Week 6/2 dq

1578185 – McGraw-Hill Professional ©

CHAPTER 141
Sepsis and Shock

Kevin Felner, MD

Robert L. Smith, MD

Key Clinical Questions

What is the definition of systemic inflammatory response syndrome (SIRS) and
how do you differentiate SIRS from sepsis, severe sepsis, and septic shock?

Which patients presenting with sepsis need admission to the intensive care unit
(ICU)?

Which septic patients require invasive monitoring (arterial catheter, central venous
catheter)?

What interventions in the treatment of sepsis improve mortality?
Which septic patients deserve empiric steroids as part of the therapeutic
regimen?

INTRODUCTION

Sepsis is a clinical syndrome that complicates severe infection and is characterized by
systemic inflammation and widespread tissue injury. The incidence and number of sepsis-

1578185 – McGraw-Hill Professional ©

related deaths has increased yearly from 1979 to 2009 with a combined peak of both
primary and secondary sepsis in 2009 greater than 1 million patients in the United States;
and sepsis is the ninth most common cause of death in the United States. Despite the
rising number of cases, earlier identification of sepsis and improved intensive medical
care has been shown to reduce the overall mortality rate to approximately 17.9%. Severity
is correlated with mortality (Table 141-1).

TABLE 141-1 Sepsis Syndromes, Definitions, and Mortality Risk

Syndrome Definition
Approximate
Mortality

Systemic
inflammatory
response syndrome
(SIRS)

At least two of the following four clinical
features:
1. Temperature >38°C or <36°C
2. Heart rate >90 beats/min
3. Respiratory rate >20 breaths/min or

PaCO2 <32 mm Hg
4. White blood cell (WBC) count >12,000

cells/mm3, or <4000 cells/mm3, or >10%
immature (band) forms

10%

Sepsis SIRS criteria plus a culture-proven infection
or presumed presence of an infection

20%

Severe sepsis Sepsis plus presence of one or more organ
dysfunctions including:
• Pulmonary dysfunction (eg, acute

respiratory distress syndrome)
• Cardiac dysfunction
• Renal dysfunction
• Hepatic dysfunction
• Neurologic dysfunction (altered

sensorium)
• Hematologic dysfunction (eg,

disseminated intravascular coagulation
[DIC], thrombocytopenia)

• Lactic acidosis (indicating end-organ
hypoperfusion)

20%-40%

Septic shock Sepsis and refractory hypotension with
mean systemic blood pressure lower than
65 mm Hg unresponsive to crystalloid fluid
challenge of 20-40 cc/kg

40%-60%

Initially successful shock resuscitation may still be associated with considerable
morbidity and mortality. Multiple organ dysfunction syndrome (MODS) refers to severe

1578185 – McGraw-Hill Professional ©

acquired dysfunction of at least two organ systems lasting at least 24 to 48 hours in the
setting of sepsis, trauma, burns, or severe inflammatory conditions so that homeostasis
cannot be maintained without intervention. Mortality is directly correlated with the number
of dysfunctional organs and the duration of dysfunction (Table 141-2). An uncontrolled
hyperinflammatory response is believed to be the cause of multiple organ dysfunction.

TABLE 141-2 Correlation between Organ Failure and Mortality in Sepsis

Organ Failure Mortality

One organ lasting more than 1 d 20%

Two organs lasting more than 1 d 40%

Three organs lasting more than 3 d 80%

The American College of Chest Physicians (ACCP) and the Society of Critical Care
Medicine (SCCM) defined the following terms to describe the spectrum of systemic
inflammation and sepsis (the International Sepsis Definitions Conference, 2001):

Systemic inflammatory response syndrome (SIRS) is a clinical syndrome that results
from activation of the immune system whether due to infection, trauma, burns, or a
noninfectious inflammatory process. This syndrome includes at least two of the
following:
(1) Temperature >38°C or <36°C
(2) Heart rate >90 beats/min
(3) Respiratory rate >20 breaths/min or PaCO2 < 32 mm Hg
(4) White blood cell count >12,000 cells/mm3, or <4000 cells/mm3, or >10%

immature (band) forms
Sepsis is a clinical syndrome that results from activation of the immune system with
a documented infection. The definition of sepsis includes the above SIRS criteria
plus a culture-proven infection or presumed presence of an infection.

