Anda di halaman 1dari 15

Review article

Mechanisms of Disease

The Hemostatic System as a Modulator

of Atherosclerosis
Julian Ilcheff Borissoff, M.D., Henri M.H. Spronk, Ph.D.,
and Hugo ten Cate, M.D., Ph.D.

From the Laboratory for Clinical Throm- ardiovascular disease is one of the leading causes of death and
bosis and Hemostasis, Departments of complications worldwide. The classic concept of atherosclerosis assigns a
Internal Medicine and Biochemistry, Car-
diovascular Research Institute of Maas- pivotal role to inflammation in the onset and progression of this disease.1,2
tricht, Maastricht University Medical Cen- Various inflammatory cell types (e.g., macrophages, neutrophils, and lymphocytes)
ter, Maastricht, the Netherlands. Address play crucial roles in the destabilization and subsequent rupture or erosion of an
reprint requests to Dr. ten Cate at the
Laboratory for Clinical Thrombosis and atherosclerotic plaque, ultimately resulting in atherothrombosis.3 Inflammation is
Hemostasis, Departments of Internal closely linked to coagulation in several pathologic conditions.4 Intriguingly, extensive
Medicine and Biochemistry, Cardiovas- bidirectional cross-talk between the two systems has been established in many com-
cular Research Institute of Maastricht,
Maastricht University Medical Center, plex diseases,5,6 including atherosclerosis.
Universiteitsingel 50, P.O. Box 616, Box 8, Although there is no clinical evidence of a role for the hemostatic system in the
Maastricht 6200 MD, the Netherlands, or progression of atherosclerosis, ample experimental data indicate that platelets and
the coagulation system are important determinants of both atherogenesis and ath-
N Engl J Med 2011;364:1746-60. erothrombosis. In numerous clinical trials, the administration of antiplatelet or
Copyright 2011 Massachusetts Medical Society. anticoagulant therapy has not been associated with attenuation or regression of
plaque growth. Nevertheless, the hemostatic system is well known for its capacity
to exert a multitude of actions on the vasculature, which may influence the mo-
lecular and cellular composition of the arterial wall and presumably of the athero-
sclerotic plaque. This review covers recent advances in this field and discusses
mechanisms of hemostasis as potential modulators of plaque phenotype.

cross- ta l k mech a nisms l ink ing the hemos tat ic s ys tem

w i th atheroscl erosis
Hemostasis is accomplished through a network of processes that include the platelet
system, coagulation, and anticoagulant and fibrinolytic pathways, which all support
the dynamic equilibrium that provides proper blood flow.7,8 Such processes evolved
to maintain the blood in a fluid state under physiologic conditions and to arrest
bleeding after vascular injury9-15 (Fig. 1A and 1B). Disruption of this well-regulated
balance leads to pathologic conditions, such as thrombosis and bleeding.

molecular and cellular responses in the Vasculature

The targeting of genes that encode distinct hemostatic factors and their effect on
arterial thrombosis in vivo has been extensively studied (see Table 1 in the Supple-
mentary Appendix, available with the full text of this article at Abundant
experimental data suggest a role for various constituents of the platelet membrane
and coagulation system in the regulation of atherosclerosis progression. Beyond
their traditional hemostatic functions, platelets are considered important in proin-
flammatory conditions, such as atherosclerosis.16 In addition, numerous coagula-

1746 n engl j med 364;18 may 5, 2011

The New England Journal of Medicine

Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
mechanisms of disease

tion proteins have been implicated in processes gests an active cell-based coagulation network,
such as the disruption of the endothelial barrier, within human atherosclerotic lesions. The role of
oxidative stress, leukocyte recruitment, inflam- these coagulation proteins in atherogenesis is in-
mation, migration and proliferation of vascular dicated by increased thrombin-generating activi-
smooth-muscle cells (VSMCs), immune responses, ty in early atherosclerotic lesions, as compared
apoptosis of platelets and other cell types, and with that in stable, advanced lesions.30 These find-
angiogenesis.17,18 Some of these actions, mostly ings are supported by experimental data31 and a
mediated by the complex of tissue factor and fac- clinical study showing that increased plaque echo-
tor VIIa (TFFVIIa), factor Xa, and thrombin, in- genicity (more fibrous structure), rather than
volve the activation of G-proteincoupled prote- plaque echolucency (lipid-rich, higher content of
ase-activated receptors (PARs) 1, 2, 3, and 4. PARs inflammatory cells and thinner fibrous caps), is
are widely distributed on vascular cells under nor- associated with thrombin generation in plasma
mal conditions and are overexpressed during ath- from patients with carotid-artery stenosis.32 The
erogenesis.19 abundance of coagulation factors within early
atherosclerotic vessels and local generation of
Platelets, the Cellular Interface between thrombin or fibrin may be attributable to pri-
Hemostasis and Atherosclerosis mary protective mechanisms against vascular
Pioneering studies have documented a prominent injury. However, the persistent inflammatory en-
role of platelets in experimental studies of ath- vironment within the arterial wall, supported in
erogenesis.20,21 Platelets exert a plethora of pro- part by coagulation-mediated actions, may main-
atherogenic activities and create an interface tain local thrombin generation, which will even-
between hemostasis, innate immunity, and in- tually turn into a vicious cycle, contributing to the
flammation in atherosclerosis.16 A systemic in- formation of intraplaque thrombi33,34 and thus
flammatory environment, independent of vessel- ultimately leading to plaque instability.
wall injury, induces a phenotypic switch to a
proatherogenic endothelium. This results in en- T issue Fac t or (E x t r insic)
hanced expression of cell-adhesion molecules, such Path wa y
as P-selectin and E-selectin. The primary adhe-
sion of platelets on a compromised vascular en- Tissue factor is a transmembrane class II cytokine
dothelial surface is accomplished through the receptor, which is considered the primary physi-
binding of platelet glycoprotein Ib receptors toologic trigger of the coagulation cascade.8 Tissue
von Willebrand factor, whereas firm adhesion is factor is also physiologically essential for vascu-
mediated through 3 integrins. Once adherent, lar development. In mice, tissue factor deficiency
platelets also secrete atherogenic mediators, such
is associated with a high rate of embryonic death
as cytokines, chemokines, growth factors, adhe- and impaired vascular integrity. Tissue factor is
sion molecules, and coagulation factors. The up- differentially distributed among the various cell
regulation of P-selectin expression on the surfaces
types of the vessel wall. Under physiologic condi-
of both platelets and endothelial cells potentiates
tions in normal blood vessels, the inner endothe-
the interactions with P-selectin glycoprotein li-lial lining does not express tissue factor, whereas
gand 1, which is expressed on leukocyte mem- the surrounding layers, consisting of VSMCs, ad-
branes. The binding between platelets and circu- ventitial fibroblasts, and pericytes, show abundant
lating leukocytes (monocytes and neutrophils), synthesis of tissue factor. This specific vascular
dendritic cells, and progenitor cells produces co-
localization of tissue factor is generally attributed
aggregates that support further leukocyte activa-to its role in the prevention of bleeding after in-
jury, also referred to as a hemostatic envelope.35
tion, adhesion, and transmigration, processes con-
sidered to be critical for plaque formation and Within the atherosclerotic lesion, tissue fac-
progression22-29 (Fig. 2). tor is predominantly localized on macrophages,
VSMCs, and foam-cellderived debris within the
coagulation system during atherosclerotic necrotic core.30,36-38 Tissue factor activity is sig-
plaque progression nificantly higher in lesions obtained from pa-
We have found a local synthesis of several func- tients with unstable angina or myocardial in-
tionally active coagulation proteins, which sug- farction than in those from patients with a stable

n engl j med 364;18 may 5, 2011 1747

The New England Journal of Medicine
Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

