Download Hematologic Changes in Sepsis and Their Therapeutic Implications

Hematologic Changes in Sepsis and Their
Therapeutic Implications
Richert E. Goyette, M.D.,1 Nigel S. Key, M.D.,2 and E. Wesley Ely, M.D., M.P.H.3
ABSTRACT
The blood and bone marrow constitute the hematologic organ system. Unlike
other organ systems, hematologic organs are distributed in space and provide for a variety
of seemingly unrelated functions. The hematologic system has both cellular and fluid-phase
elements. Cellular elements include erythrocytes, leukocytes, and platelets; fluid phase
elements include coagulation factors, natural antithrombotics, and proteins of the fibrinolytic system. The most common abnormalities of the hematologic system in patients
with sepsis are anemia, leukocytosis, thrombocytopenia, and activation of the hemostatic
system. Dysfunction of the hematologic organ system is an early manifestation of severe
sepsis and is seen in virtually all patients with this disease. In concert with alterations in the
endothelium, hematologic changes reflect both the body’s reaction to an infectious insult as
well as attempts to restore homeostasis. Dysfunction of the hematologic organ system can
contribute to multiple organ dysfunctions and death. Recognizing these sepsis-associated
changes and understanding the underlying pathophysiology are key to improving outcomes
in patients with this deadly disease.
KEYWORDS: Coagulation, erythrocytes, hematology, inflammation, leukocytes,
platelets, sepsis
Objectives: After reading this article, the reader should be able to: (1) recognize the blood and bone marrow as an organ system; (2)
understand the key role played by the endothelium in the septic process; (3) discuss alterations in red blood cells, white blood cells, and
platelets that occur in sepsis; and (4) describe the therapeutic implications of alterations in formed elements during sepsis.
Accreditation: The University of Michigan is accredited by the Accreditation Council for Continuing Medical Education to sponsor
continuing medical education for physicians.
Credits: The University of Michigan designates this educational activity for a maximum of 1 category 1 credit toward the AMA
Physician’s Recognition Award.
S
epsis with acute organ dysfunction (severe sepsis) is common, frequently fatal, and associated with
significant use of health care resources. In the United
States alone, there are more than 800,000 cases of severe
sepsis each year.1 The mortality in patients with severe
sepsis ranges from 28 to 50% or greater. Angus et al
report that each organ dysfunction increases the mortality rate of severe sepsis by 20%. It has been estimated
that with the increase in the elderly population, use of
more invasive diagnostic and therapeutic interventions,
Management of Shock; Editor in Chief, Joseph P. Lynch, III, M.D.; Guest Editors, Arthur P. Wheeler, M.D., Gordon R. Bernard, M.D. Seminars
in Respiratory and Critical Care Medicine, volume 25, number 6, 2004. Address for correspondence and reprint requests: E. Wesley Ely, M.D.,
M.P.H., Center for Health Services Research, #6109 MCE, Vanderbilt University Medical Center, Nashville, TN 37232-8300. E-mail:
[email protected]. 1Consultant, hematology and oncology, private practice, Knoxville, Tennessee; 2Department of Medicine, Division of
Hematology, Oncology, and Transplantation, University of Minnesota Medical School, Minneapolis, Minnesota; 3Department of Medicine,
Division of Allergy/Pulmonary/Critical Care Medicine and Center for Health Services Research and the VA Tennessee Valley Geriatric Research,
Education, and Clinical Center (GRECC), all at the Vanderbilt University School of Medicine, Nashville, Tennessee. Copyright # 2004 by
Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662. 1069-3424,p;2004,25,06,645,659,
ftx,en;srm00336x.
645
646
SEMINARS IN RESPIRATORY AND CRITICAL CARE MEDICINE/VOLUME 25, NUMBER 6
and increasing numbers of immunocompromised patients, the incidence of severe sepsis will increase at least
1.5% per year for the next 50 years. The hematologic
organ system is a major element in the response to a
septic insult and plays a pivotal role in the resolution
phase of severe sepsis.2 Therefore, it is important for
critical care specialists to be aware of the manifestations
and consequences of hematologic dysfunction in patients
with severe sepsis.
THE HEMATOLOGIC ORGAN
SYSTEM DEFINED
Organ systems consist of anatomically and/or physiologically related components. Because the activity of most
organ systems is focused on a single goal (e.g., gas
exchange), the concept is easy to understand. Unlike
the heart or the lungs, however, the hematologic system
is diffuse, has both cellular and fluid phase elements, and
serves a variety of unrelated functions. Consequently, its
role as an organ system is often underappreciated.
Cellular elements of the hematologic system originate
in the bone marrow, transit the blood, and then either
exit the body or are cleared by the reticuloendothelial
system. Cellular elements include red blood cells, white
blood cells, and platelets. Fluid-phase components include coagulation factors, natural anticoagulants, and
proteins of the fibrinolytic system. Disparate physiological functions of the hematologic system include gas
exchange, acid–base balance, innate and adaptive immunity, and hemostasis. Hematologic changes are almost universal in patients with severe sepsis. Although
sepsis-associated hematologic changes can be beneficial,
dysfunction, evidenced by alterations in the number of
cells, the properties of cells, fluidity of the blood, and/or
integrity of the vasculature, can contribute to both
morbidity and mortality.
THE ENDOTHELIUM
Endothelial cell function is tightly integrated with the
physiology of the hematologic system, and dysfunction
plays a key role in many of the hematologic manifestations of sepsis. Therefore, a brief discussion of endothelial cell physiology is warranted before sepsis-related
changes in the hematologic organ system are discussed.
With 1011 endothelial cells and a surface area
estimated to exceed 1,000 m2, the endothelium surpasses
the skin as the largest organ in the body.3,4 The great
majority of the endothelium is located in the microvasculature where the endothelial surface area per unit of
blood volume is 2 to 3 thousand times greater than in the
larger blood vessels.5 Endothelial dysfunction is a key
element in the pathogenesis of severe sepsis. This may be
secondary to the effects of endotoxin, proinflammatory
cytokines, reactive oxygen species, or other substances on
2004
endothelial cells. For example, a single injection of
endotoxin in rabbits produces desquamation of 25%
of the aortic endothelial surface area within 5 days.6
Alternately, organisms such as Rickettsiae can directly
infect endothelial cells and disrupt cellular homeostasis.
Severe endothelial injury from a variety of etiologies can
produce microvascular coagulopathy and acute organ
dysfunction.7
Elevated circulating levels of the soluble (s) forms
of thrombomodulin (TM), ICAM-1 (intercellular adhesion molecule-1), E-selectin, and von Willebrand factor
(vWF) are frequently measured markers of endothelial
dysfunction and injury.8 In children with septic shock,
soluble TM levels are increased and the levels correlate
with survival status and the extent of organ dysfunction.9
In a population of patients with sepsis, plasma levels of
vWF, ICAM-1, and sE-selectin were significantly increased within 8 hours of the development of the acute
respiratory distress syndrome (ARDS).10 Sepsis is associated with apoptosis of various cell types.11 In patients
with severe sepsis, endothelial cell apoptosis may produce
abnormalities in trafficking of blood cells into an infected
focus, vasoregulation, and the antithrombosis–thrombosis balance that may contribute to the pathogenesis of
septic manifestations such as multiple organ dysfunction,
impaired microvascular blood flow, and inability of the
body to restore homeostasis.
ALTERATIONS OF THE ERYTHRON
IN SEPSIS
The erythron is the organ responsible for the production
of erythrocytes. It is composed of committed red blood
cell progenitors, developing erythrocytes, reticulocytes,
and mature red blood cells. Mature red blood cells are
anucleate, hemoglobin (Hb)-containing sacs, with enzyme systems necessary to maintain the integrity of the
sac (red blood cell membrane) and prevent oxidative
damage to the contents (Hb).
Physiology
The sole function of Hb is oxygen transport; when fully
saturated, each gram of Hb can bind 1.39 mL of oxygen.
Oxygen delivery is a critical physiological process that is
a function of microcirculatory blood flow, oxygen concentration in the blood, and oxyhemoglobin dissociation.
Because the absolute amount of oxygen physically dissolved in the blood is minimal, the arterial oxygen
saturation (SaO2) essentially describes the oxygen that
is bound to Hb. The oxygen affinity of Hb is a measure
of the pO2 at which Hb is 50% saturated with oxygen
(P50). In nonsmokers, the P50 is normally 26.6 mm Hg
and varies with the concentration of 2,3-DPG (2,3diphosphoglycerate), temperature, and pH.
The Hb concentration is a function of patient age,
gender, race, activity level, and altitude above sea level.
HEMATOLOGIC CHANGES IN SEPSIS/GOYETTE ET AL
Following puberty, Hb levels are 2.0 g/dL higher in
men than in women, a difference that is a function of the
physiological effects of testosterone on the erythron.
African Americans have Hb levels that are 0.5 g/dL
lower than Caucasians. A recumbent posture for 1 hour
will result in a 6% decrease in HCT (hematocrit), with a
return of the value to normal 15 minutes after resuming an upright posture. These changes are the result of
postural fluid shifts between the intra- and extravascular
components of the extracellular fluid space. Over time,
altitude-dependent increases in erythropoiesis in individuals residing at 15,000 feet can produce a ‘‘normal’’ Hb
that may be as much as 4 g/dL greater than that
observed in healthy adults residing at sea level.
