heparin products (thrombo-
medications, specifically benzodiazepines, may improve morbid- cytopenia, severe coagulopathy, active bleeding, or recent intra-
ity and mortality rates. When sedation is needed, a standardized cerebral hemorrhage).22
protocol targeting specific endpoints should be utilized.22,47
▶▶ Stress Ulcer Prophylaxis
Neuromuscular blockade usually is reserved for patients in
whom sedation alone does not improve the effectiveness of Patients with severe sepsis are at increased risk of developing a
mechanical ventilation. Neuromuscular blockers may lead to pro- stress ulcer bleeding event. Stress ulcer prophylaxis using either
longed skeletal muscle weakness and should be avoided if pos- a histamine-receptor antagonist (H2 blocker) or proton pump
sible, especially if receiving concomitant corticosteroid therapy, inhibitor (PPI) is recommended in septic patients. Patients at the
which can increase the risk of critical illness polyneuropathy. greatest risk of stress ulcers include those who are coagulopathic,
Patients requiring neuromuscular blockade should be monitored, mechanically ventilated (> 48 hours), or hypotensive. Histamine
and intermittent bolus doses or continuous infusion should be 2–receptor antagonists (eg, ranitidine) are more efficacious than
utilized. Monitor depth of neuromuscular blockade with either sucralfate. There is an ongoing debate to determine if PPIs (eg,
train-of-four stimulation or other forms of clinical assessment omeprazole) are more efficacious than H2 blockers with low-
when using continuous infusion. Patients with early, sepsis- quality data showing conflicting outcomes.22,50 Both, however,
induced ARDS with a partial pressure of oxygen (PaO2)/fraction demonstrate equivalence in the ability to increase gastric pH.22
of inspired oxygen (FiO2) less than 150 mm Hg (20.0 kPa) may The benefit of prophylaxis must be weighed against the poten-
benefit from a short course of a neuromuscular blocking agent tial effect of increased stomach pH and development of infec-
not to exceed 48 hours.22 tious complications, such as hospital-acquired pneumonia and/
or C. difficile infection. When patients no longer have an indica-
▶▶ Renal Replacement Therapy tion for stress ulcer prophylaxis or at major transitions of care (eg,
Patients who develop AKI as a manifestation of sepsis may transition out of the ICU), it is imperative that the stress ulcer
require some type of renal replacement therapy. There are no data prophylactic medication is discontinued to avoid patients being
currently that suggest superiority for either continuous or inter- discharged home on unnecessary medications.
mittent modalities of hemodialysis as both have demonstrated
similar mortality rates. In patients who are hemodynamically ▶▶ Nutrition
unstable, continuous renal replacement therapy may be more Meeting the nutritional needs of septic patients can be challeng-
beneficial due to less overall hypotension during therapy. ing, especially in patients who are hemodynamically unstable.
When possible, early initiation of enteral nutrition should be con-
▶▶ Sodium Bicarbonate Therapy sidered to maintain gut mucosa and potentially decrease the risk
It has been theorized that vasopressors that act on catecholamines of bacterial translocation leading to infection. Patients who are
to increase blood pressure (eg, norepinephrine, epinephrine, hemodynamically unstable may not tolerate enteral nutrition and
phenylephrine, dopamine) may become less responsive at pH are at risk of gut ischemia. Parenteral nutrition alone or in con-
values less than 7.15. However, there are no robust data that have junction with enteral nutrition should not be initiated in the first
demonstrated improved hemodynamics, decreased vasopres- 7 days as this has not been shown to improve outcomes.22
sor requirements, or improved clinical outcomes when sodium
bicarbonate is administered at lower pH values. Conversely, Prognosis
bicarbonate administration has been associated with sodium and There are various factors that influence an outcome. Gram-
fluid overload although direct causality has not been established. negative bacteria are more likely to produce septic shock than
For these reasons, sodium bicarbonate therapy should only be Gram-positive bacteria (50% vs 25%) and have a higher mortality