Figure 1 (facing page). Platelets and Coagulation Factors in the Regulation of Thrombus Formation.
Panel A shows platelet adhesion and aggregation, in which atherothrombosis begins with an endothelial injury or
rupture of an atherosclerotic plaque. This process triggers transient neurohumoral vasoconstrictor mechanisms,
which are reinforced by the release of endothelium-derived factors, such as endothelin. The platelet membrane re-
ceptors glycoprotein Ib/IX/V and glycoprotein VI elicit platelet tethering to the exposed thrombogenic subendotheli-
al proteins, von Willebrand factor, and collagen. In addition, glycoprotein VI generates intracellular signals to medi-
ate platelet adhesion and aggregation through the activation of integrin receptors, such as glycoprotein Ia/IIa and
glycoprotein IIb/IIIa, with the latter also serving as a receptor for fibrinogen. These molecular events ultimately con-
tribute to the formation of the primary hemostatic plug.10 Panel B shows the tissue factor (extrinsic) pathway,
in which tissue factor, the major trigger of coagulation, is exposed at the site of plaque erosion or rupture. Tissue
factor forms a catalytic complex with factor VIIa that leads to the subsequent activation of factors IX and X. In a
so-called prothrombinase complex, activated factor X together with activated factor V promotes a downstream en
zymatic cleavage of prothrombin, which yields small amounts of thrombin.11 Thrombin is a pleiotropic, central co
agulation enzyme12 that not only converts fibrinogen into fibrin but also has a substantial role in the activation of
platelets and activates factor XIII to induce fibrin polymerization, a fundamental process for the formation of a stable
clot, or thrombus. Furthermore, by supporting positive-feedback activation of the upstream factors V, VIII, and XI,
thrombin plays a crucial part in the amplification and propagation phases of coagulation. The activated platelet sur-
face is also a critical catalyst for the coagulation cascade. Platelets actively participate in the clotting process by in-
troducing extra amounts of tissue factor, factor V, fibrinogen, and factor XIII into the system, derived from various
local sources (fibrinogen and factors V and XIII stored in granules),13 and facilitating the direct activation of factor
XI by thrombin and the subsequent activation of factor IX on the platelet surface. Factor IXa forms the so-called te-
nase complex together with factor VIIIa, thereby igniting a burst of additional thrombin generation, which is essen-
tial in forming sufficient fibrin and sealing the defect. Panel C shows the contact activation (intrinsic) pathway,
which is not considered to be essential for protection against bleeding in vivo, even though its components may be
involved in the pathogenesis of arterial thrombosis.14 The exposure of plasma prekallikrein, high-molecular-weight
kininogen, and factors XI and XII to anionic surfaces15 results in the conversion of prekallikrein to kallikrein, which
activates factor XII into factor XIIa but also cleaves high-molecular-weight kininogen, leading to the release of the
inflammatory mediator and vasodilator bradykinin. Factor XIIa activates factor XI and favors the conversion of more
prekallikrein to kallikrein, thereby reciprocally amplifying the cascade. This sequence of proteolytic reactions leads to
the activation of factor IX, which ultimately cleaves factor X into its active form and culminates in the convergence
of both coagulation pathways. Gray circles indicate the inactive form of a coagulation protein, and green circles in
dicate the active form.

form of cardiovascular disease,39-41 suggesting a of TFFVIIa on atherosclerosis progression. Lev-

role of this coagulation protein in plaque throm- els of plasma tissue factor antigen, modulated by
bogenicity. Factor VII is also extrahepatically ex- known polymorphisms of the tissue factor gene,
pressed within both normal and atherosclerotic are positively associated with both an increased
vessels and colocalizes with tissue factor on risk of death from cardiovascular causes44 and
macrophages and VSMCs.30 Apart from its co- an increased carotid intimamedia thickness,45
agulation properties, the TFFVIIa complex is which is considered a marker of subclinical ath-
multifunctional, with a capacity to promote cell erosclerosis. A similar relation between factor VII
signaling, gene transcription, and subsequent and increased intimamedia thickness has been
protein synthesis. PAR-2 activation is essential in documented both in healthy young adults and in
the mediation of TFFVIIainduced signaling. The patients with peripheral arterial disease.46,47
latter may engage several proatherogenic pro-
cesses, such as monocyte and fibroblast chemo- C om mon C oagul at ion Path wa y
taxis, inflammation, VSMC migration and pro-
liferation (vascular remodeling), angiogenesis Pleiotropic Factor X a
(contributing to plaque destabilization), induction Once activated, factor Xa initiates intracellular
of oxidative stress in macrophages, and apopto- signaling in various cell types of the cardiovascu-
sis42 (Fig. 3). Surprisingly, reduced vascular ex- lar system, preferentially mediated by PAR-2 or,
pression of tissue factor does not affect athero- when in ternary complex with TFFVIIa, through
sclerosis progression in transgenic mice.43 both PAR-1 and PAR-2.17 PAR-1, PAR-2, or both are
There are few clinical data regarding the role present in abundance on endothelial cells, leuko-

1748 n engl j med 364;18 may 5, 2011

The New England Journal of Medicine

Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
mechanisms of disease



Fibrinogen Glycoprotein Prekallikrein Kallikrein

Ib/ IX/ V XIa
XII Anionic
Glycoprotein VI
weight kininogen Bradykinin
von Willebrand Collagen

B Stable Clot

Prothrombinase Prothrombin

factor Tissue
Positive thrombin feedback
V Va

Tenase IXa

Additional amount of factor IX
provided by platelets


cytes, VSMCs, fibroblasts, and dendritic cells. Fac- flammation, leukocyte transmigration,
Draft 4 resteno-
Author Borissoff
tor Xadependent, PAR-mediated signaling con- sis, and angiogenesis (Fig.
Fig #
3). 1
Of note, vascular
tributes to the production of proinflammatory remodeling and neointimalTitle formation were
cytokines, including interleukin-6, interleukin-8, reduced on targeted delivery
ME of nonspecific fac-
and chemokine (C-C motif) ligand 2 (CCL2), and tor Xa inhibitors (heparin
DE and Longolow-molecular-
Artist Knoper
to the expression of cell-adhesion molecules, in- weight heparins) coupled to AUTHORan antifibrin
cluding E-selectin, intracellular adhesion molecule body.48 Figure has been redrawn and type has been reset
Please check carefully

1 (ICAM-1), and vascular-cell adhesion molecule Issue date 5/05/11

1 (VCAM-1), along with tissue factor up-regulation, Thrombin

VSMC proliferation, and the release of growth fac- Thrombin is a unique serine protease that is piv-
tors (vascular endothelial growth factor, platelet- otal to coagulation and that may also display vari-
derived growth factor, and transforming growth ous actions toward other systems (e.g., immune,
factor ).17 All these may contribute to the pro- nervous, gastrointestinal, and musculoskeletal sys-
gression of atherosclerotic plaque, involving in- tems). Governed by the interaction and proteo-

n engl j med 364;18 may 5, 2011 1749

The New England Journal of Medicine
Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

Homeostasis Imbalance
Resting platelets Platelet activation
ADP Hemoglobinhaptoglobin
Thrombin complex clearance
Plasmin Further
Platelet activating factor activation
Thromboxane A2 CD163
Nitric Platelet Monocyte-to-
ATP Prostacyclin factor 4
oxide macrophage
Adenosine differentiation
CD39 CD73

Healthy endothelium Impaired endothelium

Atherosclerotic plaque development

Adhesion to endothelial Leukocyte recruitment

cells and monocytes ADAM15, CCL2 and 3, P-selectin,
P-selectin ICAM-1, VCAM-1

Inflammation Differentiation to foam cells

Interleukin-1, platelet factor 4, Platelet Monocyte Platelet factor 4
neutrophil-activating peptide 2,
RANTES, CD40L, TNF-, interleukin-8
Thrombosis Plaque destabilization
Tissue factor Matrix metalloproteinases