Pathophysiology
Alterations in the properties of individual red blood cells
may contribute to the pathophysiology of sepsis. These
alterations include abnormalities of oxygen carrying
capacity and changes in red cell mechanical properties
that may alter the rheology of the microcirculation. In
addition, changes in the erythron can affect other components of the hematologic system. For example, the
HCT level is inversely correlated with the bleeding time,
a test that reflects both platelet function and microvascular integrity.
ALTERATIONS IN OXYGEN CARRYING CAPACITY
The capacity of the blood to carry oxygen can be
impaired by changes in Hb’s oxygen affinity or its
concentration or both. Fever and acidosis in patients
with sepsis shift the oxygen dissociation curve to the
right. As a consequence of the decreased oxygen affinity,
red blood cells release oxygen to the tissues more readily.
Increased levels of erythrocyte 2,3-DPG also favor oxygen release. Unfortunately, 2,3-DPG is depleted in
banked blood and is not restored to normal levels for
6 to 12 hours following transfusion. Consequently,
immediately following transfusion the Hb in transfused
red blood cells fails to release oxygen as readily and at
the same pO2 levels as Hb in metabolically intact
erythrocytes.
ALTERATIONS IN PROPERTIES OF ERYTHROCYTES
In patients with sepsis, alterations in the mechanical
properties of the red blood cells may impair microcirculatory blood flow, which may in turn decrease tissue
oxygen delivery. Cellular deformability, the ability of a
cell to change its shape in response to applied forces, is
an important determinant of microcirculatory blood
flow. Decreased red blood cell deformability occurs early
in sepsis and has the potential to reduce blood flow and
increase the time required for cells to transit the microcirculation.12 Alterations in the red blood cell membrane
responsible for increased cellular rigidity probably
include damage to membrane proteins from reactive
oxygen species generated by inflammatory cells and
ischemic tissues. Thus, in addition to its negative effect
on oxygen delivery, decreased red blood cell deformability may contribute to organ dysfunction outside of
the hematologic system. Although its clinical utility has
not been established, altered red blood cell deformability
has been proposed as a tool to detect sepsis before the
appearance of more classical signs and symptoms.13
Experimentally, red blood cell aggregates that
could impair microcirculatory blood flow have also
been reported in sepsis.14 This tendency to form aggregates likely has a nonimmunologic etiology and may be a
consequence of increased acute phase proteins such as
fibrinogen that can decrease the erythrocytes’ negative
charge (zeta potential), the electrical potential that
promotes mutual red blood cell repulsion. Although
the significance of red cell aggregates in sepsis pathophysiology is unclear, one laboratory manifestation of
the decrease in erythrocyte zeta potential is an increased
erythrocyte sedimentation rate (ESR).
Clinical and Laboratory Manifestations
Extravascular leak of fluids at the onset of sepsis has the
potential to produce hemoconcentration. This situation
can result in a relative erythrocytosis during the initial
phase of sepsis. With time, however, the most common
alteration in the erythron in patients with sepsis is
anemia. Anemia results in a decrease in the oxygencarrying capacity of the blood. Although anemia can be
identified by decreases in the number of erythrocytes or a
fall in the HCT, the diagnosis is best established by the
Hb concentration. This does not reflect superiority of
one parameter or another; instead, the Hb concentration
is preferred because it most closely reflects the blood’s
oxygen carrying capacity.
Diagnostic Considerations
Apart from hemodilution, anemia may be the result of
blood loss, decreased production (hypoproliferative anemia), or increased destruction (hemolytic anemia) of red
blood cells. In most instances, therefore, the anemia in
patients with sepsis is multifactorial in origin.
BLOOD LOSS
Anemia in patients with severe sepsis may be a consequence of bleeding. In most instances, the diagnosis and
source of blood loss are readily apparent. The source of
the anemia may be less evident in patients who become
septic following major trauma with bleeding into the soft
tissues of the thigh or the retroperitoneum, for example.
Significant blood loss can also occur as a result of
multiple phlebotomies necessary to obtain blood to
diagnose and monitor patients with critical illnesses
647
648
SEMINARS IN RESPIRATORY AND CRITICAL CARE MEDICINE/VOLUME 25, NUMBER 6
2004
such as severe sepsis. It has been estimated that repeated
phlebotomies result in a loss of 24 to 41 mL of blood
daily.15–17 In a study of patients with arterial lines
hospitalized in the intensive care unit (ICU), the mean
blood loss was 944 mL.18 This is equivalent to almost
20% of the blood volume of a 70 kg man and contributed
to transfusion requirements in this population.
ANEMIA OF SYSTEMIC INFLAMMATION
Patients with the systemic inflammatory response syndrome (SIRS) may be anemic. The anemia of systemic
inflammation has also been called the anemia of chronic
disease. However, the latter term is a misnomer because
this syndrome can develop within days of the inciting
event in patients with sepsis and other inflammatory
disorders.19 Characteristic pathophysiological features of
this hypoproliferative anemia include a reticuloendothelial block in iron transport, decreased sensitivity of the
erythron to erythropoietin, and shortened red blood cell
survival.20 It is tempting to speculate that because this
cytokine-mediated process effectively sequesters iron
required for microbial growth (as well as erythropoiesis),
it may have evolved as a protective mechanism.21
The anemia is generally mild with Hb and HCT values
of 8 g/dL and 24% or greater, respectively. The block in
reticuloendothelial iron transport is significant enough
in approximately one third of patients to produce microcytic red blood cell indices suggestive of iron deficiency.
Other laboratory features include low serum iron and
iron-binding capacity and increased serum ferritin.
Multiorgan dysfunction in patients with severe sepsis
can contribute to the onset or progression of the anemia
of systemic inflammation.
Figure 1 Peripheral blood smear from a patient with disseminated intravascular coagulation. Important manifestations include
the presence of schistocytes (fragmented erythrocytes) and the
relative absence of platelets.
with underlying glucose-6-phosphate dehydrogenase
(G6PD) deficiency (more common in those of Mediterranean or African descent) or an unstable hemoglobin. In addition, endogenous oxidants and acid
metabolites produced during acute infection may contribute to hemolysis in susceptible patients. During the
workup of suspected hemolysis in a patient with sepsis, it
is important to bear in mind the possibility that a
positive direct antiglobulin (Coombs’) test may be produced not only by autoantibodies but also by adsorption
of certain antibiotics (e.g., cephalosporins) onto the
surface of red blood cells. In most cases of severe sepsis,
infection-associated hemolysis plays only a minor role in
clinical manifestations and outcome. However, experimental evidence suggests that hemoglobinemia can increase lethality of endotoxemia through a tumor necrosis
factor (TNF)-mediated mechanism.23
HEMOLYSIS
Sepsis may precipitate disseminated intravascular coagulation (DIC) with hemolysis due to red blood cell
fragmentation. Approximately 25% of patients with
DIC will have clinical evidence of microangiopathic
hemolysis manifested by the presence of schistocytes
on their peripheral blood smear (Fig. 1). Although the
clinical presentation is usually not that of severe sepsis,
babesiosis, bartonellosis, leishmaniasis, and malaria can
also be associated with a prominent hemolytic component. Clinically significant hemolysis can occur in patients with sepsis due to Clostridium perfringens or type B
Haemophilus influenzae. In the former, intravascular
hemolysis is due to the effects of bacterial lysolecithins
that directly lyse red blood cell membranes. In the latter,
polyribosyl ribosyl phosphate (PRP) from bacterial capsular polysaccharides is released during infection and
binds to the red cell surface.22 Subsequently, these
patients may develop antibodies to PRP, which may
result in complement-mediated hemolysis. Treatment of
infections with sulfonamides and other antibiotics can
precipitate Heinz body hemolytic anemia in patients
Treatment Considerations
Critically ill patients typically receive multiple transfusions of packed red blood cells (PRBCs). In one study,
85% of patients hospitalized in the ICU for more than 1
week received blood transfusions (9.5 0.8 U per patient).24 These were not single-unit transfusions nor
were they generally administered because of bleeding.
Instead, patients were transfused a constant 2 to 3 units
of PRBCs per week with approximately one third of the
units administered in the absence of a clear indication.
Packed red blood cells are prepared by differential
centrifugation, which removes 80% of the plasma
resulting in a volume and HCT of 350 mL and
60%, respectively. When transfused into a 70 kg adult,
one unit of packed red blood cells will increase the Hb
and HCT by 1 g/dL and 3%, respectively. Obviously,
continued bleeding or aggressive fluid resuscitation will
blunt this response.
Until recently, many critical care specialists chose
a relatively liberal transfusion strategy based upon the
HEMATOLOGIC CHANGES IN SEPSIS/GOYETTE ET AL
assumption that a more normal oxygen-carrying capacity
would benefit the recovery process.25 Despite its lifesaving benefits, however, transfusion of red blood cells is
associated with several potential complications. In addition to alloimmunization and transmission of infectious
disease, complications of particular importance in the
ICU setting include acute hemolytic transfusion reactions, transfusion-related acute lung injury (TRALI),
and nosocomial infections secondary to transfusionmediated immunomodulation.26 Hébert et al recently
compared restrictive and liberal transfusion strategies in
a population of critically ill subjects.27 They reported
that a restrictive transfusion strategy with a Hb trigger of
< 7 g/dL was as effective as a more liberal strategy that
used a Hb of < 10 g/dL to trigger transfusion of red cells
in patients without an acute coronary syndrome. However, the mortality benefit of this intervention remains to
be confirmed by a large-scale, randomized clinical trial.