A recent study has brought into the question the sensitivity of the current definition,
suggesting that many patients, usually older, do not actually even have two out of four
SIRS criteria when they are septic. These caveats have not been adopted into any formal
guidelines at this time. Clinicians should have a lower threshold for considering sepsis,
especially in older patients with a suggestive presentation despite not fulfilling the above
criteria.

The severity of sepsis is graded according to the associated organ dysfunction and
hemodynamic compromise. Severe sepsis refers to the presence of sepsis and one or
more organ dysfunctions. Organ dysfunction may be defined as hypotension, acute lung
injury including acute respiratory distress syndrome (ARDS), disseminated intravascular
coagulation (DIC), thrombocytopenia, altered mental status, mottled skin, capillary refill
greater than 3 seconds, renal dysfunction, hepatic dysfunction, cardiac dysfunction based
on echocardiography or measurement of cardiac index, or lactic acidosis indicating
hypoperfusion. The phenomenon of sepsis-induced myocardial dysfunction occurs when
patients have normal cardiac function prior to their infection and the sepsis induces a

1578185 – McGraw-Hill Professional ©

global cardiac dysfunction, seen on echocardiography as global hypokinesis, which may
impair the expected high cardiac output usually associated with a vasodilated circulation.

Septic shock refers to the presence of sepsis and refractory hypotension with mean
systemic blood pressure lower than 65 mm Hg that is unresponsive to crystalloid fluid
challenge of 20 to 40 cc/kg. Septic shock leads to acute circulatory collapse.

PATHOPHYSIOLOGY
Sepsis is as an uncontrolled inflammatory response to an infection in which a
dysregulated host immune response leads to multiorgan involvement not limited to the
source infected organ. Microbial antigens such as lipopolysaccharides (LPS) from Gram-
negative bacteria bind to Toll-like receptors on inflammatory cells, thereby causing a
complex immune reaction involving T-cells, macrophages, neutrophil, endothelial cells, and
dendritic cells. Cytokines (such as IL-1, IL-6, IL-8), growth factors (such as TNFa), high-
mobility group box-1 (HMGB-1), arachidonic acid metabolites, and nitric oxide and host
genetics likely determine the nature of the response. The complement cascade,
coagulation cascades, platelets, and leukocytes interact at the vascular endothelium level
resulting in microvascular injury, thrombosis, and loss of endothelial integrity, which
altogether results in tissue ischemia. This diffuse endothelial disruption is responsible for
the various organ dysfunctions and global tissue hypoxia that accompany severe sepsis
and septic shock. Multiple mechanisms including decreased preload, vasoregulatory
dysfunction, myocardial depression, and impaired tissue extraction due to
microcirculatory dysfunction or mitochondrial dysfunction (cytopathic hypoxia) cause
global tissue hypoxia. Some noninfectious processes (eg, pancreatitis) may also lead to a
dysregulated host immune response and multiorgan dysfunction, and these conditions are
categorized using the term SIRS. These patients appear septic without a clear infectious
source.

DIFFERENTIAL DIAGNOSIS

The differential diagnosis for conditions that cause sepsis includes conditions that
present with high-output nonshock states. Common disorders that meet SIRS criteria
include nonmassive pulmonary embolus, alcohol withdrawal, even COPD exacerbations.
Thyrotoxicosis, aortic regurgitation, arteriosclerosis, and cirrhosis may mimic sepsis with
high cardiac output state and wide pulse pressure without shock.

Conditions that belong to the category of vasodilatory or high cardiac output shock
include anaphylaxis, adrenal insufficiency, and neurogenic shock in addition to septic
shock. The other causes of shock all fall into a category of low-output states, including
cardiogenic shock, hypovolemic shock, and obstructive shock (Table 141-3).