Chisholm_Ch084_p1321-1332.indd 1328 11/10/21 4:37 PM

 CHAPTER 84 | SEPSIS AND SEPTIC SHOCK  1329

rate than other pathogens. This may be related to the severity of SARS-CoV-2 infection is associated with significant arterial and
the underlying condition. Patients with rapidly fatal conditions, venous complications, occurring in as much as 35% of hospi-
such as leukemia, aplastic anemia, and burns, have a worse prog- talized critically ill patients. Prothrombotic divergence is char-
nosis than patients with nonfatal underlying conditions, such as acterized by decreased concentrations of protein C, protein S,
diabetes mellitus or chronic renal insufficiency. Other factors that and antithrombin coupled with elevated d-dimer, fibrinogen,
worsen the prognosis of septic patients include advanced age, and factor VIII concentrations. This coagulation imbalance
malnutrition, resistant bacteria, utilization of medical devices, leads to microthrombi findings upon autopsy in multiple organ
and immunosuppression. Data for long-term mortality are lack- systems.53 These findings are primarily responsible for the sig-
ing (it is estimated that the mortality for sepsis survivors within nificant morbidity and mortality in COVID-19 patients compli-
the first year is 20%).51 Patients may have prolonged physical dis- cated by sepsis syndromes.
ability related to muscle weakness and posttraumatic stress. Treatment is divided primarily into nonpharmacologic and
pharmacologic therapies for severe COVID-19. Nonpharma-
cologic therapy is similar to ARDS with a specific emphasis on
SEPSIS AND SEPTIC SHOCK: CORONAVIRUS proning, which is recommended for patients with persistent
DISEASE 2019 (COVID-19) hypoxemia despite increasing supplemental oxygen requirements
While most patients infected with severe acute respiratory who otherwise are not indicated for intubation. Pharmacologic
syndrome coronavirus 2 (SARS-CoV-2) will experience mild– treatment for septic shock is similar to other etiologies (including
moderate COVID-19 infection, a significant percentage (~5%) VTE prophylaxis/anticoagulation when indicated). A number of
will require intensive care unit management due to mechanical treatments aimed at critically ill patients with severe COVID-19
ventilation and/or septic shock. In these patients, viral replica- have been attempted, including convalescent plasma, remdesivir,
tion rapidly increases, highlighting the profound inflamma- and dexamethasone. While data are mixed with regard to clinical
tory response (ie, cytokine storm) leading to diffuse alveolar outcomes of convalescent plasma and remdesivir, dexamethasone
wall thickening and macrophage distribution into airspaces is currently recommended for all hospitalized patients requir-
with subsequent pulmonary edema and formation of ground- ing supplemental oxygen, with the greatest benefit of decreased
glass infiltrates. These pathophysiologic features can lead ulti- mortality demonstrated in those patients receiving mechanical
mately to ARDS and profound respiratory failure.52 In addition, ventilation.54

Patient Care Process

Collect Information: •• Formulate appropriate dosing regimens for medications
•• Retrieve available diagnostic and laboratory data, especially involved in therapy and revise as needed.
CBC with differential, CMP, ABG, and lactic acid. •• Consider the need for adjunctive therapies such as
•• Measure vital signs, including temperature, heart rate, analgesics, sedatives, insulin, neuromuscular blockers, blood
respiratory rate, and blood pressure. product administration, renal replacement therapy, sodium
bicarbonate, venous thromboembolism prophylaxis, stress
•• Speak with patient/family and review records to identify
ulcer prophylaxis, and nutrition.
previous antimicrobial history, microbiologic cultures, and
medication allergies. Implement the Care Plan:
Assess the Information: •• Patient parameters may change frequently, thus requiring
modifications of therapy. Reassess vital signs and laboratory
•• Based on physical examination and review of systems,
values often and adjust medications and dosing regimens as
determine whether the patient meets criteria for sepsis.
needed.
•• Assess for a potential source of infection.
•• Involve the patient and family members in each step of the
•• Identify comorbidities, concomitant medications, and/or care plan.
laboratory findings that may impact antimicrobial choices.
Follow-up: Monitor and Evaluate:
•• Understand which parameters indicate effective/ineffective
resuscitation. Plan for additional resuscitation therapy if the •• Continually monitor patient parameters to ensure optimal
patient remains hypotensive. therapy to minimize morbidity and mortality.
•• Continue antimicrobial therapy on average for a total of 7
Develop a Care Plan:
to 10 days, depending on the site of infection and response
•• Formulate an appropriate plan for antimicrobial therapy, to therapy. Monitor for ongoing signs and symptoms of
which should include starting broad-spectrum agents that infection, including leukocytosis, fever, hypotension, and
cover the most likely pathogens within 1 hour of suspecting tachycardia.
sepsis.
•• Step-down therapy from IV to oral antimicrobials when the
•• Make recommendations to control the source of infection. patient is hemodynamically stable, afebrile, has a normalized
•• Develop a strategy to optimize hemodynamics. Crystalloid WBC, has a pathogen susceptible to oral therapy, and is able
fluids should be used first, followed by vasopressors and/or to take oral medications.
inotropes. Corticosteroids may be utilized in patients who •• Escalate or deescalate antimicrobial therapy based on culture
remain hemodynamically unstable. and susceptibility results, which should be assessed daily.