Figure 2. Platelets in Atherogenesis. COLOR FIGURE

Intact endothelium normally expresses CD39 (ecto-ATPase) and CD73 (ecto-5-nucleotidase), which act in tandemDraft to induce
6 the 4/19/11
down of the prothrombotic adenosine 5-triphosphate (ATP) and adenosine diphosphate (ADP) into the largelyAuthor Borissoff
antiinflammatory ade-
Fig # 2
nosine, thus preventing platelet activation and aggregation. Healthy endothelium also secretes vasodilators, such as prostacyclin and ni-
tric oxide, which have potent antiadhesive and antiaggregating effects. At the time of activation, platelets undergo a substantial change
in shape and promptly release a variety of autocrine and paracrine mediators such as ADP, epinephrine, and thromboxane
ME A 2. Studies
investigating how platelets orchestrate these widely differing atherogenic actions have provided an increased understanding Longo of the
Artist Knoper
mechanisms involved. Much attention has focused on cytokine-like and chemokine systems such as the CD40CD40L dyad, CCL5
(RANTES), and platelet factor 4.23,24 Platelet factor 4 supports monocyte differentiation into macrophages and down-regulates the
Figure has been redrawn and type has athe-
been reset
Please check carefully
roprotective receptor CD163, which accounts for the clearance of hemoglobinhaptoglobin complexes. Transgenic mice
Issue date 5/05/11
lacking platelet
factor 4 have diminished progression of atherosclerosis. Furthermore, CD40 and its ligand, CD40L, which belongs to the superfamily of
tumor necrosis factor receptor and ligand, is widely expressed in the vessel wall (e.g., in endothelial cells, vascular smooth-muscle cells,
and fibroblasts) and several immune constituents (monocytes or macrophages, neutrophils, mast cells, T and B cells, and dendritic
cells).25 The complex array of proinflammatory, immune-modulating effects and prothrombotic features26 assert an integral role for
CD40CD40L in atherogenesis. Overall, these findings support the hypothesis that platelets are important proinflammatory players that
elicit multifaceted cellular interactions and are directly involved in the early development of atherosclerotic lesions. Platelets are primary
mediators in both adaptive and innate immunity.27 Hence, the targeting of platelet chemokines appears to be therapeutically unsuitable
in the context of atherosclerosis because of the severe impairment of multiple systemic immune responses, which may also result in car-
cinogenesis.28,29 ADAM15 denotes ADAM metallopeptidase domaincontaining protein 15, CCL2/3 chemokine (C-C motif) ligand 2/3,
ICAM-1 intercellular cell-adhesion molecule 1, TNF- tumor necrosis factor , and VCAM-1 vascular-cell adhesion molecule 1.

lytic activation of its direct cellular targets (PAR-1, bin favors the transformation of protein C into
3, and 4),49,50 thrombin is entwined with the reg- activated protein C, a potent anticoagulant and
ulation of vascular physiology and pathophysi- antiinflammatory molecule. Moreover, thrombin
ology51 (Fig. 3). Thrombin is an example of a can diminish the release of interleukin-12 and
multifaceted molecule with broad physiologic promote the up-regulation of interleukin-10 in
properties. By binding to thrombomodulin, throm- monocytes, thus inducing immunosuppressive

1750 n engl j med 364;18 may 5, 2011

The New England Journal of Medicine

Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
mechanisms of disease

X Fibrinogen
factor Endothelial permeability
VII VIIa Monocyte transmigration
Tissue Prothrombin LDL accumulation
factor Platelet reactivity

Indicates activation of corresponding
protease-activated receptor below

Protease-activated Protease-activated Protease-activated Protease-activated

receptor 1 receptor 2 receptor 3 receptor 4
Endothelial cells Endothelial cells Endothelial cells Endothelial cells
Leukocytes Leukocytes Fibroblasts Leukocytes
Vascular smooth- Vascular smooth- Dendritic cells Vascular smooth-
muscle cells muscle cells muscle cells
Fibroblasts Dendritic cells Platelets
Dendritic cells

Proatherogenic cellular responses

Endothelial dysfunction Angiogenesis

Endothelial permeability Oxidative stress
Platelet activation Apoptosis
Monocyte recruitment Proteolysis
Vascular remodeling Inflammation

Stable plaque Unstable plaque

Figure 3. Nonhemostatic Actions Triggered by the Tissue Factor and Common Activation Pathways in the Phenotypic
Modulation of the Arterial Wall.
Thrombin, factor Xa, and the tissue factorfactor VIIa complex can activate protease-activated receptors, which are
Draft 4 4/04/11
widely expressed on endothelial cells, leukocytes, vascular smooth-muscle cells, fibroblasts, dendritic cells, and
Author Borissoff
platelets, resulting in a plethora of proatherogenic actions. Gray circles indicate the inactive form
Fig # 3 of a coagulation
protein, and green circles indicate the active form. LDL denotes low-density lipoprotein. Title
and antiinflammatory actions. Thrombin may also genesis, which suggests an important
Artist Knoper role in the
play a role in normal vasomotor regulation.18 pathogenesis of cardiovascularAUTHOR disease. 18 Throm-
Figure has been redrawn and type has been reset
The endothelial decay of thrombomodulin dur- bin, factor Xa, factor XIa, factor IXa, and plas-
Please check carefully

Issue date 5/05/11

ing atherogenesis may allow thrombin to poten- min also show enzymatic activity for cleavage of
tiate atherogenic processes, such as endothelial complement proteins C3 and C5 into their active
dysfunction and barrier disruption, oxidative forms.52 Proteins C3 and C5 are known to in-
stress, apoptosis, inflammation (overexpression duce inflammation and chemotaxis of inflam-
of cytokines or chemokines), activation of plate- matory cells. Human coronary atherosclerotic le-
lets and leukocytes, leukocyte recruitment, mi- sions overexpress anaphylatoxin receptors C3aR
gration and proliferation of VSMCs, and angio- and C5aR, as compared with healthy vessels,

n engl j med 364;18 may 5, 2011 1751

The New England Journal of Medicine
Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