TABLE 141-3 Differential Diagnosis of Shock

Vasodilatory shock Sepsis
Anaphylaxis
Adrenal insufficiency
Neurogenic

Low-output shock states Cardiogenic (eg, massive myocardial infarction, myocarditis,

1578185 – McGraw-Hill Professional ©

valvular disease)
Hypovolemic (eg, hemorrhagic, gastrointestinal losses, burns,
pancreatitis)
Obstructive (eg, massive PE, tension pneumothorax, auto-
PEEP, tamponade, abdominal compartment syndrome)

PE, pulmonary embolus; PEEP, positive end expiratory pressure.

DIAGNOSIS

Presentation of sepsis often varies according to infection source, patient age, underlying
comorbidities (including immune system function and cardiac status), and timing of
presentation relative to onset of sepsis. Early manifestations of sepsis (tachycardia,
oliguria, and hyperglycemia) may be subtle and easily overlooked in the hospitalized
patient. In addition, a patient with underlying poor cardiac function who becomes septic
may not be able to generate the high cardiac outputs expected in sepsis and may not have
the typical findings on physical examination of a septic patient. Signs of established
sepsis include altered mental status, metabolic acidosis and respiratory alkalosis,
hypotension with decreased systemic vascular resistance (SVR) and elevated cardiac
output, and coagulopathy. Late manifestations include acute lung injury (ALI), ARDS,
acute renal failure, hepatic dysfunction, and refractory shock.

Sepsis may be related to a systemic inflammatory response to any infectious source.
Less than 50% of septic patients will have positive blood cultures, and 20% to 30% of
patients will have no microbial cause identified from any source. Aggressive clinical
evaluation includes a detailed history and review of systems. A complete physical
examination can assess for sometimes inconspicuous and missed infection sources,
including skin and soft tissue, central nervous system, gastrointestinal tract, and
indwelling devices.

It is critical to stabilize the patient and identify the cause of the ongoing immunologic
response. Obtaining cultures for blood, urine, and other fluids early, prior to administration
of antibiotics, should be a high priority and helps preserve the integrity of results, but the
evaluation should not be at the expense of administering antibiotics expediently.
Identification of the underlying source remains paramount, and lack of source
identification and control may render choice of antibiotics meaningless. The most
common sites of infection in sepsis are the urinary and respiratory tracts, but any organ
system may be involved. Urinary sources include cystitis, pyelonephritis, and perinephric
abscess. Patients with kidney stones may develop Gram-negative septicemia. Sinusitis,
mastoiditis, pneumonia, lung abscess, and empyema may be associated with sepsis.
Gastrointestinal sources of sepsis may include esophageal rupture or perforation
following a procedure or after vomiting, cholangitis, cholecystitis, intestinal infarction or
perforation, acute pancreatitis, Clostridium difficile colitis, diverticulitis, and intra-
abdominal abscess. Postoperative mediastinitis and acute bacterial endocarditis may
lead to sepsis. Skin and soft tissue sources of sepsis include infected decubitus ulcer,
postoperative wound infection, soft tissue abscess, or necrotizing fasciitis. Vascular
causes of infection include central and peripheral lines, arterial catheters, dialysis
catheters, ventriculoperitoneal shunts and septic thrombophlebitis. Infected articular

1578185 – McGraw-Hill Professional ©

prosthetic devices have also been associated with sepsis. Meningitis and intracranial
abscess, sometimes associated with neurosurgery, are also considerations.

Relevant diagnostic studies based on symptoms, signs, and clinical suspicion in
patients may include chest or abdominal radiography and culture of blood, urine, sputum,
or other relevant body fluids that may be infected such as cerebrospinal fluid (CSF)
analysis, paracentesis in patients with ascites, or thoracentesis in patients with pleural
effusions. When plain films, blood cultures, and fluid cultures do not yield a likely
infectious culprit, advanced imaging with chest and abdominal computed tomography
may identify pulmonary infiltrates, intra-abdominal abscesses, and obstructing renal
stones. Biliary pathology may be better imaged with ultrasound. In hemodynamically
stable patients, magnetic resonance imaging (MRI), or endoscopic retrograde
pancreatography (ERCP) may be indicated. Many patients undergo echocardiography to
assess cardiac function and to identify the presence of vegetations.