One strategy to improve oxygen carrying capacity
without transfusion is the administration of recombinant
human erythropoietin (rHuEPO). Exogenous rHuEPO
compensates for the body’s relatively blunted erythropoietic drive produced by the anemia of systemic inflammation. In a recent randomized controlled trial of
critically ill patients, administration of 40,000 units of
rHuEPO decreased the total number of transfused units
of PRBCs by 19% compared with placebo.28 In addition,
patients who received rHuEPO had higher Hb levels,
despite receiving fewer transfusions than the control
group. However the outcomes were similar between
the groups, and a cost–benefit analysis of this strategy
is not available at this time.
The complex systems that balance tissue oxygen
delivery with tissue oxygen demand are impaired and can
produce tissue hypoxia in patients with sepsis and other
critical illnesses. Given the importance of oxygen availability to the tissues, several randomized, controlled
clinical trials have evaluated the role of increased global
oxygen delivery (DO2). Surprisingly, trials of supranormal DO2 have reported conflicting results, with some
showing decreased mortality, others showing no change,
and one reporting increased mortality.29 Recently, Rivers
et al reported that early goal-directed therapy (EGDT)
consisting of standard measures to optimize the central
venous pressure (CVP), mean arterial pressure (MAP),
and urine output (UOP) plus treatment to increase
central venous oxygen saturation (SCVO2) to 70% or
greater significantly decreased mortality in patients with
severe sepsis.30 Critical elements of the process included
initial treatment for 6 or more hours in the emergency
department, continuous monitoring of SCVO2, transfusions of PRBCs to a target HCT of at least 30%, and
inotropic support with dobutamine if the SCVO2 remained < 70% despite optimized CVP, MAP, UOP,
and HCT. Thus early identification and treatment of
global tissue hypoxia despite stable vital signs with
transfused red blood cells and other appropriate measures can help to restore the balance between oxygen
delivery and oxygen demand.
ALTERATIONS OF LEUKOCYTES IN SEPSIS
Leukocyte alterations are common in patients with
severe sepsis. In fact, white blood cells 12.0 109 /L,
4.0 109 /L, or a left shift with > 10 immature
neutrophils (bands) are one of the four SIRS criteria
employed to establish a diagnosis of sepsis in patients
with infection.31
Pathophysiology and Background
The initial response to a septic challenge is met by
elements of the innate limb of the immune system. In
humans, cellular effectors of innate immunity include
neutrophils, monocytes/macrophages, natural killer
(NK) cells, and platelets. Neutrophils play important
roles by phagocytosing infectious organisms, crystalline
material (e.g., uric acid), and immune complexes. Cytokines released during inflammation and infection promote migration of bands and mature neutrophils from
the maturation-storage pool in the bone marrow into the
circulation, whereas metamyelocytes are not released
into the blood except in extreme circumstances. Following release, granulocytes randomly enter either the
circulating or the marginating pools; the concentration
of cells in each pool is approximately equal. A neutrophilic leukocytosis is an appropriate response to many
types of infections. However, when the neutrophilic
response is exaggerated, lysosomal enzymes and toxic
oxygen metabolites released by the excess of inflammatory cells can damage host tissue adjacent to the inflammatory focus. Oxidants released from granulocytes may
also inactivate tissue protease inhibitors and thereby
contribute to injury of endothelial and parenchymal cells
by allowing unchecked enzymatic activity of elastase,
collagenase, and other matrix metalloproteinases. In
addition, leukocyte counts in excess of 50 109 /L increase blood viscosity with the potential to impair
microcirculatory blood flow.
The monocyte/macrophage is the human analog
of the invertebrate amoebocyte: an innate immune cell
that is involved in response to pathogens, wound repair,
and hemolymph clotting. By virtue of their ability to
upregulate synthesis and release of cytokines such as
tumor necrosis factor alpha (TNF-a), as well as expression of tissue factor (TF), monocytes provide one of
the critical links between inflammation and coagulation.
Monocytes originate in the bone marrow and then
transit the peripheral blood to lodge in tissues as longlived macrophages, such as Kupffer’s cells and
other reticuloendothelial elements. Monocytes possess
pathogen-associated molecular pattern signaling
649
650
SEMINARS IN RESPIRATORY AND CRITICAL CARE MEDICINE/VOLUME 25, NUMBER 6
receptors such as toll-like receptors that are a key
component of the innate immune system.32 Unlike the
antigen receptors of B- and T-cells, these receptors are
encoded in the germline and trigger an immediate
effector function.
Clinical and Laboratory Manifestations
Neutrophilic leukocytosis is a common manifestation of
sepsis. It appears to result from a combination of factors,
including recruitment of mature neutrophils from the
marginating pool into the circulating pool, mobilization
of mature and developing neutrophils from the bone
marrow, and eventually increased leukopoiesis. In most
instances, the leukocytosis is moderate. Examination of
the peripheral blood smear in patients with sepsis may
show persistence of the primary granules that first appear
at the stage of the promyelocyte (toxic granulation),
residual blue-gray areas of ribonucleic acid (RNA)-rich
cytoplasm (Döhle’s bodies), and cytoplasmic vacuoles
(i.e., phagolysosomes, autophagic vacuoles). In some
instances, intracellular bacteria can be identified in a
stained buffy-coat smear. Neutropenia in patients with
sepsis can be the result of depletion of bone marrow
granulocyte precursors, a granulocytic ‘‘maturation arrest,’’ or migration of leukocytes into the infected focus
in numbers in excess of the bone marrow’s ability to
replace them in a timely fashion. Although neutropenia
can occur in adult patients with severe sepsis, it is more
common in the pediatric population.33
Diagnostic Considerations
Sepsis may be the presenting manifestation of a primary
hematologic disorder. For example, infection may be the
first clue to the neutropenia of aplastic anemia, agranulocytosis, or acute leukemia. In most instances, the
etiology of the white blood cell changes is obvious. For
example, patients with aplastic anemia are pancytopenic,
and the typical patient with sepsis will not have the
circulating blasts, promyelocytes, myelocytes, or basophilia that may be variably present depending on the type
of leukemia.
LEUKEMOID REACTION
A patient with sepsis may sometimes present with
leukocyte counts in excess of 50 109/L. This extreme
nonneoplastic leukocytosis is known as a leukemoid
(leukemia-like) reaction. Although infection secondary
to granulocyte dysfunction is uncommon in patients
with chronic myelogenous leukemia (CML), patients
with previously undiagnosed CML may also develop
sepsis. In this case, it is important to differentiate a
leukemoid reaction secondary to sepsis from CML in a
septic patient. Clues from the initial examination and
laboratory studies that are suggestive of coexistent CML
2004
include splenomegaly, granulocyte counts > 50 109 /L,
basophilia, and the presence of small numbers of circulating blasts and promyelocytes. Additional laboratory
studies are not usually required. However, unlike the
septic patient who has a normal or elevated leukocyte
alkaline phosphatase (LAP), decreased or absent stainable LAP is a feature of granulocytes in CML. Other
laboratory studies that can establish the diagnosis of
CML but are rarely required on an emergent basis
include identification of the Philadelphia chromosome
by cytogenetic analysis or fluorescent in situ hybridization, or the presence of the bcr-abl fusion oncogene, the
Philadelphia chromosome’s molecular equivalent.
Treatment Considerations
White blood cell changes of sepsis resolve with effective
therapy of the primary disorder. Although the use
of hematopoietic growth factors such as granulocytecolony-stimulating factor (G-CSF) or granulocyte
macrophage-colony-stimulating factor (GM-CSF) to
activate leukocytes and increase their numbers is theoretically attractive, particularly for patients with sepsis
and neutropenia, there is no evidence that this approach
improves outcomes.34
ALTERATIONS OF PLATELETS IN SEPSIS
Thrombocytopenia is a frequent accompaniment of
critical illness and is commonly employed in clinical
trials of severe sepsis therapies as a marker of hematologic organ system dysfunction.35,36 In the ICU setting,
platelet counts < 100,000/mm3 (< 100 109/L) are
identified in 20 to 40% of patients.37–39 In a study of
an ICU population, sepsis was identified as a major risk
factor for thrombocytopenia.37 Platelet counts in critically ill patients are also related to prognosis independent
of the Simplified Acute Physiology Score II (SAPS II)
and Acute Physiology and Chronic Health Evaluation II
(APACHE II) severity scores. In addition, a blunted rise
in previously depressed platelet counts is associated with
poorer outcomes in acutely ill patients.40
Pathophysiology and Background
Megakaryocytes are derived from multipotential hematopoietic progenitor cells. Acting under the influence of
thrombopoietin (TPO) and other cytokines, such as
stem cell factor, interleukin (IL)-3, IL-6, IL-11, and
erythropoietin, progenitor cells progressively differentiate into promegakaryoblasts and mature megakaryocytes.41 Platelet formation results from a combination
of budding of platelets from the megakaryocytes’ surface,
release of platelets from small preformed regions of
the mature megakaryocyte, and by formation of proplatelets, long filaments of megakaryocyte cytoplasm.42
HEMATOLOGIC CHANGES IN SEPSIS/GOYETTE ET AL
Proplatelets enter the circulation and progressively
fragment to yield individual platelets. In a healthy adult,
the normal platelet count is 150,000 to 450,000/mm3
(150–450 109 /L).