primarily localized on macrophages but also on brand factor and cardiovascular disease have been
endothelial cells, intimal VSMCs, T cells, and mast inconsistent.44,60 More experimental and clinical
cells. Overall, these data establish a new inter- data are needed to clarify these relationships.
face between coagulation and inflammation in
atherosclerosis. Fibrinogen, Fibrin, and Factor XIII
The administration of thrombin-specific in- In clinical studies, there have been strong asso-
hibitors reduces restenosis in rabbits with ath- ciations between increased plasma fibrinogen lev-
erosclerosis after angioplasty.53,54 Another piece els and the risk of cardiovascular disease, which
of evidence for the in vivo relevance of these ef- suggests hyperfibrinogenemia as an independent
fects comes from a study showing that the direct predictor of vascular events.62 Furthermore, the
thrombin inhibitor melagatran reduces athero- distribution of fibrinogen and fibrin degradation
sclerosis progression in apolipoprotein Eknock- products in atherosclerotic lesions during pro-
out mice and promotes plaque stability by inhib- gression has been clearly documented.63,64 Ele-
iting proinflammatory transcription factors and vated levels of plasma fibrinogen, a major de
attenuating the synthesis of matrix metallopro- terminant of the amount of thrombin that is
teinases.55 Furthermore, mice with combined defi- formed,65 are closely related to an enhanced rate
ciency of factor VIII and apolipoprotein E had of coronary-artery calcification and increased in-
significantly less development of atherosclerotic timamedia thickness, both measures of prema-
lesions than control mice, despite having more ture atherosclerosis.66 From a cellular and mo-
pronounced hyperlipidemia.56 In contrast, hyper- lecular perspective, fibrinogen may affect the
coagulability has been linked with atherosclero- plaque phenotype through several distinct mech-
sis progression in murine studies, showing that anisms: favoring the permeability of endothelial
homozygosity for factor V Leiden, a known pro- cells, extracellular accumulation of low-density li-
thrombotic mutation, promotes atherogenesis.57 poprotein (LDL) cholesterol, and the formation of
However, a recent study showed an increase in foam cells; inducing the migration of monocytes
the size of atherosclerotic plaques in procoagu- and VSMCs; increasing platelet reactivity or aggre-
lant mice, indicating that a hypercoagulable state gation; and enhancing inflammation67 (Fig. 3).
contributes to a more stable plaque phenotype.31 Studies in animals have shown distinct results
Overall, these findings suggest that hemostasis on the role of fibrinogen in atherosclerosis, with
exerts various effects on the vasculature and, by some studies indicating that fibrinogen deficien-
the action of distinct regulators, may ultimately cy in transgenic mice is associated with accel
contribute to determining the plaque phenotype. erated atherogenesis in a thrombin-dependent
The clinical evidence in this regard remains manner,68 and others showing that fibrinogen
inconsistent. Despite the fact that prothrombotic deficiency is not a prerequisite for the develop-
genetic variants have not been consistently linked ment of advanced atherosclerotic plaque.69 In-
to the progression of cardiovascular disease in creased plasma levels of d-dimer fragments are
patients,44 clinical data show a positive associa- also associated with enhanced inflammation
tion between markers of thrombin generation and an increased incidence of cardiovascular dis-
and the atherosclerotic plaque burden.58,59 Low ease and are considered a biomarker of athero-
levels of factor VIII have not shown atheropro- thrombosis.70 However, the effect of fibrin deg-
tective effects in patients with hemophilia,44 radation products on the vascular-wall phenotype
whereas there is clinical evidence that elevated is less clear. Although the results of one study
levels of factor VIII promote cardiovascular dis- suggested that d-dimers promote a proathero-
ease.60 In plasma, factor VIII circulates in a com- genic phenotype in human monocytes,71 other
plex with von Willebrand factor, which modulates studies have shown that both fragments D and E
factor VIII activity in the circulation. Since mice may prevent the proliferation of VSMCs in vitro.72
that are deficient in von Willebrand factor have Finally, blood coagulation factor XIII may
significantly fewer atherosclerotic plaques than also be related to atherogenesis. Factor XIII not
control mice, von Willebrand factor may also play only cross-links fibrin chains to fibrin on activa-
a role in atherosclerosis.61 Like the data regard- tion, which contributes to clot stability, but also
ing factor VIII and other coagulation proteases, appears to facilitate the formation of hyperactive
clinical data on the association between von Wille dimers of angiotensin II type 1 receptor, thus

1752 n engl j med 364;18 may 5, 2011

The New England Journal of Medicine

Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
mechanisms of disease

leading to chronic sensitization of circulating

monocytes and exacerbating atherosclerosis.73
Misfolded proteins
Platelet polyphosphates
C on tac t Ac t i vat ion (In t r insic) Anionic surfaces
Path wa y

The contact activation pathway is considered non- Complement

essential for hemostasis in vivo (Fig. 1C and Fig. 4).
However, it may be involved in the pathogenesis Epidermal growth
factor receptor
of arterial thrombosis.14 Although experimental
data have clearly shown that mice deficient in receptor
XII XIIa Angiogenesis
factor XII are protected against arterial thrombo-
sis and stroke,14 in several epidemiologic studies,
data on the association between factor XII and
the risk of cardiovascular disease in humans are
inconsistent.74-76 Although additional research is
needed in this field, the pharmacologic inhibition
of factor XII activation represents a potential ther- Prekallikrein Kallikrein
apeutic target,77,78 considering that hereditary de-
ficiency of factor XII is not associated with bleed-
ing disorders or other pathologic conditions.
At a molecular level, factor XII influences High-molecular-
distinct processes mostly through the plasma kal- weight kininogen Bradykinin
likreinkinin system.79 Factor XIImediated bra-
dykinin formation not only regulates vasodilata-
tion and vascular permeability but also induces Vascular permeability
activation of the complement and fibrinolytic sys-
tems by activating components C3 and C5 and
facilitating the synthesis of tissue-type plasmino- Figure 4. Contact Activation Pathway and Its Proinflammatory COLOR FIGURE

and Proangiogenic Properties. Draft 3 4/04/11

gen activator from endothelial cells, whereas Borissoff
Author processes,
The contact system plays a role in various physiologic such as
kallikrein activates urokinase-type plasminogen Fig # 4
blood-pressure regulation, coagulation, fibrinolysis, angiogenesis, and in-
activator and plasminogen. Platelet-derived inor- Title
flammation. It consists of factor XII, prekallikrein, and high-molecular-weight
ganic polyphosphates80 and misfolded proteins, kininogen. The activation of the proinflammatoryME kallikreinkinin and com-
DE Longo
which are found abundantly in atherosclerotic plement systems is triggered by the proteolytic cleavage
of factor XII (auto-
arteries,81 can also activate factor XII, leading to activation) in reaction to contact with negatively charged artificial or bio-
logic surfaces. The gray circle indicates the inactiveFigure
form of
has been a coagulation
redrawn and type has been reset
kallikrein formation without triggering coagula- protein, and green circles indicate the active form.
Please check carefully

Issue date 5/05/11 with plus

Green circles
tion.82 Levels of tissue kallikrein and plasma signs indicate either positive-feedback reactions or induction of a process.
prekallikrein are associated with the severity of
cardiovascular disease83,84 and have been found
to be critical in the process of vascular repair.85
Given the proangiogenic and proinflammatory (Fig. 5A). Although TFPI is expressed on endo-
nature of factor XII86 and the plasma kallikrein thelial cells, VSMCs, and macrophages in the fi-
kinin system, chronic stimulation of these re- brous cap and shoulder areas of the plaques, it
sponses may promote a proatherogenic intraar- also colocalizes with tissue factor and attenuates
terial environment over time. its activity within atherosclerotic lesions.30,96,97
This finding suggests a role for TFPI not only in
the regulation of tissue factor procoagulant ac-
A n t ic oagul a n t Path wa ys
in Va scul a r Infl a m m at ion tivity but also in the control of tissue factor
induced proatherogenic signaling. The adminis-
Tissue factor pathway inhibitor (TFPI), which is tration of recombinant TFPI has reduced the rates
widely distributed in healthy arterial vessels, tends of inflammation and death in an animal model
to be overexpressed in atherosclerotic lesions87 by decreasing the expression of tumor necrosis

n engl j med 364;18 may 5, 2011 1753

The New England Journal of Medicine
Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

Figure 5 (facing page). Anticoagulant Pathways and Their Nonhemostatic Features.

The regulation of coagulation operates at three levels: inhibition of thrombin, factor Xa, and factor IXa by antithrom-
bin; inhibition of factor Xa, the tissue factorfactor VIIa (TFFVIIa) complex, and hence thrombin formation by tis-
sue factor pathway inhibitor; and proteolytic inactivation of factor V and factor VIII by activated protein C. As shown
in Panel A, antithrombin is a serine protease that inhibits key coagulation enzymes such as thrombin, factor Xa, and
factor IXa. Its action is amplified by as much as 4000 times in the presence of heparin or heparin-like substances,
such as heparan sulfate proteoglycan. Antithrombin has apparent antiinflammatory effects,88 as seen in an increase
in the release of prostacyclin and a decrease in nuclear factor B signaling, which is known to have multiple proinflam-
matory responses. Similar effects have been found after the administration of synthetic direct thrombin inhibitors,
which has contributed to plaque stability in vivo.55 Antithrombin attenuates leukocyte recruitment during inflamma-
tion, which hints at another potential atheroprotective role. Heparin also stimulates the release from endothelial
cells of tissue factor pathway inhibitor, which then binds to factor Xa and the TFFVIIa complex to form an inactive
quaternary complex, thus showing a multitude of antiatherogenic functions. Like antithrombin, heparin cofactor II
has the ability to inactivate thrombin, factor Xa, and factor IXa, whereas the plasma form of heparin cofactor II is an
inefficient inhibitor in the absence of glycosaminoglycans (e.g., heparan sulfate and dermatan sulfate). Heparin co-
factor II is implicated both in vascular remodeling and in atherogenesis. Mice that are deficient in heparin cofactor II
have enhanced intimal hyperplasia after vascular injury. 89 Such mice have increased neointima formation and en-
hanced atherogenesis, as compared with control mice. However, the findings in clinical studies have been inconsis-
tent, with some indicating that heparin cofactor II is a strong predictive marker against atherosclerosis90,91 and one
indicating that its presence is not predictive.92 Protein Z is a cofactor of another protein, named protein Zrelated
protease inhibitor, which inhibits factor Xa and factor XIa in the coagulation cascade. Although the roles of protein
Z and protein Zrelated protease inhibitor in inflammation and the onset of atherosclerosis are poorly understood,
a few clinical trials have shown a significant inverse relationship between levels of these proteins and the clinical se-
verity of atherosclerosis.93-95 As shown in Panel B, thrombin also behaves as an anticoagulant molecule physiologi-
cally. Binding to the endothelial protein C receptor, protein C is transformed into activated protein C by an activation
complex established between thrombin and thrombomodulin. This process is followed by dissociation of activated
protein C from the endothelial protein C receptor and the formation of a complex between activated protein C and
protein S. The latter allows the inactivation of factor Va and factor VIIIa and thus limits further thrombin generation.
Gray circles indicate the inactive form of a coagulation protein, and green circles indicate the active form.