TRIAGE AND HOSPITAL ADMISSION
All patients with a presentation of severe sepsis or septic shock should be admitted to or
transferred to a monitored setting that is capable of continuous vital sign monitoring with
the ability to measure central venous pressure (CVP) and central venous oxygen
saturations (ScvO2).

PRACTICE POINT

Recent data suggest that most septic shock patients may be managed without the use
of CVP or ScvO2 monitoring; however the values may still be used to assess response
to therapy in selected patients with undifferentiated or mixed shock and in patients
with underlying organ dysfunction such as chronic kidney disease and cognitive
impairment.

Those patients with SIRS and sepsis should be monitored closely if not placed in an
intensive care unit setting so that they can be treated promptly if they start to show signs
of deterioration. Vital signs should be monitored frequently in addition to telemetry and
continuous pulse oximetry. Intermediate care units (sometimes called step-down units or
transitional care units) vary from facility to facility in their capabilities for invasive
monitoring and use of vasoactive agents. The protocols and policies at individual
institutions will help determine placement of these patients, based on monitoring
requirements and response to initial resuscitation in the emergency department or on a
medical floor.

PRACTICE POINT

Early aggressive resuscitation, early antibiotics (within 1 hour of severe sepsis or
septic shock identification), and early source identification and control improve
outcomes in patients with severe sepsis and septic shock. These patients require
timely evaluation to determine admission location, and patients with marginally stable

1578185 – McGraw-Hill Professional ©

clinical parameters should be admitted to an intensive care unit setting to
expeditiously meet early care goals to improve outcomes.

MANAGEMENT
Management of severe sepsis and septic shock requires a structured approach that
ensures proper diagnostic evaluation and implementation of evidence-based interventions
in an expedient manner to improve outcomes (Figure 141-1). This approach requires (1)
empiric antibiotic coverage of an infectious source while cultures are pending, (2) optimal
fluid resuscitation, (3) pressor and/or inotrope therapy for selected patients, and (4)
consideration of additional therapies such as drainage of abscesses, removal of lines,
moderate (but not intensive) control of hyperglycemia, and (5) consideration of steroids in
selected patient subsets when indicated.

Figure 141-1 Sepsis algorithm.

1578185 – McGraw-Hill Professional ©

ANTIBIOTIC THERAPY
Initial emergency department management of patients with sepsis begins with a
heightened awareness of the condition by assessing all patients for the presence of SIRS
criteria. Numerous studies have shown that early and appropriate antibiotics are
associated with markedly improved outcomes. Antibiotics should be directed against the
likely organisms based upon the presumptive infection source. In many situations,
especially when the presumptive source of infection is not obvious, multiple antibiotic
agents should be initiated to offer broad antimicrobial coverage. Such broad coverage
should then be re-evaluated daily to optimize dosing and minimize drug interactions and
the development of resistance. Choice of antibiotic depends upon penetration into the
suspected infection site, local resistance patterns, efficacy against the most likely
organisms, prior exposure to specific antibiotics, and risks of side effects. Therapeutic
drainage of an infected space is critical to diagnose the source of infection, guide the
choice of antibiotic therapy, and facilitate recovery. In patients with devices, clinicians may
need to evaluate and consider early and rapid removal of potentially or known infected
invasive devices including central venous catheters (CVCs), peripherally inserted central
venous catheters (PICCs), urinary catheters, and other implanted hardware.

Recent evidence suggests that mortality increases with delay of antibiotics more than
1 hour after identification and management of severe sepsis or septic shock. Patients at
risk of fungal infections (ie, recent abdominal surgery, total parenteral nutrition (TPN)
administration, chronic steroid use) may benefit from empiric antifungal agents in
addition to the antimicrobial regimen.

More data is needed before recommending use of procalcitonin levels in septic
patients. While there is reasonable evidence that procalcitonin may be useful in the
management of community acquired pneumonia and COPD exacerbations, the evidence
for its use in decisions to discontinue antibiotics in septic patients is less robust. Studies
comparing a calcitonin-guided algorithm with standard management show no difference
in the amount of time spent on antibiotics.