Clinical and Laboratory Manifestations
Sepsis-associated thrombocytopenia is multifactorial in
origin. In experimental models of sepsis, platelets adhere
to activated endothelium in multiple organs.43,44 Following adhesion, activated platelets may either dislodge
and return to the circulation or release their granule
contents and undergo irreversible aggregation with viscous metamorphosis. Inflammatory mediators and bacterial products such as endotoxin can contribute to
sepsis-associated thrombocytopenia by enhancing platelet reactivity and adhesivity.45 Phagocytosis of platelets
by reticuloendothelial elements may also contribute to
the cytopenias of sepsis. Microscopy of bone marrow
aspirates of patients with sepsis often demonstrates
hemophagocytic histiocytes.46 In a prospective study of
50 patients with sepsis and thrombocytopenia, hemophagocytosis was identified in 32 patients (64%). This
process appears to be a function of elevated levels of
macrophage-colony-stimulating factor and is correlated
with the presence and extent of multiorgan dysfunction.47 However, a cause and effect relationship has not
been demonstrated. Instead, hemophagocytosis of platelets may represent a mechanism through which nonpathogenic immunoglobulin bound to bacterial products
on the surface of platelets is cleared.46 In some patients
with sepsis, however, antibodies to specific platelet antigens such as GP IIb/IIIa or GP Ib/IX have been
detected.48 Because similar antibodies have been implicated in the pathogenesis of immune thrombocytopenic
purpura (ITP), these autoantibodies may contribute to
sepsis-associated thrombocytopenia in some patients.
Other Diagnostic Considerations
HEPARIN-INDUCED THROMBOCYTOPENIA
Critically ill patients, including those with severe sepsis,
are often exposed to heparin administered for thromboprophylaxis or to maintain patency of indwelling lines or
to coat vascular catheters. Although the prophylactic
value of unfractionated heparin and its low-molecular
weight derivatives has in general a very favorable benefit–
risk profile, even minimal amounts of these drugs may
induce heparin-induced thrombocytopenia (HIT) that
may be mistaken for the thrombocytopenia of sepsis.
Although thrombocytopenia can also be induced by other
drugs (e.g., quinidine), HIT is unique because the thrombocytopenia can be associated with significant morbidity
and mortality resulting from thrombosis. Two types of
HIT have been identified. Type I HIT presents 1 to
4 days following heparin exposure in 10 to 20% of
patients.49 The disorder is transient, non–immune mediated, and reversible despite ongoing heparin exposure. In
general, patients with type I HIT are asymptomatic and
platelet counts do not drop below 100 109 /L.
On the other hand, type II HIT is a potentially
fatal disorder that is associated with significant morbidity that may be confused with thrombotic DIC in a
patient with severe sepsis. As an immune-mediated
clinicopathological syndrome, it typically develops 5 to
14 days after heparin exposure—a time interval that can
be as short as several hours in patients who have been
recently exposed to the drug. Type II HIT develops in 1
to 5% of patients treated with unfractionated heparin,
and in a much lower proportion of patients exposed to
low molecular weight heparin (LMWH). Thrombocytopenia is more severe with platelet counts below 60 109 /L (although usually not below 20 109 /L) and
unlike type I HIT, the thrombocytopenia persists for
days after heparin has been discontinued. Paradoxically,
bleeding in patients with HIT II is rare. Instead, 50%
of patients present with arterial and venous thrombosis
that may be isolated or develop in multiple vessels. The
mortality and limb loss rate from amputation in patients
with HIT II have each been estimated to be as high as
30%. Pathophysiologically, HIT II results from antibodies, predominantly immunoglobulin G (IgG), that
develop in response to an antigen formed by a complex
between platelet factor-4 (PF-4) and heparin. Binding of
antibody to the PF-4-heparin complex induces platelet
aggregation with thrombocytopenia and platelet activation with release of platelet microparticles, which are
believed to be responsible for the thrombotic diathesis.50
When clinically suspected, HIT II should be treated
accordingly. However, the diagnosis requires subsequent
laboratory confirmation, preferably with both a functional assay of serotonin release from normal donor
platelets exposed to patient plasma in the presence of
heparin, and an assay for the presence of the PF-4heparin antibody; for example, by enzyme-linked immunosorbent assay (ELISA). In practice, the ELISA
assay (which has a moderately high sensitivity with low
specificity) is widely used as a screening test, whereas the
‘‘gold standard’’ serotonin release assay is offered by
relatively few laboratories. Treatment of confirmed cases
of HIT II (with or without confirmed thrombosis)
should include cessation of all heparin and administration of a direct thrombin inhibitor such as lepirudin or
argatroban by continuous infusion. In Canada and
Europe, danaparoid is another alternative agent that
may be used. LMWH should be avoided because antibodies can cross-react with the PF-4–LMWH complex.
Warfarin anticoagulation should be avoided until the
platelet count has recovered to at least > 100 109 /L on
antithrombin therapy because the drug may diminish
circulating protein C levels in the short term, which may
651
652
SEMINARS IN RESPIRATORY AND CRITICAL CARE MEDICINE/VOLUME 25, NUMBER 6
in turn exacerbate the procoagulant state and result in
venous limb gangrene.51
PSEUDOTHROMBOCYTOPENIA
Pseudothrombocytopenia should be suspected when a
low platelet count is reported by the clinical laboratory in
a patient without clinical manifestations of thrombocytopenia or when other organ dysfunctions are absent.
Artifactually low platelet counts may be due to several
causes. They include platelet aggregation in finger-stick
samples, giant platelets not recognized as thrombocytes
by automated cell counters, platelet satellitosis, or platelet
clumping. The latter is produced by the presence of
anticoagulant-dependent platelet agglutinins.52 Although
it is most commonly seen with EDTA, platelet clumping
can also occur with other anticoagulants.
HYPOPROLIFERATIVE THROMBOCYTOPENIA
The bone marrow of some patients with severe sepsis
and thrombocytopenia may be hypocellular with decreased numbers of megakaryocytes. Several processes
can be responsible for hypoproliferative thrombocytopenia. For example, it may be a consequence of the effects
of cytokines, drug-induced injury, preexisting disease, or
an acute folic acid deficiency.
OTHER CAUSES
Thrombocytopenia due to accelerated platelet destruction may be due to immunologic or nonimmunologic
causes in infected patients. Immunologic disorders producing thrombocytopenia can be autoimmune (e.g.,
collagen vascular disorders, drugs, etc.) or alloimmune
(e.g., posttransfusion purpura). Nonimmunologic etiologies include microangiopathic disorders such as DIC
and thrombotic thrombocytopenic purpura (TTP).
Treatment Considerations
As the platelet count progressively falls below 150 109 /L, the risk of bleeding increases; however, platelet
counts above 60 109 /L enable patients to withstand
most minor hemostatic challenges. Risk–benefit assessment should be employed when considering the administration of anticoagulants or drugs that impair platelet
function when counts are less than 30 109 /L. Platelet
counts below 20 109 /L can be accompanied by petechiae, ecchymoses, epistaxis, and serious internal bleeding. The relationships between platelet count and
bleeding risk are largely derived from observations of
patients with untreated acute leukemia in the early
1960s.53,54 Although no clear threshold for bleeding
was identified, the percentage of days in which patients
experienced thrombocytopenia-associated hemorrhage
rose steadily after platelet counts dropped below a range
of 20 109 /L to 50 109 /L. This ‘‘trigger’’ for prophylactic platelet transfusions of 20 109 /L was widely
2004
incorporated into clinical practice for patients with and
without malignancy. Since that time, and due to the
more accurate automated methods of determining platelet count and subsequent risk of spontaneous bleeding,
most clinicians now use a threshold of < 10 109 /L to
trigger prophylactic platelet transfusion in a nonbleeding, nonseptic patient with hypoproliferative thrombocytopenia. However, platelet counts of 15 to 20 109 /L
are recommended in patients with concomitant fever or
infection.55 During the study of treatment of patients
with severe sepsis with activated protein C, the safety
data were generated with platelet counts maintained
above 30 109 /L.
WATCHFUL WAITING
A septic patient with a low absolute risk for thrombocytopenic bleeding with a stable platelet count of
20 109 /L may be carefully observed. However,
other risk factors commonly associated with sepsis such
as fever, increased platelet consumption, a declining
platelet count, renal failure, and/or treatment with drugs
that inhibit platelet function or an anticoagulant warrant
prophylactic platelet transfusion.
PLATELET TRANSFUSIONS
Platelet transfusions remain the ‘‘gold standard’’ for the
treatment of clinically significant thrombocytopenia.
However, their use is associated with several complications, including pyrogenic reactions, allergic reactions,
and disease transmission.56 In fact, the risk of transmission of viral diseases is the same as that associated with
transfusion of a single unit of PRBCs (in the case of a
single donor platelet pheresis unit) or five to eight donor
exposures (in the case of pooled random donor platelets).