factor (TNF-), chemokines, and myeloperoxi- tion,104,105 whereas low levels of total TFPI are
dase.88 Moreover, TFPI is a potent inhibitor of associated with an increased risk of athero-
matrix metalloproteinases, which are considered thrombosis.106,107
key players in plaque destabilization and athero- In addition to its anticoagulant properties,
thrombotic complications. Decreased TFPI expres- the protein C pathway is known for its protective
sion has been associated with up-regulation of the effects on vascular gene-expression profiles in-
synthesis of matrix metalloproteinases in plaques volving antiapoptotic and antiinflammatory re-
with a vulnerable phenotype. In addition, TFPI sponses, as well as its stabilizing effect on the
has inhibited endothelial migration and angio- endothelial barrier (Fig. 5B).108 Studies of ath-
genesis in mice. Several studies in animals have erosclerosis have shown a substantial down-
shown that TFPI attenuates neointimal hyperpla- regulation of the local expression of endothelial
sia and stenosis but also suppresses the release protein C receptor and thrombomodulin within
of proatherogenic platelet-derived growth factor atherosclerotic vessels, suggesting impaired ac-
BB, CCL2, and matrix metalloproteinase 2.98-101 tivation of protein C and hence a reduced anti-
In agreement with these findings, TFPI-deficient atherogenic response. Several mechanisms, such
mice have significantly more atherosclerotic as enhanced shedding of thrombomodulin from
plaques than control mice,102 whereas vascular- dysfunctional endothelium, an abundance of LDL-
directed TFPI overexpression appears to regulate cholesterol deposits, and local inflammation with-
lipoprotein clearance and temporarily lowers in the arterial wall, may account for the attenu-
plasma cholesterol levels, also reducing athero- ation of the anticoagulant activities of protein C
sclerotic plaque development.103 Clinical data within the atherosclerotic plaque. Overexpression
suggest that plasma TFPI is a marker of endothe- of thrombomodulin has been shown to limit
lial dysfunction; high levels of both free and total neointimal formation in rabbits,109 whereas a
TFPI levels are associated with an increased athero- genetic impairment of the protein Cactivating
sclerotic burden and coronary-artery calcifica- cofactor function of thrombomodulin, resulting

1754 n engl j med 364;18 may 5, 2011

The New England Journal of Medicine

Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
mechanisms of disease

Tissue factor pathway inhibitor
Heparin stimulates release from
endothelial cells
Protein Zrelated Vascular remodeling
protease inhibitor Tissue
and protein Z Inflammation
factor Tissue
factor Lipoprotein clearance

Antithrombin and Xa quaternary Action is amplified 4000 times in the
heparin cofactor II Tissue complex presence of heparin or heparin-like
factor substances (e.g., heparan sulfate or
pathway dermatan sulfate)
Protein Zrelated Inflammation
protease inhibitor Va V
and protein Z
Heparin cofactor II
Prothrombin Thrombin Plasma form requires
glycosaminoglycans (e.g., heparan
sulfate or dermatan sulfate) to
function as an efficient inhibitor
Antithrombin and XIIIa
heparin cofactor II Vascular remodeling

Fibrinogen Fibrin

B Activated protein C
Protein C
Antiinflammatory activity Gene-expression regulation
Antiapoptotic activity Endothelial barrier protection

Thrombomodulin Inactivation of Thrombin

Activated protein C Va and VIIIa generation
Protein S
protein receptor C

Vascular endothelium


Draft 6 4/19/11
in diminished formation of activated protein C, cal studies have confirmed a significant associa-
Author Borissoff
is associated with an increased atherosclerotic tion between circulating low levels of activated
Fig # 5

burden in mice.31 protein C and a greater extent or severity ofTitle

The determinants of soluble levels of throm- erosclerosis.117-119 ME
DE Longo
bomodulin in patients with atherosclerosis are Furthermore, protein S, which has been Artist de- Knoper
poorly understood. The results of various clini- scribed as linking hemostasis, inflammation,Figure and
has been redrawn and type has been reset
cal studies that have examined the relationship apoptosis, forms a complex with the complement Please check carefully
Issue date 5/05/11
between thrombomodulin and the extent of ath- system regulator C4b-binding protein (C4BP),
erosclerotic burden have been inconsistent.110-114 a major inhibitor of the classical complement
In monkeys, progressive atherosclerosis is asso- pathway, localizing it on the surface of apoptotic
ciated with impaired formation of activated pro- cells120 and thus promoting phagocytic activity
tein C, whereas dietary regression of atheroscle- by macrophages.121 Intriguingly, protein S sig-
rosis was found to enhance the anticoagulant nificantly inhibits the expression of macrophage
response.115 Mice with a heterozygous deficiency scavenger receptor A and diminishes the uptake
in protein C have enhanced focal arterial inflam- of acetylated LDL cholesterol mediated by this re-
mation and thrombosis, leading to increased neo- ceptor, resulting in a decreased intracellular lipid
intima formation and localized thrombosis.116 content in macrophages.122 These actions are
In agreement with these findings, several clini- mostly attributable to the ability of protein S to

n engl j med 364;18 may 5, 2011 1755

The New England Journal of Medicine
Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