INTRAVENOUS FLUIDS

Volume resuscitation should begin simultaneously with empiric antibiotic therapy in
patients suspected of having sepsis. In the vasodilatory state low blood pressures with
decreased venous return lead to an underfilled, but hyperdynamic heart. Rivers and
colleagues showed that early goal-directed therapy (EGDT), initiated in the emergency
department, improved mortality in patients with severe sepsis and septic shock.

For routine use in sepsis, crystalloid fluid should be used first due to evidence of
benefit, markedly lower expense, and demonstrated safety (lacking the inherent risks of
blood product administration with albumin). The Saline Versus Albumin Fluid Evaluation
(SAFE) study evaluated nearly 7,000 critically ill patients, 18% of whom had severe sepsis.
Patients were randomly assigned to receive 4% albumin versus normal saline, and
investigators reported no differences in mortality at 28 days. Additionally, there were no
significant differences seen in the sepsis subgroup. Despite a theoretic benefit to using
albumin in highly selected patients with significant volume overload, no studies support
this approach. The Albumin Replacement in Severe Sepsis or Septic Shock (ALBIOS) study
showed no mortality benefit, though there was a suggestion that patients with septic
shock benefitted from albumin.

1578185 – McGraw-Hill Professional ©

The choice of crystalloid has recently come into question based on the results of
recent data suggesting that normal saline (NS) is associated with renal insufficiency as
well as hyperchloremic metabolic acidosis. Alternatives include lactated ringers (LR) and
plasmalyte and Normosol. LR contains 4 mEq/L of potassium; however, this is unlikely to
cause a meaningful increase in serum potassium levels due to the volume of distribution
in the intracellular space, even in patients with renal failure, and any rise is offset by the
alkalinizing effect of LR. Three small randomized control trials comparing NS versus LR in
perirenal transplant surgery patients showed that patients receiving several liters of LR
had marginally lower potassium levels. There is currently insufficient data to support the
routine use of the more costly alternatives, plasmalyte and Normosol, which are balanced
crystalloid solutions; they may offset acidosis with anions that are converted to
bicarbonate.

Early goal-directed therapy includes early aggressive volume resuscitation in the first 6
hours of care, and other measures over the first hours and days of care (see Figure 141-1).
Close monitoring of central venous pressure (CVP) is accomplished with a central venous
catheter placed in the internal jugular or subclavian vein. Central venous pressure and
ScvO2 monitoring allows adjustment of or addition of interventions based on the
parameters measured within the individual patient to achieve the goal of ScvO2 at 70%, if
the patient remains hypotensive (mean arterial pressure [MAP] < 65 mm Hg) after a
reasonable fluid challenge with crystalloid (approximately 20-40 cc/kg) to optimize filling
pressures.

The EGDT algorithm, whose utility has come into question with three recent trials, uses
a CVP goal of 8 to 12 cm H2O, which is a reasonable estimate goal. However, that goal
should not be applied blindly to all patients without knowledge of coexisting conditions
including pulmonary arterial hypertension, dilated cardiomyopathy, and old right
ventricular infarction. Clinicians may follow the trend of the CVP and correlate it with the
ScvO2, patient hemodynamics, and evidence of organ perfusion including mental status
and urine output. Ample data suggest that the CVP serves as a poor predictor of volume
responsiveness, and multiple factors are necessary to determine the need for continued
volume resuscitation including passive leg raising and pulse pressure variation. Passive
leg raise is a technique in which a spontaneously breathing patient is placed with the legs
elevated, essentially transferring approximately 300 cc of intravascular fluid into the
thorax, followed by measurement of cardiac output. This technique avoids the
administration of exogenous fluid. The measurement of ScvO2 carries valuable
information and weight, offering the clinician an assessment of cardiac function and
oxygen delivery balanced against oxygen consumption.