One unit of platelets contains the number of platelets in
a unit of fresh whole blood in a volume of 50 mL. In
the absence of a consumptive process (e.g., fever, DIC,
bleeding, or alloimmunization) or sequestration (e.g.,
splenomegaly), one unit of platelets should increase the
platelet count by 10 109 /L/m2 ( 7 109 /L in a
70 kg adult). A pheresis pack contains the equivalent of 6
to 10 units of platelets, and is obtained from a single
donor by platelet pheresis. In general, one platelet
pheresis pack will increase the platelet count of a stable
70 kg adult by 60 109 /L.
ALTERATIONS OF THE HEMOSTATIC
SYSTEM IN SEPSIS
Pathophysiology and Background
The coagulopathy of sepsis results from the effects of
cytokines, bacterial products, and in some cases, direct
endothelial cell injury. In addition to the thrombocytopenia described earlier, hemostatic changes of
sepsis-induced coagulopathy include prothrombotic
HEMATOLOGIC CHANGES IN SEPSIS/GOYETTE ET AL
manifestations and impairment of the body’s natural
fibrinolytic potential.
ACTIVATION OF PROCOAGULANT PATHWAYS
The initial step in the coagulopathy of sepsis involves
activation of the tissue factor pathway of coagulation
(formerly known as the ‘‘extrinsic pathway’’). Tumor
necrosis factor and IL-1 are believed to play a crucial
role in the activation of coagulation during sepsis.
Animal models of sepsis have established that enhanced
TF expression occurs in vivo, and that blockade of its
procoagulant activity prevents the development of coagulopathy, organ dysfunction, and death.57–60 The primary role of the contact pathway in sepsis is probably to
activate fibrinolysis and generate vasoactive kinins.61,62
Although multiple cytokines (TNFa, IL-1a, IL1-b,
IL-6, IL-8) and endotoxin itself are capable of upregulating TF synthesis by endothelial cells and monocytes in
vitro, it has not been established that these cells are the
dominant source of TF in vivo. Endotoxin and cytokineinduced activation of endothelial cell TF synthesis ia
another possibility, although detectable expression of TF
was limited to splenic endothelium in animals dying
from Escherichia coli sepsis.63 While inhibitors targeted
at enzymes ‘‘downstream’’ from TF/VIIa (such as factor
Xa and thrombin) are capable of inhibiting the coagulopathic response, they do not prevent organ damage
or death.64 Laboratory changes associated with this
prothrombotic diathesis include elevated levels of prothrombin (PT) fragment F1.2 and increased concentrations of thrombin-antithrombin (TAT) complexes. PT
fragment F1.2 is a peptide that is cleaved when PT is
converted to thrombin; elevated plasma levels of PT
fragment F1.2 are evidence of a prothrombotic diathesis.
Increased TAT complexes result from binding and
inactivation of thrombin by circulating antithrombin.
Because D-dimers are formed by plasmin-mediated
cleavage of cross-linked fibrin, elevated levels of Ddimer also support the presence of enhanced intravascular activation of the coagulation system.
receptor (EPCR) and circulating protein S, protein C is
converted to its active form and exerts several antiinflammatory, antithrombotic, and profibrinolytic properties. Levels of protein C decrease 12 hours (median)
before the onset of the clinical symptoms of severe sepsis
and septic shock and are low in more than 90% of
patients with severe sepsis.65,66 The Protein C Worldwide Evaluation in Severe Sepsis (PROWESS) trial
demonstrated that treatment with recombinant human
APC (rhAPC), drotrecogin alfa (activated), significantly
decreases all-cause 30-day mortality in patients with
severe sepsis.33 Failure of clinical trials to demonstrate
that either AT or TFPI can improve outcomes in
patients with severe sepsis suggests that drotrecogin
alfa (activated) not only possesses antithrombotic properties but also modulates endothelial cell function—a
property supported by experimental evidence.67,68
TNF-a and IL-1 have been shown to downregulate endothelial TM and EPCR expression in vitro.
EPCR, which binds to protein C, enhances thrombinTM dependent activation of protein C. Whether impairment of protein C activation on the endothelial cell
surface actually occurs in vivo during sepsis has only
recently been studied, but the data do not allow a firm
conclusion to be drawn. In meningococcemia, decreased
immunohistochemical expression of TM and EPCR in
the microvasculature of purpuric skin biopsy specimens
has been demonstrated. These findings were interpreted
to indicate that protein C activation is impaired, a
conclusion supported by in vivo data in two patients
treated with protein C concentrate.69 However, a recent
post hoc analysis of data from the PROWESS study, in
which the ratio of plasma concentrations of zymogen
protein C to APC was measured in individuals with
severe sepsis, suggested that the ability to activate
protein C may be quite variable.70 Some individuals
had no apparent defect in protein C activation because
their elevated APC levels mirrored the plasma levels of
thrombin (measured indirectly as F1.2). Others had lower
APC levels than expected for the F1.2 levels, likely
indicative of a relative failure to activate protein C.71
DOWNREGULATION OF ANTICOAGULANT PATHWAYS
The body possesses several natural antithrombotic proteins that modulate hemostasis and help to localize
coagulation to areas of tissue injury. They include
antithrombin (AT), tissue factor pathway inhibitor
(TFPI), and activated protein C (APC). Antithrombin
inactivates several serine proteases formed during coagulation and is the substrate required for heparin to exert
its anticoagulant activity. TFPI first binds and inactivates factor Xa; subsequently, the TFPI-Xa complex
inhibits factor VIIa within the factor VIIa/ TF complex.
Protein C is a vitamin K-dependent zymogen synthesized in the liver. APC is formed from protein C by
thrombin bound to TM, an endothelial cell transmembrane protein. In concert with the endothelial protein C
INHIBITION OF FIBRINOLYSIS
Inhibition of fibrinolysis promotes persistence of microvascular clots, a consequence that may significantly
impair oxygen delivery and lead to organ dysfunction.
Experimental animal and human models of sepsis have
demonstrated an early and transient activation of fibrinolysis, probably due to release of plasminogen activators, followed by prolonged inhibition. This inhibition is
probably explained by a sustained TNFa, IL-1, and
lipopolysaccharide-mediated stimulation of plasminogen activator inhibitor-1 (PAI-1) synthesis by endothelium and monocytes, and is independent of the
coagulation response. In meningococcemia, higher
PAI-1 levels have been shown to correlate with the
653
654
SEMINARS IN RESPIRATORY AND CRITICAL CARE MEDICINE/VOLUME 25, NUMBER 6
development of septic shock rather than meningitis, and
a higher risk of death. Furthermore, PAI-1 levels are at
least partly genetically determined, with higher values
identified in individuals homozygous for the 4G/4G
polymorphism of the PAI-1 gene.72,73 These clinical
observations agree with in vitro experiments using cultured cell lines, in which basal and inducible levels of
PAI-1 are higher in cell lines expressing this polymorphism. A similar observation regarding the prognostic
relevance of inhibited fibrinolysis in nonmeningococcal
sepsis has been presented.74
Clinical and Laboratory Manifestations
A sepsis-associated coagulopathy is a sign of dysfunction
of the hematologic organ system. In the Ibuprofen in
Sepsis Trial, a coagulopathy with elevated levels of Ddimer and decreased levels of protein C was almost
universal in patients with severe sepsis (Fig. 2).66 The
coagulopathy in patients with this disease is linked to
endothelial cell dysfunction and fibrin deposition,
pathological processes that contribute to multiple organ
system dysfunction and death. For example, involvement
of the pulmonary vasculature is a prominent pathological
manifestation of pulmonary dysfunction in patients with
severe sepsis and ARDS.75,76 The first manifestations
occur at the level of the endothelial cell and include
increased capillary permeability. Diffuse alveolar damage
subsequently follows and bronchoalveolar lavage at this
time demonstrates the presence of procoagulants and
decreased fibrinolytic activity. Vascular lesions correlate
with the temporal phase of diffuse alveolar damage and
include thrombotic, fibroproliferative, and obliterative
changes.
2004
sepsis. Therefore, it may be important to differentiate
sepsis-associated coagulopathy from overt DIC.
Although there is no single clinical manifestation or
laboratory value that is diagnostic of overt DIC, the
following changes are reasonably consistent with the
diagnosis: (1) presence of a disorder that is known to
be linked with DIC (such as severe sepsis), (2) mucocutaneous bleeding and dysfunction of one or more organ
systems, and (3) ongoing consumption of platelets and
coagulation factors. The first two criteria are detected by
clinical examination, whereas the third will be manifest
as thrombocytopenia and/or prolongation of one or
more global tests of coagulation. On behalf of the
International Society of Thrombosis and Haemostasis,
Taylor et al have proposed a scoring system for the
diagnosis of DIC77 (Table 1). This system utilizes the
features already described to provide a DIC score; values
5 are characteristic of DIC. Currently, the prognostic
value of this system is undergoing prospective validation.
SEVERE LIVER DISEASE
Severe liver disease can produce alterations in the hemostatic system that can resemble sepsis coagulopathy or
DIC. In fact, in the absence of clinical features it is
difficult to differentiate the two disorders with hemostatic assays alone. Patients with severe liver disease may
have hematologic changes similar to those of severe
sepsis including anemia, thrombocytopenia, and elevated
PT and aPTT values. Results of factor VII and VIII
functional assays can assist in the differential diagnosis.