bind to and induce phosphorylation of the Mer trials has indicated that the use of aspirin is as-
receptor tyrosine kinase. In addition, protein S sociated with a reduction of approximately 30%
plays a role in the protection of the integrity of the in the risk of myocardial infarction, with more
bloodbrain barrier.123 The expression of protein S limited effects on the risk of stroke.135 In addi-
is reduced within atherosclerotic plaques obtained tion to aspirins antiplatelet actions, the efficacy
from patients with unstable angina, as compared of this drug may be due in part to its antiin-
with specimens from patients with stable angi- flammatory actions.136-138 It is difficult to dis-
na.124 Hereditary deficiency of both proteins C sect the contribution of platelets in any of these
and S has been associated with an increased inci- antiinflammatory effects of aspirin. Also, for
dence in arterial thromboembolic events125 and more selective antiplatelet drugs, including clopi-
peripheral-artery disease (Fig. 5B).89-95,126 dogrel, prasugrel, and ticagrelor, which target
platelet receptors, resulting in impaired platelet
F u t ur e Per spec t i v e s activation, antiinflammatory and atherosclero-
sis-delaying effects have been reported.139 How-
Hemostasis is anatomically and functionally en- ever, clinical trials of platelet inhibitors for the
twined with the vasculature. Besides its essential prevention of atherosclerosis progression have not
roles in protecting vascular integrity and main- shown diminished development of plaque with
taining normal blood flow, accumulating data any consistency.140
suggest an intimate cross-talk between hemosta- For many years, oral anticoagulants have been
sis and inflammation, underscoring the role of used for short- and long-term indications. Studies
both systems in many complex diseases, includ- of heparin and vitamin K antagonists have shown
ing atherothrombosis. Intriguingly, numerous that short-term use of these drugs is not likely
studies in animals have also documented that he- to have a major effect on chronic disorders such
mostasis is closely linked to the pathophysiology as atherosclerosis.141,142 Despite the fact that long-
of atherogenesis. Is this association, mostly based term administration of vitamin K antagonists
on experimental data, corroborated by clinical did not have any visible effects on angiographic
data as well? progression in patients who had undergone coro-
The current concept of a vulnerable plaque sug- nary-artery bypass grafting, an additional follow-
gests that repeated plaque microruptures, fol- up assessment 3 years after discontinuation of
lowed by subclinical thrombosis, are critical for therapy showed a significant 35% reduction in
plaque growth and vulnerability.127-129 In agree- overall mortality in the warfarin group.143 Given
ment with these findings, histopathological stud- the powerful effects on risk reduction in throm-
ies showed that two thirds of coronary thrombi botic cardiovascular outcomes, one might specu-
obtained from patients who died suddenly from late that this effect was at least partially medi-
cardiovascular causes were in later stages of matu- ated by effects of vitamin K antagonists on plaque
ration, suggesting that thrombi may exist long phenotype rather than plaque size. At the same
before a rupture occurs.33,34 In addition, the con- time, the principal vascular side effect of the
temporary understanding of atherothrombosis long-term administration of these drugs is ac-
has evolved substantially, establishing new roles celerated calcification. This effect is mainly due
for the hemostatic system beyond thrombosis. to direct inhibition of other vitamin Kdependent
We have summarized the potential array of ac- proteins in the vessel wall, including matrix Gla
tions of hemostasis in relation to the phenotype protein. It is not known whether any additional
of the atherosclerotic vascular wall, presumably influence of inhibition of thrombin formation
linked to plaque stability. But is all of this clini- may occur.144,145 The role of the hemostatic system
cally relevant? in atherosclerosis in humans requires further
Antithrombotic therapy with the use of anti- investigation. Only a handful of molecules rele-
platelet or anticoagulant agents is the key to vant to hemostasis are targeted by existing medi-
atherothrombosis prevention in various clinical cations. As more specific interventions are devel-
situations.130-133 The role of antiplatelet therapy oped, new therapeutic avenues and research
in secondary prevention is no longer questioned, approaches may open up. With the introduction
given the strong overall effect of drugs such as of new oral anticoagulants (e.g., direct inhibitors
aspirin.134 A meta-analysis of primary-prevention of factor Xa and thrombin),146,147 which are small

1756 n engl j med 364;18 may 5, 2011

The New England Journal of Medicine

Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
mechanisms of disease

molecules that can access the vessel wall, it will trasonography has been a major tool in vascular
be possible to document the effects of these drugs imaging. Unfortunately, this approach is charac-
on plaque formation and especially on plaque terized by poor tissue penetration, providing no
stability. Since both thrombin inhibition55 and a information on plaque characteristics, and is sub-
prothrombotic state31 have been suggested as pro- ject to intraobserver and interobserver variability.
moters of plaque stability in atherogenic mice, the With the development of high-resolution magnetic
net effects in humans, if any, are unpredictable. resonance imaging, the assessment of plaque
In conclusion, given the potential of hemosta- characteristics will improve vessel-wall phenotyp-
sis to influence molecular and cellular responses ing as a means of addressing the role of the
in the vasculature, new scientific approaches are hemostatic system in atherosclerosis.
required. Notably, the majority of experimental Supported by a Marie Curie fellowship (MEST-CT-2005-020706)
data are entirely based on quantification of plaque from the European Commission (to Dr. Borissoff).
burden, rather than on extensive phenotyping of Disclosure forms provided by the authors are available with
the full text of this article at
the lesions. This is a major drawback in vascular We thank Jo G.R. De Mey of the Department of Pharmacology
medicine. Furthermore, most clinical studies pre- and Toxicology, Tilman M. Hackeng of the Department of Bio-
dominantly focus on establishing the thrombotic chemistry, and Mat J.A.P. Daemen of the Cardiovascular Re-
search Institute Maastricht all at Maastricht University Medi-
and mortality outcomes, whereas few investigate cal Center for their helpful suggestions during the preparation
plaque progression. During the past decade, ul- of the manuscript.

1. Ross R. Atherosclerosis an inflam- 14. Gailani D, Renne T. Intrinsic pathway and vascular disease. Trends Mol Med
matory disease. N Engl J Med 1999;340: of coagulation and arterial thrombosis. 2008;14:530-8.
115-26. Arterioscler Thromb Vasc Biol 2007;27: 25. Lievens D, Eijgelaar WJ, Biessen EA,
2. Libby P. Inflammation in atheroscle- 2507-13. Daemen MJ, Lutgens E. The multi-func-
rosis. Nature 2002;420:868-74. 15. van der Meijden PE, Munnix IC, Auger tionality of CD40L and its receptor CD40
3. Hansson GK. Inflammation, athero- JM, et al. Dual role of collagen in factor in atherosclerosis. Thromb Haemost 2009;
sclerosis, and coronary artery disease. XII-dependent thrombus formation. Blood 102:206-14.
N Engl J Med 2005;352:1685-95. 2009;114:881-90. 26. Antoniades C, Bakogiannis C, Tou-
4. Levi M, ten Cate H. Disseminated in- 16. Gawaz M, Langer H, May AE. Platelets soulis D, Antonopoulos AS, Stefanadis C.
travascular coagulation. N Engl J Med in inflammation and atherogenesis. J Clin The CD40/CD40 ligand system: linking
1999;341:586-92. Invest 2005;115:3378-84. inflammation with atherothrombosis.
5. Levi M, van der Poll T, Buller HR. Bi- 17. Borensztajn K, Peppelenbosch MP, J Am Coll Cardiol 2009;54:669-77.
directional relation between inflamma- Spek CA. Factor Xa: at the crossroads be- 27. Semple JW, Freedman J. Platelets and
tion and coagulation. Circulation 2004; tween coagulation and signaling in physi- innate immunity. Cell Mol Life Sci 2010;
109:2698-704. ology and disease. Trends Mol Med 2008; 67:499-511.
6. Esmon CT. The interactions between 14:429-40. 28. Tyner JW, Uchida O, Kajiwara N, et al.
inflammation and coagulation. Br J Hae- 18. Borissoff JI, Spronk HM, Heeneman CCL5-CCR5 interaction provides anti-
matol 2005;131:417-30. S, ten Cate H. Is thrombin a key player in apoptotic signals for macrophage survival
7. Dav G, Patrono C. Platelet activation the coagulation-atherogenesis maze? during viral infection. Nat Med 2005;11:
and atherothrombosis. N Engl J Med 2007; Cardiovasc Res 2009;82:392-403. 1180-7.
357:2482-94. 19. Ossovskaya VS, Bunnett NW. Prote- 29. Chiodoni C, Iezzi M, Guiducci C, et al.
8. Furie B, Furie BC. Mechanisms of ase-activated receptors: contribution to Triggering CD40 on endothelial cells con-
thrombus formation. N Engl J Med 2008; physiology and disease. Physiol Rev 2004; tributes to tumor growth. J Exp Med 2006;
359:938-49. 84:579-621. 203:2441-50.
9. Rosenberg RD, Aird WC. Vascular- 20. Massberg S, Brand K, Grner S, et al. 30. Borissoff JI, Heeneman S, Kilinc E, et
bedspecific hemostasis and hypercoagu- A critical role of platelet adhesion in the al. Early atherosclerosis exhibits an en-
lable states. N Engl J Med 1999;340:1555- initiation of atherosclerotic lesion forma- hanced procoagulant state. Circulation
64. tion. J Exp Med 2002;196:887-96. 2010;122:821-30.
10. Ruggeri ZM. Platelets in atherothrom- 21. Huo Y, Schober A, Forlow SB, et al. 31. Seehaus S, Shahzad K, Kashif M, et
bosis. Nat Med 2002;8:1227-34. Circulating activated platelets exacerbate al. Hypercoagulability inhibits monocyte
11. Monroe DM, Hoffman M, Roberts atherosclerosis in mice deficient in apoli- transendothelial migration through pro-
HR. Platelets and thrombin generation. poprotein E. Nat Med 2003;9:61-7. tease-activated receptor-1-, phospholipase-
Arterioscler Thromb Vasc Biol 2002;22: 22. Weber C. Platelets and chemokines in Cbeta-, phosphoinositide 3-kinase-, and
1381-9. atherosclerosis: partners in crime. Circ nitric oxide-dependent signaling in mono-
12. Crawley JT, Zanardelli S, Chion CK, Res 2005;96:612-6. cytes and promotes plaque stability. Cir-
Lane DA. The central role of thrombin in 23. Gleissner CA, von Hundelshausen P, culation 2009;120:774-84.
hemostasis. J Thromb Haemost 2007;5: Ley K. Platelet chemokines in vascular 32. With Not AT, Mathiesen EB, sterud
Suppl 1:95-101. disease. Arterioscler Thromb Vasc Biol B, Amiral J, Vissac AM, Hansen JB. In-
13. Mackman N, Tilley RE, Key NS. Role 2008;28:1920-7. creased thrombin generation in persons
of the extrinsic pathway of blood coagula- 24. Rizvi M, Pathak D, Freedman JE, with echogenic carotid plaques. Thromb
tion in hemostasis and thrombosis. Arterio- Chakrabarti S. CD40-CD40 ligand inter- Haemost 2008;99:602-8.
scler Thromb Vasc Biol 2007;27:1687-93. actions in oxidative stress, inflammation 33. Kramer MC, Rittersma SZ, de Winter