Despite strong evidence from the original Rivers et al. EGDT trial more than a decade
ago, recent publications of the PROCESS (Protocolized Care for Early Septic Shock) trial,
the ARISE (Australasian Resuscitation in Early Septic Shock) trial, and the PRoMISe
(Protocolized Management in Septic Shock) trial have challenged the utility of the Rivers
EGDT algorithm. The three trials showed that with early recognition of septic shock in the
emergency department, management according to the Rivers’ EGDT algorithm, including
placement of central lines and measurement of CVP and ScvO2, did not improve any
outcome measures when compared to standard management. The results of these three
large, multicenter, randomized control trials have seriously challenged what had become
axiomatic since the publication of Rivers original EGDT trial. A change in management,
however, may not be immediate as more than 50% of the patients in the EGDT and non-

1578185 – McGraw-Hill Professional ©

EGDT arms of the trials had central lines placed, and use of vasopressors without
placement of a central line has not gained mainstream acceptance. That said, some
patients with early positive response to IV fluids and other aggressive measures may not
require a central line. Furthermore, use of lactate levels (rather than ScvO2 ± CVP
measurements) should provide adequate evaluation of response to therapy, supported by
recent studies.

In addition, there has been a change in the paradigm that calls for the use of large
volume resuscitation in the treatment of septic patients. Due to the ample data regarding
the dangers of over resucitation and its deleterious effects on organ function, fluid
management has shifted to somewhat less aggressive volume resuscitation and earlier
use of vasoconstrictors.

PRACTICE POINT

Central venous saturation (ScvO2) is a measure of oxygen saturation taken from the
distal tip of a central venous line inserted just proximal to the right atrium. The ScvO2
measures the balance between oxygen delivery and oxygen consumption, with normal
ScvO2 ranging between 65% and 75%. Lower values reflect a high oxygen extraction
state, usually seen in states of shock with low cardiac output (cardiogenic,
hypovolemic, obstructive).
In sepsis, as in other vasodilatory or high cardiac output states, low oxygen extraction
—possibly due to mitochondrial dysfunction—leads to higher values of ScvO2. Often,
these higher values of ScvO2 are not apparent until the patient has been adequately
resuscitated with intravascular volume expansion. The mean ScvO2 in the Rivers study
was 55%, which is lower than values seen in other sepsis trials.
Early goal-directed therapy (EGDT) studies initially suggested that clinicians should
augment therapeutic interventions when ScvO2 is less than 70% in patients with severe
sepsis or septic shock. Three recent studies have shown that outcomes are no worse
when ScvO2 is not used to guide management. More recent studies suggest that
lactate clearance of at least 10% at a minimum of 2 hours after beginning volume
resuscitation is a valid way to assess the efficacy of intravenous fluid administration.
For EGDT the order of therapy augmentation included: volume expansion (to achieve
CVP 8-12 mm Hg) → pressor agents (to achieve MAP ≥ 65 mm Hg) → transfusion of
packed RBCs (to achieve an ScvO2 ≥ 70%) → inotropic agents (to achieve an ScvO2 ≥
70%). This sequence has been challenged by the same three studies comparing EGDT
versus standard treatment. A less codified algorithm might include 20-30 cc/kg fluid
administration, pressor administration for patients who remain hypotensive with signs
of hypoperfusion, further evaluation of the need for additional fluid, along with lactate
clearance after the first 2 hours of therapy. Placement of a CVL and measurement of
ScvO2 should be individualized, and not routinely used in the care of many patients
with sepsis. Of highest importance is early antibiotic administration and intravenous
fluids via a secure peripheral or central venous line.

BLOOD TRANSFUSION

1578185 – McGraw-Hill Professional ©

Patients with severe sepsis or septic shock who have been resuscitated adequately will
usually demonstrate the physiology of a low oxygen extraction state with high ScvO2
values, but importantly, preresuscitation values may make patients appear as high oxygen
extractors, with low ScvO2 values more consistent with a low cardiac output state. Early
goal-directed therapy protocol includes transfusing red blood cells if the hematocrit is less
than 30% and the ScvO2 remains less than 70% after meeting CVP and blood pressure
goals. While a subgroup analysis in the original EGDT trial favored transfusions to
improve outcomes, potential deleterious effects from red blood cell transfusions, including
questionable efficacy of older stored blood, the immunomodulating effects of red blood
cell transfusions, and the risk of transfusion reactions, make this part of the EGDT
protocol more difficult to recommend broadly for every patient meeting EGDT criteria. The
2013 Surviving Sepsis Guidelines were revised for red cell transfusions due to the
controversy and conflicting data regarding the benefits and risks of red blood cell
transfusions. Current recommendations employ a transfusion threshold of 7 gm/dL once
tissue hypoperfusion has resolved, except in the setting of active cardiac ischemia, blood
loss, severe hypoxemia, and ischemic heart disease. The target goal recommendation is 7
gm/dl to 9 mg/dL, and transfusion for hemoglobin threshold less than 7 g/dL has been
shown to have equivalent outcomes for mortality and other relevant outcomes as
transfusion for a hemoglobin threshold less than 9 g/dL in patients with septic shock
based on the 2014 TRISS trial.