Metabolic alterations in severe liver disease usually result
in a low level of factor VII and increased concentrations
of factor VIII. Because factor VIII is consumed during
clotting, patients with severe sepsis and a coagulopathy,
on the other hand, are more likely to have decreased
concentrations of factor VIII.78,79
Diagnostic Considerations
DISSEMINATED INTRAVASCULAR COAGULATION
Thrombocytopenia, prolongation of the PT and/or
activated partial thromboplastin time (aPTT) or any
combination thereof is common in patients with severe
General Treatment Considerations
In the absence of bleeding, patients with severe sepsisassociated coagulopathy can simply be monitored. In
most instances, once the infection is controlled the
Figure 2 Coagulopathy is
common in patients with
severe sepsis. Note that elevated D-dimers and decreased levels of protein C
are present in the great majority of patients with the disease. Classical markers of
disseminated intravascular
coagulation occur in only
20% of patients with the
disease. (Illustration courtesy of the National Initiative
in Sepsis Education.)
HEMATOLOGIC CHANGES IN SEPSIS/GOYETTE ET AL
Table 1 Scoring Criteria for Overt Disseminated
Intravascular Coagulation73
Does the patient have severe sepsis or another disorder
known to be associated with DIC?. If so, proceed to scoring
global coagulation test results and calculate score
Parameter
Score
PLATELET COUNT
0
> 100,000/mm3 (> 100 109 /L)
1
3
2
9
< 100,000/mm (< 100 10 /L)
3
9
< 50,000/mm (< 50 10 /L)
ELEVATED MARKER OF FIBRIN DEGRADATION*
0
No increase
1
Moderate increase
Strong increase
2
PROLONGED PROTHROMBIN TIME
0
< 3 seconds
1
> 3 seconds and < 6 seconds
2
> 6 seconds
FIBRINOGEN CONCENTRATION
0
> 100 mg/dL (> 1.0 g/L)
1
< 100 mg/dL (< 1.0 g/L)
*D-dimers, fibrin split products (FSPs), fibrin degradation products
(FDPs).
Scores of 5 or more are compatible with overt DIC and should be
followed on a daily basis.
Scores < 5 are suggestive of nonovert DIC or another disorder and
should be repeated in the next 1–2 days or as clinically indicated.
abnormal laboratory parameters return toward normal.
In patients who are bleeding despite active treatment of
their infection, a careful examination should be conducted to rule out a discrete and correctable anatomic
lesion (e.g., peptic ulcer) as the cause of the bleed. If
bleeding is from multiple sites, treatment is that of DIC
and is generally subdivided into three general areas: (1)
treatment of the primary disease, (2) platelets and
coagulation factor replacement, and (3) antithrombotics.
Platelet transfusions are often used to maintain the
platelet count above 20 109 /L, though the risk of
spontaneous bleeding in the absence of anticoagulants
usually occurs at platelet counts below 10 to 20 109 /L.
Platelets should be administered to patients with thrombocytopenia who are bleeding. Low-dose unfractionated
heparin has been used in the past to interrupt the vicious
cycle of accelerated activation of coagulation, thrombosis, and factor depletion in patients with DIC. However,
its efficacy has not been demonstrated in large-scale,
randomized clinical trials.80 Fibrinolytic inhibitors are
also not recommended in patients with severe sepsis and
DIC because they may convert a hemorrhagic form of
DIC to a microvascular thrombotic form with worsened
end organ perfusion.
Drotrecogin Alfa (Activated) Treatment-Related
Considerations
Practice guidelines from the Surviving Sepsis Campaign,
an international collaboration of professional societies
whose members are involved in the treatment of severe
sepsis, have recently strongly endorsed the administration of drotrecogin alfa (activated) to patients with
severe sepsis at high risk of death.81 Consequently, we
will briefly review hematologic issues related to drotrecogin alfa (activated) administration.
BLEEDING
As an antithrombotic agent, drotrecogin alfa (activated)
is associated with an increased risk of bleeding. In the
PROWESS trial, serious bleeding was reported in 2.0%
of placebo recipients and 3.5% of subjects who were
treated with the drug.35 Risk factors for bleeding include
platelet counts < 30 109 /L; concomitantly administered medications that increase the risk of bleeding such
as GP IIb/IIIa inhibitors, oral anticoagulants, therapeutic (not thromboprophylactic) doses of heparin; surgical
interventions; and other disorders that increase the risk
of bleeding in any candidate for systemic anticoagulation
or thrombolysis. Importantly, the increased rate of
bleeding in patients treated with drotrecogin alfa (activated) is primarily procedure-related and predominantly
develops during drug infusion. Because of its short halflife, there is no significant residual drug effect if the
infusion is temporarily interrupted 2 hours prior to any
surgical procedure. The decision as to when to restart
the infusion should be based upon the nature of the
procedure and assessment of the patient.82 Cases of
intracranial hemorrhage, a potentially life-threatening
complication, are rare and are most often associated with
platelet counts < 30 109 /L or meningitis.
USE IN PATIENTS WITH DIC
A study by Aoki et al reported that low-dose rhAPC
corrected the clinical and laboratory abnormalities in
patients with DIC more effectively than unfractionated
heparin.83 The 28-day all-cause mortality rates in patients treated with APC and heparin were 20.4% and
40%, respectively (p < 0.05). In addition, patients treated
with APC had significantly less bleeding (p ¼ 0.009).
Joyce et al performed a retrospective subgroup analysis to
determine the effect of drotrecogin alfa (activated) in
patients with severe sepsis and DIC.84 Treatment with
drotrecogin alfa (activated) produced a 42% reduction in
the relative risk of death compared with placebo (95%
confidence interval: 58 to 19).
USE IN THE ELDERLY
The incidence of sepsis and associated mortality increases significantly with age. As pointed out by Ely
et al, there are age-related inflammatory and coagulopathic changes that may increase vulnerability of older
patients to sepsis-associated hematologic changes and
adverse outcomes.85 Advancing age is accompanied by
both prothrombotic and impaired fibrinolytic activity. In
a comparative study of healthy centenarians with healthy
655
656
SEMINARS IN RESPIRATORY AND CRITICAL CARE MEDICINE/VOLUME 25, NUMBER 6
adult controls, hypercoagulability could be identified in
the very elderly by significant increases in coagulation
factor VIIa, activation peptides of factors IX and X, F1.2,
and TAT complexes.86 These changes are accompanied
by a balanced and significant increase in fibrinolytic
activity reflected in elevated D-dimers and plasmin–
antiplasmin complexes. These hemostatic alterations
may be linked to increased frailty and functional impairment in the elderly.87
Further activation of the hemostatic system of the
aged can be produced by bacterial products released
during sepsis. Yamamoto et al have demonstrated experimentally that with age there is enhanced lipopolysaccharide signaling.88 The result may be increased levels
of inflammatory cytokines and upregulation of the PAI1 gene. When the hemostatic alterations of advanced age
are coupled with the coagulopathy of sepsis, the elderly
population is placed at significant increased risk for
mortality.
Ely et al evaluated the short- and long-term
outcomes of drotrecogin alfa (activated) therapy of 386
adult PROWESS enrollees aged 75 years and older with
severe sepsis.85 Subjects treated with drotrecogin alfa
(activated) had absolute risk reductions in 28-day and
in-hospital mortality of 15.5% and 15.6%, respectively
(for both), compared with placebo recipients (p ¼ 0.002).
Long-term follow-up data indicated that survival rates
for patients treated with drotrecogin alfa (activated)
recipients were significantly higher over a 2-year period
(p ¼ 0.02). The incidences of serious bleeding events
during the 28-day study period were 3.9% and 2.2%
(p ¼ 0.34), slightly higher than the overall PROWESS
population. However, there was no interaction between
age and bleeding rates (p ¼ 0.97). Presumably as a result
of the drug’s antithrombotic properties, elderly patients
treated with drotrecogin alfa (activated) had significantly
fewer thrombotic events than controls (p ¼ 0.019). In
the presence of a favorable benefit/risk profile, this study
supports the use of the drug in the elderly at high risk of
death when the patient, family, and health care team
have chosen aggressive care.
2004
Because drotrecogin alfa (activated) inactivates coagulation factors Va and VIIIa, treatment has been shown to
prolong the aPTT by a mean of 7 seconds. The PT tests
the tissue factor and final common pathways of coagulation. Inactivation of factor Va by the drug can lead to a
prolongation of the PT. However, in the PROWESS
enrollees, a median of only 1 second separated the
placebo and intervention groups. Therefore, the PT
cannot be expected to be an accurate surrogate of drug
effect.
In the PROWESS trial, drotrecogin alfa (activated) benefited patients with normal and depressed
levels of protein C. Following treatment with the drug,
levels of zymogen protein C increased in PROWESS
enrollees, reflecting restoration of homeostasis. Antithrombin is consumed during the coagulopathy of sepsis.
Treatment with drotrecogin alfa (activated) also increased levels of this natural antithrombotic protein.
However, clinical benefit from drotrecogin alfa (activated) was seen even in subjects with normal baseline AT
activity. Thus there is no scientific basis to assay protein
C or AT levels to stratify patients for treatment or to
monitor therapy. The antithrombotic properties of drotrecogin alfa (activated) produce significant improvements in sepsis-associated elevations of the markers of
intravascular thrombin generation such as concentrations of PT fragment F1.2 and TAT complexes. Analogously therefore, routine assay of these substances is not
indicated or clinically useful.