Chisholm_Ch084_p1321-1332.indd 1329 11/10/21 4:37 PM

 1330  SECTION 15 | DISEASES OF INFECTIOUS ORIGIN

Abbreviations Introduced 15. Opal SM, Girard TD, Ely EW. The immunopathogenesis of sepsis
in elderly patients. Clin Infect Dis. 2005;41(Suppl 7):S504–S512.
in This Chapter 16. Kim PK, Deutschman CS. Inflammatory responses and mediators.
Surg Clin North Am. 2000;80(3):885–894.
ABG Arterial blood gas
17. van der Poll T, van Deventer SJ. Cytokines and anticytokines
AKI Acute kidney injury
ARDS Acute respiratory distress syndrome in the pathogenesis of sepsis. Infect Dis Clin North Am.
CBC Complete blood count 1999;13(2):413–426, ix.
CMP Comprehensive metabolic panel 18. Rubio I, Osuchowski MF, Shankar-Hari M, et al. Current gaps in
CVP Central venous pressure sepsis immunology: new opportunities for translational research.
DIC Disseminated intravascular coagulation Lancet Infect Dis. 2019;19(12):e422–e436.
EGDT   Early goal–directed therapy 19. Holtzman C, Whitney D, Barlam T, Miller NS. Assessment of
Fio2 Fraction of inspired oxygen impact of peptide nucleic acid fluorescence in situ hybridization
HES Hydroxyethyl starch for rapid identification of coagulase-negative staphylococci in
IV Intravenous the absence of antimicrobial stewardship intervention. J Clin
MAP Mean arterial pressure Microbiol. 2011;49(4):1581–1582.
MDR Multidrug-resistant 20. Harbarth S, Garbino J, Pugin J, et al. Inappropriate initial
Pao2 Partial pressure of oxygen antimicrobial therapy and its effect on survival in a clinical trial
PPI Proton pump inhibitor of immunomodulating therapy for severe sepsis. Am J Med.
SOFA Sequential (Sepsis-related) Organ Failure 2003;115(7):529–535.
Assessment 21. Garnacho-Montero J, Garcia-Garmendia JL, Barrero-Almodovar
VTE Venous thromboembolism A, et al. Impact of adequate empirical antibiotic therapy on the
outcome of patients admitted to the intensive care unit with sepsis.
REFERENCES Crit Care Med. 2003;31(12):2742–2751.
1. Russell JA. Management of sepsis. N Engl J Med. 2006;355(16): 22. Rhodes A, Evans LE, Alhazzani W, et al. Surviving sepsis
1699–1713. campaign: international guidelines for management of sepsis and
2. Angus DC, van der Poll T. Severe sepsis and septic shock. septic shock: 2016. Crit Care Med. 2017;45(3):486–552.
N Engl J Med. 2013;369(21):2063. 23. Jimenez MF, Marshall JC, International Sepsis Forum. Source
3. Singer M, Deutschman CS, Seymour CW, et al. The Third control in the management of sepsis. Intensive Care Med. 2001;
International Consensus Definitions for Sepsis and Septic Shock 27(Suppl 1):S49–S62.
(Sepsis-3). JAMA. 2016;315(8):801–810. 24. O’Grady NP, Alexander M, Dellinger EP, et al. Guidelines for the
4. Hotchkiss RS, Karl IE. The pathophysiology and treatment of prevention of intravascular catheter-related infections. Centers
sepsis. N Engl J Med. 2003;348(2):138–150. for Disease Control and Prevention. MMWR Recomm Rep.
5. Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology 2002;51(RR-10):1–29.
of sepsis in the United States from 1979 through 2000. N Engl J 25. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy
Med. 2003;348(16):1546–1554. in the treatment of severe sepsis and septic shock. N Engl J Med.