n engl j med 364;18 may 5, 2011 1757

The New England Journal of Medicine
Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

RJ, et al. Relationship of thrombus heal- tients with peripheral arterial disease. Sharrett AR, Chambless LE. Prospective
ing to underlying plaque morphology in Stroke 1996;27:450-4. study of hemostatic factors and incidence
sudden coronary death. J Am Coll Cardiol 47. Green D, Foiles N, Chan C, Kang J, of coronary heart disease: the Atheroscle-
2010;55:122-32. Schreiner PJ, Liu K. An association be- rosis Risk in Communities (ARIC) Study.
34. Rittersma SZ, van der Wal AC, Koch tween clotting factor VII and carotid inti- Circulation 1997;96:1102-8.
KT, et al. Plaque instability frequently oc- ma-media thickness: the CARDIA study. 61. Methia N, Andre P, Denis CV, Econo-
curs days or weeks before occlusive coro- Stroke 2010;41:1417-22. mopoulos M, Wagner DD. Localized re-
nary thrombosis: a pathological throm- 48. Thomas AC, Campbell JH. Targeted duction of atherosclerosis in von Wille-
bectomy study in primary percutaneous delivery of heparin and LMWH using a brand factor-deficient mice. Blood 2001;
coronary intervention. Circulation 2005; fibrin antibody prevents restenosis. Ath- 98:1424-8.
111:1160-5. erosclerosis 2004;176:73-81. 62. Danesh J, Lewington S, Thompson
35. Mackman N. Role of tissue factor in 49. Coughlin SR. Thrombin signalling SG, et al. Plasma fibrinogen level and the
hemostasis, thrombosis, and vascular de- and protease-activated receptors. Nature risk of major cardiovascular diseases and
velopment. Arterioscler Thromb Vasc Biol 2000;407:258-64. nonvascular mortality: an individual par-
2004;24:1015-22. 50. Wang D, Paria BC, Zhang Q, et al. ticipant meta-analysis. JAMA 2005;294:
36. Annex BH, Denning SM, Channon A role for Gab1/SHP2 in thrombin activa- 1799-809.
KM, et al. Differential expression of tis- tion of PAK1: gene transfer of kinase- 63. Bini A, Fenoglio JJ Jr, Mesa-Tejada R,
sue factor protein in directional atherec- dead PAK1 inhibits injury-induced reste- Kudryk B, Kaplan KL. Identification and
tomy specimens from patients with stable nosis. Circ Res 2009;104:1066-75. distribution of fibrinogen, fibrin, and
and unstable coronary syndromes. Circu- 51. Hirano K. The roles of proteinase- fibrin(ogen) degradation products in ath-
lation 1995;91:619-22. activated receptors in the vascular physi- erosclerosis: use of monoclonal antibod-
37. Moreno PR, Bernardi VH, Lopez- ology and pathophysiology. Arterioscler ies. Arteriosclerosis 1989;9:109-21.
Cuellar J, et al. Macrophages, smooth Thromb Vasc Biol 2007;27:27-36. 64. Lepedda AJ, Cigliano A, Cherchi GM,
muscle cells, and tissue factor in unstable 52. Huber-Lang M, Sarma JV, Zetoune FS, et al. A proteomic approach to differenti-
angina: implications for cell-mediated et al. Generation of C5a in the absence of ate histologically classified stable and
thrombogenicity in acute coronary syn- C3: a new complement activation path- unstable plaques from human carotid ar-
dromes. Circulation 1996;94:3090-7. way. Nat Med 2006;12:682-7. teries. Atherosclerosis 2009;203:112-8.
38. Marmur JD, Thiruvikraman SV, Fyfe 53. Chen X, Ren S, Ma MG, et al. Hirulog- 65. Dielis AW, Castoldi E, Spronk HM, et
BS, et al. Identification of active tissue like peptide reduces restenosis and ex- al. Coagulation factors and the protein C
factor in human coronary atheroma. Cir- pression of tissue factor and transform- system as determinants of thrombin gen-
culation 1996;94:1226-32. ing growth factor-beta in carotid artery eration in a normal population. J Thromb
39. Ardissino D, Merlini PA, Ariens R, of atherosclerotic rabbits. Atherosclerosis Haemost 2008;6:125-31.
Coppola R, Bramucci E, Mannucci PM. 2003;169:31-40. 66. Green D, Chan C, Kang J, et al. Longi-
Tissue-factor antigen and activity in human 54. Thome LM, Gimple LW, Bachhuber tudinal assessment of fibrinogen in rela-
coronary atherosclerotic plaques. Lancet BG, et al. Early plus delayed hirudin re- tion to subclinical cardiovascular disease:
1997;349:769-71. duces restenosis in the atherosclerotic the CARDIA study. J Thromb Haemost
40. Toschi V, Gallo R, Lettino M, et al. rabbit more than early administration 2010;8:489-95.
Tissue factor modulates the thromboge- alone: potential implications for dosing 67. de Moerloose P, Boehlen F, Neerman-
nicity of human atherosclerotic plaques. of antithrombin agents. Circulation 1998; Arbez M. Fibrinogen and the risk of
Circulation 1997;95:594-9. 98:2301-6. thrombosis. Semin Thromb Hemost 2010;
41. Ardissino D, Merlini PA, Bauer KA, et 55. Bea F, Kreuzer J, Preusch M, et al. 36:7-17.
al. Thrombogenic potential of human Melagatran reduces advanced atheroscle- 68. Iwaki T, Sandoval-Cooper MJ, Brech-
coronary atherosclerotic plaques. Blood rotic lesion size and may promote plaque mann M, Ploplis VA, Castellino FJ. A fibrin-
2001;98:2726-9. stability in apolipoprotein E-deficient ogen deficiency accelerates the initiation
42. Monroe DM, Key NS. The tissue fac- mice. Arterioscler Thromb Vasc Biol 2006; of LDL cholesterol-driven atherosclerosis
tor-factor VIIa complex: procoagulant 26:2787-92. via thrombin generation and platelet acti-
activity, regulation, and multitasking. 56. Khallou-Laschet J, Caligiuri G, Tupin vation in genetically predisposed mice.
J Thromb Haemost 2007;5:1097-105. E, et al. Role of the intrinsic coagulation Blood 2006;107:3883-91.
43. Tilley RE, Pedersen B, Pawlinski R, et pathway in atherogenesis assessed in 69. Xiao Q, Danton MJ, Witte DP, Kowala
al. Atherosclerosis in mice is not affected hemophilic apolipoprotein E knockout MC, Valentine MT, Degen JL. Fibrinogen
by a reduction in tissue factor expression. mice. Arterioscler Thromb Vasc Biol 2005; deficiency is compatible with the develop-
Arterioscler Thromb Vasc Biol 2006;26: 25:e123-e126. ment of atherosclerosis in mice. J Clin
555-62. 57. Eitzman DT, Westrick RJ, Shen Y, et al. Invest 1998;101:1184-94.
44. Konings J, Govers-Riemslag JWP, ten Homozygosity for factor V Leiden leads 70. Folsom AR. Hemostatic risk factors
Cate H. Novel insights into genetics of to enhanced thrombosis and athero for atherothrombotic disease: an epide-
arterial thrombosis. In: Baars HF, ed. sclerosis in mice. Circulation 2005;111: miologic view. Thromb Haemost 2001;86:
Clinical cardiogenetics. New York: Spring- 1822-5. 366-73.
er, 2010. 58. Di Tullio MR, Homma S, Jin Z, Sacco 71. Zhou D, Yang PY, Zhou B, Rui YC. Fi-
45. Gertow K, Amato M, Werba JP, et al. RL. Aortic atherosclerosis, hypercoagula- brin D-dimer fragments enhance inflam-
Tissue factor gene promoter haplotype as- bility, and stroke the APRIS (Aortic Plaque matory responses in macrophages: role in
sociates with carotid intima-media thick- and Risk of Ischemic Stroke) study. J Am advancing atherosclerosis. Clin Exp Phar-
ness in subjects in cardiovascular risk Coll Cardiol 2008;52:855-61. macol Physiol 2007;34:185-90.
prevention. Atherosclerosis 2009;207:168- 59. Pramo JA, Orbe J, Beloqui O, et al. 72. Ishida T, Tanaka K. Effects of fibrin
73. Prothrombin fragment 1+2 is associated and fibrinogen-degradation products on
46. Cortellaro M, Baldassarre D, Cofran- with carotid intima-media thickness in the growth of rabbit aortic smooth mus-
cesco E, et al. Relation between hemo- subjects free of clinical cardiovascular cle cells in culture. Atherosclerosis 1982;
static variables and increase of common disease. Stroke 2004;35:1085-9. 44:161-74.
carotid intima-media thickness in pa- 60. Folsom AR, Wu KK, Rosamond WD, 73. Abdalla S, Lother H, Langer A, el Far-