CONCLUSIONS
The hematologic organ system is a central player in the
clinical manifestations of sepsis, sepsis pathophysiology,
and recovery from sepsis. Identification of hematologic
organ dysfunction in a patient with two or more SIRS
criteria and infection indicates that the diagnosis is
severe sepsis. Knowledge of the hematologic manifestations of sepsis can improve diagnosis and therapy of
patients with this common and frequently fatal disorder.
FINANCIAL SUPPORT
EFFECT OF DROTRECOGIN ALFA (ACTIVATED)
ON TESTS OF HEMOSTASIS
There is no need to monitor drotrecogin alfa (activated)
therapy with tests of hemostasis including protein C
assays. However, because both the sepsis coagulopathy
and the drug can alter routine tests of hemostasis,
relevant drug-related effects on coagulation parameters
are briefly described.82
Drotrecogin alfa (activated) has no direct effect
on platelet counts; however, because the drug restores
homeostasis, treatment can be accompanied by decreased
platelet consumption and restoration of platelet counts
into the normal range. The aPTT tests the contact
(intrinsic) and final common pathways of coagulation.
Dr. Ely is a recipient of the Paul Beeson Faculty Scholar
Award from the Alliance for Aging Research. He is a
recipient of a K23 from the National Institute of Health
(#AG01023–01A1) and is the Associate Director of
Research for the VA Tennessee Valley Geriatric
Research and Education Clinical Center (GRECC).
No other external funding was provided.
REFERENCES
1. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G,
Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the
United States: analysis of incidence, outcome, and associated
costs of care. Crit Care Med 2001;29:1303–1310
HEMATOLOGIC CHANGES IN SEPSIS/GOYETTE ET AL
2. Aird WC. The hematologic system as a marker of organ
dysfunction in sepsis. Mayo Clin Proc 2003;78:869–881
3. Sporn L, Huber P. Endothelial cell biology. In Colman RW,
Hirsh J, Marder VJ, Clowes AW, George JN, eds. Hemostasis
and Thrombosis. Philadelphia: Lippincott: Williams &
Wilkins; 2001
4. Hack CE, Zeerleder S. The endothelium in sepsis: source
of and a target for inflammation. Crit Care Med 2001;
29(7 Suppl):S21–S27
5. Esmon CT. The normal role of activated protein C in
maintaining homeostasis and its relevance to critical illness.
Crit Care 2001;5:S7–12
6. Leclerc J, Pu Q, Corseaux D, et al. A single endotoxin injection in the rabbit causes prolonged blood vessel dysfunction
and a procoagulant state. Crit Care Med 2000;28:3672–3678
7. Vincent JL. Microvascular endothelial dysfunction: a renewed
appreciation of sepsis pathophysiology. Crit Care 2001;5:
S1–S5
8. Reinhart K, Bayer O, Brunkhorst F, Meisner M. Markers of
endothelial damage in organ dysfunction and sepsis. Crit Care
Med 2002;30(5 Suppl):S302–S312
9. Krafte-Jacobs B, Brilli R. Increased circulating thrombomodulin in children with septic shock. Crit Care Med 1998;26:
933–938
10. Moss M, Gillespie MK, Ackerson L, Moore FA, Moore EE,
Parsons PE. Endothelial cell activity varies in patients at risk
for the adult respiratory distress syndrome. Crit Care Med
1996;24:1782–1786
11. Hotchkiss RS, Karl IE. The pathophysiology and treatment
of sepsis. N Engl J Med 2003;348:138–150
12. Baskurt OK, Gelmont D, Meiselman HJ. Red blood cell
deformability in sepsis. Am J Respir Crit Care Med 1998;157:
421–427
13. Langenfeld JE, Livingston DH, Machiedo GW. Red cell
deformability is an early indicator of infection. Surgery 1991;
110:398–403; discussion 403–404
14. Baskurt OK, Temiz A, Meiselman HJ. Red blood cell
aggregation in experimental sepsis. J Lab Clin Med 1997;130:
183–190
15. van Iperen CE, Gaillard CA, Kraaijenhagen RJ, Braam BG,
Marx JJ, van de Wiel A. Response of erythropoiesis and iron
metabolism to recombinant human erythropoietin in intensive
care unit patients. Crit Care Med 2000;28:2773–2778
16. Zimmerman JE, Seneff MG, Sun X, Wagner DP, Knaus
WA. Evaluating laboratory usage in the intensive care unit:
patient and institutional characteristics that influence frequency of blood sampling. Crit Care Med 1997;25:737–748
17. Vincent JL, Baron JF, Reinhart K, et al. Anemia and blood
transfusion in critically ill patients. JAMA 2002;288:1499–
1507
18. Smoller BR, Kruskall MS. Phlebotomy for diagnostic
laboratory tests in adults: pattern of use and effect on
transfusion requirements. N Engl J Med 1986;314:1233–1235
19. Olivares M, Walter T, Osorio M, Chadud P, Schlesinger L.
Anemia of a mild viral infection: the measles vaccine as a
model. Pediatrics 1989;84:851–855
20. Krantz SB. Pathogenesis and treatment of the anemia of
chronic disease. Am J Med Sci 1994;307:353–359
21. Andrews NC. Disorders of iron metabolism. N Engl J Med
1999;341:1986–1995
22. Shurin SB, Anderson P, Zollinger J, Rathbun RK. Pathophysiology of hemolysis in infections with Hemophilus
influenzae type b. J Clin Invest 1986;77:1340–1348
23. Su D, Roth RI, Levin J. Hemoglobin infusion augments the
tumor necrosis factor response to bacterial endotoxin (lipopolysaccharide) in mice. Crit Care Med 1999;27:771–778
24. Corwin HL, Parsonnet KC, Gettinger A. RBC transfusion in
the ICU: is there a reason? Chest 1995;108:767–771
25. Gazmuri RJ, Shakeri SA. Blood transfusion and the risk of
nosocomial infection: an underreported complication? Crit
Care Med 2002;30:2389–2391
26. Goodnough LT, Brecher ME, Kanter MH, AuBuchon JP.
Transfusion medicine: first of two parts–blood transfusion.
N Engl J Med 1999;340:438–447
27. Hébert PC, Wells G, Blajchman MA, et al. A multicenter,
randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical
Care Investigators, Canadian Critical Care Trials Group.
N Engl J Med 1999;340:409–417
28. Corwin HL, Gettinger A, Pearl RG, et al. Efficacy of
recombinant human erythropoietin in critically ill patients: a
randomized controlled trial. JAMA 2002;288:2827–2835
29. Russell JA. Adding fuel to the fire: the supranormal oxygen
delivery trials controversy. Crit Care Med 1998;26:981–983
30. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed
therapy in the treatment of severe sepsis and septic shock.
N Engl J Med 2001;345:1368–1377
31. Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and
organ failure and guidelines for the use of innovative therapies
in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of
Critical Care Medicine. Chest 1992;101:1644–1655
32. Medzhitov R, Janeway C Jr. Innate immunity. N Engl J Med
2000;343:338–344
33. Funke A, Berner R, Traichel B, Schmeisser D, Leititis JU,
Niemeyer CM. Frequency, natural course, and outcome of
neonatal neutropenia. Pediatrics 2000;106(1 Pt 1):45–51
34. Root RK, Lodato RF, Patrick W, et al. Multicenter, doubleblind, placebo-controlled study of the use of filgrastim in
patients hospitalized with pneumonia and severe sepsis. Crit
Care Med 2003;31:367–373
35. Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and safety
of recombinant human activated protein C for severe sepsis.
N Engl J Med 2001;344:699–709
36. Warren BL, Eid A, Singer P, et al. Caring for the critically ill
patient: high-dose antithrombin III in severe sepsis: a
randomized controlled trial. JAMA 2001;286:1869–1878
37. Baughman RP, Lower EE, Flessa HC, Tollerud DJ.
Thrombocytopenia in the intensive care unit. Chest 1993;
104:1243–1247
38. Stephan F, Hollande J, Richard O, Cheffi A, MaierRedelsperger M, Flahault A. Thrombocytopenia in a surgical
ICU. Chest 1999;115:1363–1370
39. Vanderschueren S, De Weerdt A, Malbrain M, et al.
Thrombocytopenia and prognosis in intensive care. Crit Care
Med 2000;28:1871–1876
40. Nijsten MW, ten Duis HJ, Zijlstra JG, Porte RJ, Zwaveling
JH, Paling JC. The TH: blunted rise in platelet count in
critically ill patients is associated with worse outcome. Crit
Care Med 2000;28:3843–3846
41. Long MW. Thrombopoietin stimulation of hematopoietic
stem/progenitor cells. Curr Opin Hematol 1999;6:159–163
42. Cramer EM. Megakaryocyte structure and function. Curr
Opin Hematol 1999;6:354–361
43. Shibazaki M, Kawabata Y, Yokochi T, Nishida A, Takada
H, Endo Y. Complement-dependent accumulation and
657
658
SEMINARS IN RESPIRATORY AND CRITICAL CARE MEDICINE/VOLUME 25, NUMBER 6
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
degradation of platelets in the lung and liver induced by injection of lipopolysaccharides. Infect Immun 1999;67:5186–5191
Tsujikawa A, Kiryu J, Yamashiro K, et al. Interactions
between blood cells and retinal endothelium in endotoxic
sepsis. Hypertension 2000;36:250–258
Salat A, Murabito M, Boehm D, et al. Endotoxin enhances in
vitro platelet aggregability in whole blood. Thromb Res 1999;
93:145–148
Stephan F, Thioliere B, Verdy E, Tulliez M. Role of
hemophagocytic histiocytosis in the etiology of thrombocytopenia in patients with sepsis syndrome or septic shock. Clin
Infect Dis 1997;25:1159–1164
Francois B, Trimoreau F, Vignon P, Fixe P, Praloran V,
Gastinne H. Thrombocytopenia in the sepsis syndrome: role
of hemophagocytosis and macrophage colony-stimulating
factor. Am J Med 1997;103:114–120
Stephan F, Cheffi MA, Kaplan C, et al. Autoantibodies
against platelet glycoproteins in critically ill patients with
thrombocytopenia. Am J Med 2000;108:554–560
Brieger DB, Mak KH, Kottke-Marchant K, Topol EJ.