6. Rhee C, Dantes R, Epstein L, et al. Incidence and trends of sepsis 2001;345(19):1368–1377.
in US hospitals using clinical vs claims data, 2009-2014. JAMA. 26. Hollenberg SM, Ahrens TS, Annane D, et al. Practice parameters
2017;318(13):1241–1249. for hemodynamic support of sepsis in adult patients: 2004
7. Hoste EA, Lameire NH, Vanholder RC, et al. Acute renal failure update. Crit Care Med. 2004;32(9):1928–1948.
in patients with sepsis in a surgical ICU: predictive factors, 27. PRISM Investigators, Rowan KM, Angus DC, et al. Early, goal-
incidence, comorbidity, and outcome. J Am Soc Nephrol. directed therapy for septic shock—a patient-level meta-analysis.
2003;14(4):1022–1030. N Engl J Med. 2017;376(23):2223–2234.
8. Poutsiaka DD, Davidson LE, Kahn KL, et al. Risk factors for 28. Finfer S, Bellomo R, Boyce N, et al. A comparison of albumin and
death after sepsis in patients immunosuppressed before the onset saline for fluid resuscitation in the intensive care unit. N Engl J
of sepsis. Scand J Infect Dis. 2009;41(6–7):469–479. Med. 2004;350(22):2247–2256.
9. Wafaisade A, Lefering R, Bouillon B, et al. Epidemiology and risk 29. Ferrada P, Anand RJ, Whelan J, et al. Qualitative assessment
factors of sepsis after multiple trauma: an analysis of 29,829 patients of the inferior vena cava: useful tool for the evaluation of fluid
from the Trauma Registry of the German Society for Trauma status in critically ill patients. Am Surg. 2012;78(4):468–470.
Surgery. Crit Care Med. 2011;39(4):621–628. 30. Schefold JC, Storm C, Bercker S, et al. Inferior vena cava diameter
10. Lin MT, Albertson TE. Genomic polymorphisms in sepsis. Crit correlates with invasive hemodynamic measures in mechanically
Care Med. 2004;32(2):569–579. ventilated intensive care unit patients with sepsis. J Emerg Med.
11. Rudd KE, Johnson SC, Agesa KM, et al. Global, regional, 2010;38(5):632–637.
and national sepsis incidence and mortality, 1990-2017: 31. Caironi P, Tognoni G, Masson S, et al. Albumin replacement
analysis for the Global Burden of Disease Study. Lancet. in patients with severe sepsis or septic shock. N Engl J Med.
2020;395(10219):200–211. 2014;370(15):1412–1421.
12. Bodey GP, Mardani M, Hanna HA, et al. The epidemiology 32. Patel A, Laffan MA, Waheed U, Brett SJ. Randomised trials of
of Candida glabrata and Candida albicans fungemia in human albumin for adults with sepsis: systematic review and
immunocompromised patients with cancer. Am J Med. meta-analysis with trial sequential analysis of all-cause mortality.
2002;112(5):380–385. BMJ. 2014;349:g4561.
13. Costa SF, Marinho I, Araujo EA, et al. Nosocomial fungaemia: a 33. Perner A, Haase N, Guttormsen AB, et al. Hydroxyethyl starch
2-year prospective study. J Hosp Infect. 2000;45(1):69–72. 130/0.42 versus Ringer’s acetate in severe sepsis. N Engl J Med.
14. Marie C, Muret J, Fitting C, Payen D, Cavaillon JM. Interleukin-1 2012;367(2):124–134.
receptor antagonist production during infectious and 34. Myburgh JA, Finfer S, Bellomo R, et al. Hydroxyethyl starch or
noninfectious systemic inflammatory response syndrome. Crit saline for fluid resuscitation in intensive care. N Engl J Med.
Care Med. 2000;28(7):2277–2282. 2012;367(20):1901–1911.

Chisholm_Ch084_p1321-1332.indd 1330 11/10/21 4:37 PM