VASOACTIVE MEDICATIONS
An important aspect of sepsis management includes vasoactive medications.
Vasopressors are often required to maintain mean arterial blood pressures (MAP) above a
target value and the choice of agent depends on the physiologic need (Table 141-4). The
EGDT protocol recommends vasopressor agents to maintain MAP ≥ 65 mm Hg. There is
no firm evidence favoring one vasopressor agent over another, but norepinephrine likely
has the greatest vasoconstrictor potency along with some inotropic effect. The most
recent Surviving Sepsis Guidelines recommend epinephrine as the second line
vasopressor of choice after norepinephrine based on several randomized studies
suggesting worse outcomes with use of dopamine (compared to norepinephrine).
Epinephrine’s most concerning side effects include arrhythmias and elevated lactate
levels, which are due to beta receptor agonism rather than ongoing organ ischemia.
Dopamine may be used if a more inotropic and chronotropic effect is desired and should
be avoided if cardiogenic shock is suspected due to demonstrated increased mortality and
arrhythmias in that patient population. Low-dose dopamine does not provide renal
protection, and should not be used solely for that purpose. Phenylephrine may be the
preferred agent for blood pressure elevation in patients with prohibitive tachycardia or
arrhythmias. Vasopressin, a pure vasoconstrictor, has been used to lessen the doses of
adrenergic vasopressor agents; however, current available data does not support its
routine use in severe sepsis or septic shock.

TABLE 141-4 Vasoactive Medications in Sepsis

Medication (Dose) Inotropy Chronotropy
Arterial
Vasoconstriction Practical Uses

1578185 – McGraw-Hill Professional ©

Norepinephrine 0.01-
3.00 mcg/kg/min; 8-
30 mcg/min typical
dosing

Yes Yes (but less
than
dopamine)

Yes • First line for many
patients with severe
sepsis or septic shock

• Significant
vasoconstriction with
inotropy which is
helpful for patients
with poor left
ventricular reserve or
sepsis-related
cardiomyopathy

Dopamine 1-5
mcg/kg/min,
increased renal blood
flow; 5-10
mcg/kg/min,
increased
chronotropy/inotropy;
>10 mcg/kg/min,
predominant
vasoconstriction,
increased blood
pressure

Yes Yes Yes • Not a first line
vasopressor for severe
sepsis or septic shock.
May be useful for
severe bradycardia
and mild hypotension

• Randomized
comparison to
norepinephrine
showed no significant
differences in
mortality, but
increased arrhythmias
with dopamine and
increased mortality in
cardiogenic shock

• More tachycardia than
with norepinephrine

• More potent inotrope
than norepinephrine

• Differing effects at
escalating doses with
vasoconstriction at
highest dose

• Available in premixed
or preprepared bags
and therefore can be
initiated quickly during
emergent need

Epinephrine Yes Yes Yes • Second line
vasopressor after
norepinephrine in
severe sepsis, septic
shock. Similar to

1578185 – McGraw-Hill Professional ©

dopamine with
differing effects with
escalating doses

• Increased production
of lactate and
significant tachycardia
has kept this as a
second-line
medication. Lactate
often related to beta
receptor agonism
rather than
hypoperfusion

Phenylephrine 0.4-9.1
mcg/kg/min

No No Yes • Pure vasoconstrictor
• Used primarily in

sepsis in patients with
excessive tachycardia
or arrhythmias to
avoid medications
with chronotropic
effect

• Used in severe aortic
stenosis