Heparin-induced thrombocytopenia. J Am Coll Cardiol
1998;31:1449–1459
Warkentin TE, Hayward CP, Boshkov LK, et al. Sera from
patients with heparin-induced thrombocytopenia generate
platelet-derived microparticles with procoagulant activity: an
explanation for the thrombotic complications of heparininduced thrombocytopenia. Blood 1994;84:3691–3699
Warkentin TE, Elavathil LJ, Hayward CP, Johnston MA,
Russett JI, Kelton JG. The pathogenesis of venous limb
gangrene associated with heparin-induced thrombocytopenia.
Ann Intern Med 1997;127:804–812
Berkman N, Michaeli Y, Or R, Eldor A. EDTA-dependent
pseudothrombocytopenia: a clinical study of 18 patients and a
review of the literature. Am J Hematol 1991;36:195–201
Gaydos LA, Freireich EJ, Mantel N. The quantitative
relation between platelet count and hemorrhage in patients
with acute leukemia. N Engl J Med 1962;266:905–909
Rinder HM, Arbini AA, Snyder EL. Optimal dosing and
triggers for prophylactic use of platelet transfusions. Curr
Opin Hematol 1999;6:437–441
Drews RE. Critical issues in hematology: anemia, thrombocytopenia, coagulopathy, and blood product transfusions in
critically ill patients. Clin Chest Med 2003;24:607–622
Triulzi D. Optimizing Platelet Transfusion Therapy. Baltimore, MD: Institute for Transfusion Medicine; 1997
Osterud B, Bjorklid E. The tissue factor pathway in
disseminated intravascular coagulation. Semin Thromb Hemost
2001;27:605–617
Taylor FB Jr, Chang A, Ruf W, et al. Lethal E. coli septic
shock is prevented by blocking tissue factor with monoclonal
antibody. Circ Shock 1991;33:127–134
Creasey AA, Chang AC, Feigen L, Wun TC, Taylor FB Jr,
Hinshaw LB. Tissue factor pathway inhibitor reduces mortality
from Escherichia coli septic shock. J Clin Invest 1993;91:2850–
2860
Taylor FB, Chang AC, Peer G, Li A, Ezban M, Hedner U.
Active site inhibited factor VIIa (DEGR VIIa) attenuates the
coagulant and interleukin-6 and -8, but not tumor necrosis
factor, responses of the baboon to LD100 Escherichia coli.
Blood 1998;91:1609–1615
Pixley RA, De La Cadena R, Page JD, et al. The contact
system contributes to hypotension but not disseminated
intravascular coagulation in lethal bacteremia: in vivo use of
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
2004
a monoclonal anti-factor XII antibody to block contact
activation in baboons. J Clin Invest 1993;91:61–68
van der Poll T, de Jonge E, Levi M. Regulatory role of
cytokines in disseminated intravascular coagulation. Semin
Thromb Hemost 2001;27:639–651
Drake TA, Cheng J, Chang A, Taylor FB Jr. Expression of
tissue factor, thrombomodulin, and E-selectin in baboons with
lethal Escherichia coli sepsis. Am J Pathol 1993;142:1458–1470
Taylor FB Jr, Chang AC, Peer GT, et al. DEGR-factor Xa
blocks disseminated intravascular coagulation initiated by
Escherichia coli without preventing shock or organ damage.
Blood 1991;78:364–368
Mesters RM, Helterbrand J, Utterback BG, et al. Prognostic
value of protein C concentrations in neutropenic patients at
high risk of severe septic complications. Crit Care Med 2000;
28:2209–2216
Yan SB, Helterbrand JD, Hartman DL, Wright TJ, Bernard
GR. Low levels of protein C are associated with poor outcome
in severe sepsis. Chest 2001;120:915–922
Yan SB, Dhainaut JF. Activated protein C versus protein C in
severe sepsis. Crit Care Med 2001;29(7 Suppl):S69–S74
Cheng T, Liu D, Griffin JH, et al. Activated protein C blocks
p53-mediated apoptosis in ischemic human brain endothelium and is neuroprotective. Nat Med 2003;9:338–342
Faust SN, Levin M, Harrison OB, et al. Dysfunction of
endothelial protein C activation in severe meningococcal
sepsis. N Engl J Med 2001;345:408–416
Dhainaut JF, Yan SB, Margolis BD, et al. Drotrecogin alfa
(activated) (recombinant human activated protein C) reduces
host coagulopathy response in patients with severe sepsis.
Thromb Haemost 2003;90:642–653
Liaw P, Ferrell G, Loeb M, Foley R, Weitz J, Esmon CT.
Patients with sepsis vary markedly in their ability to generate
activated protein C [abstract]. Blood 2001;98:445a
Hermans PW, Hibberd ML, Booy R, et al. 4G/5G promoter
polymorphism in the plasminogen-activator-inhibitor-1 gene
and outcome of meningococcal disease. Meningococcal
Research Group. Lancet 1999;354:556–560
Westendorp RG, Hottenga JJ, Slagboom PE. Variation in
plasminogen-activator-inhibitor-1 gene and risk of meningococcal septic shock. Lancet 1999;354:561–563
Raaphorst J, Johan Groeneveld AB, Bossink AW, Erik Hack
C. Early inhibition of activated fibrinolysis predicts microbial
infection, shock and mortality in febrile medical patients.
Thromb Haemost 2001;86:543–549
Tomashefski JF Jr. Pulmonary pathology of the adult respiratory
distress syndrome. Clin Chest Med 1990;11:593–619
Hasegawa N, Husari AW, Hart WT, Kandra TG, Raffin TA.
Role of the coagulation system in ARDS. Chest 1994;105:
268–277
Taylor FB Jr, Toh CH, Hoots WK, Wada H, Levi M.
Towards definition, clinical and laboratory criteria, and a
scoring system for disseminated intravascular coagulation.
Thromb Haemost 2001;86:1327–1330
Mammen EF. Coagulation abnormalities in liver disease.
Hematol Oncol Clin North Am 1992;6:1247–1257
Clarke BJ, Sridhara S, Woskowska Z, Blajchman MA.
Consumption of plasma factor VII in a rabbit model of
nonovert disseminated intravascular coagulation. Thromb Res
2002;108:329–334
Feinstein DI. Diagnosis and management of disseminated
intravascular coagulation: the role of heparin therapy. Blood
1982;60:284–287
HEMATOLOGIC CHANGES IN SEPSIS/GOYETTE ET AL
81. Dellinger RP, Carlet JM, Masur H, et al. Surviving sepsis
campaign management guidelines committee. Surviving
sepsis campaign guidelines for management of severe sepsis
and septic shock. Crit Care Med 2004;32:858–873
82. Laterre PF, Heiselman D. Management of patients with
severe sepsis, treated by drotrecogin alfa (activated). Am J
Surg 2002;184(6A Suppl):S39–S46
83. Aoki N, Matsuda T, Saito H, et al. A comparative doubleblind randomized trial of activated protein C and unfractionated heparin in the treatment of disseminated intravascular
coagulation. Int J Hematol 2002;75:540–547
84. Joyce D, Yan B, Basson BR, et al. Disseminated intravascular
coagulation in severe sepsis patients treated with recombinant
human activated protein C [abstract]. Blood 2001;98:445a
85. Ely EW, Angus DC, Williams MD, Bates B, Qualy R,
Bernard GR. Drotrecogin alfa (activated) treatment of older
patients with severe sepsis. Clin Infect Dis 2003;37:187–
195
86. Mari D, Mannucci PM, Coppola R, Bottasso B, Bauer KA,
Rosenberg RD. Hypercoagulability in centenarians: the
paradox of successful aging. Blood 1995;85:3144–3149
87. Cohen HJ, Harris T, Pieper CF. Coagulation and activation of
inflammatory pathways in the development of functional decline
and mortality in the elderly. Am J Med 2003;114:180–187
88. Yamamoto K, Shimokawa T, Yi H, et al. Aging accelerates
endotoxin-induced thrombosis: increased responses of
plasminogen activator inhibitor-1 and lipopolysaccharide
signaling with aging. Am J Pathol 2002;161:1805–1814
659