1758 n engl j med 364;18 may 5, 2011

The New England Journal of Medicine

Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
mechanisms of disease

amawy Y, Quitterer U. Factor XIIIA trans- 87. Crawley J, Lupu F, Westmuckett AD, Natt JM, Willerson JT. Local gene transfer
glutaminase crosslinks AT1 receptor Severs NJ, Kakkar VV, Lupu C. Expres- of tissue factor pathway inhibitor regu-
dimers of monocytes at the onset of sion, localization, and activity of tissue lates intimal hyperplasia in atheroscle-
atherosclerosis. Cell 2004;119:343-54. factor pathway inhibitor in normal and rotic arteries. Proc Natl Acad Sci U S A
74. Zito F, Lowe GD, Rumley A, McMahon atherosclerotic human vessels. Arterio- 2001;98:4078-83.
AD, Humphries SE. Association of the scler Thromb Vasc Biol 2000;20:1362-73. 101. Kopp CW, Hlzenbein T, Steiner S,
factor XII 46C>T polymorphism with risk 88. Okajima K. Regulation of inflamma- et al. Inhibition of restenosis by tissue
of coronary heart disease (CHD) in the tory responses by natural anticoagulants. factor pathway inhibitor: in vivo and in
WOSCOPS study. Atherosclerosis 2002; Immunol Rev 2001;184:258-74. vitro evidence for suppressed monocyte
165:153-8. 89. Aihara K, Azuma H, Akaike M, et al. chemoattraction and reduced gelatinolyt-
75. Govers-Riemslag JW, Smid M, Cooper Strain-dependent embryonic lethality and ic activity. Blood 2004;103:1653-61.
JA, et al. The plasma kallikrein-kinin sys- exaggerated vascular remodeling in hepa- 102. Westrick RJ, Bodary PF, Xu Z, Shen
tem and risk of cardiovascular disease in rin cofactor II-deficient mice. J Clin Invest YC, Broze GJ, Eitzman DT. Deficiency of
men. J Thromb Haemost 2007;5:1896-903. 2007;117:1514-26. tissue factor pathway inhibitor promotes
76. Siegerink B, Govers-Riemslag JW, 90. Takamori N, Azuma H, Kato M, et al. atherosclerosis and thrombosis in mice.
Rosendaal FR, ten Cate H, Algra A. In- High plasma heparin cofactor II activity Circulation 2001;103:3044-6.
trinsic coagulation activation and the risk is associated with reduced incidence of 103. Pan S, White TA, Witt TA, Chiriac A,
of arterial thrombosis in young women: in-stent restenosis after percutaneous cor- Mueske CS, Simari RD. Vascular-directed
results from the Risk of Arterial Throm- onary intervention. Circulation 2004;109: tissue factor pathway inhibitor overex-
bosis in relation to Oral contraceptives 481-6. pression regulates plasma cholesterol and
(RATIO) case-control study. Circulation 91. Aihara K, Azuma H, Takamori N, et reduces atherosclerotic plaque develop-
2010;122:1854-61. al. Heparin cofactor II is a novel protec- ment. Circ Res 2009;105:713-20.
77. Hagedorn I, Schmidbauer S, Pleines I, tive factor against carotid atherosclerosis 104. Sakata T, Mannami T, Baba S, et al.
et al. Factor XIIa inhibitor recombinant in elderly individuals. Circulation 2004; Potential of free-form TFPI and PAI-1 to
human albumin Infestin-4 abolishes oc- 109:2761-5. be useful markers of early atherosclerosis
clusive arterial thrombus formation with- 92. Giri TK, Ahn CW, Wu KK, Tollefsen in a Japanese general population (the
out affecting bleeding. Circulation 2010; DM. Heparin cofactor II levels do not pre- Suita Study): association with the intimal-
121:1510-7. dict the development of coronary heart medial thickness of carotid arteries. Ath-
78. Gailani D, Renn T. The intrinsic disease: the Atherosclerosis Risk in Com- erosclerosis 2004;176:355-60.
pathway of coagulation: a target for treat- munities (ARIC) study. Arterioscler Thromb 105. Mitchell CT, Kamineni A, Palmas W,
ing thromboembolic disease? J Thromb Vasc Biol 2005;25:2689-90. Cushman M. Tissue factor pathway in-
Haemost 2007;5:1106-12. 93. Sofi F, Cesari F, Pratesi G, et al. Low hibitor, vascular risk factors and subclini-
79. Schmaier AH. The elusive physiologic protein Z levels in patients with periph- cal atherosclerosis: the Multi-Ethnic Study
role of factor XII. J Clin Invest 2008;118: eral arterial disease. Thromb Haemost of Atherosclerosis. Atherosclerosis 2009;
3006-9. 2007;98:1114-7. 207:277-83.
80. Mller F, Mutch NJ, Schenk WA, et al. 94. Pardos-Gea J, Ordi-Ros J, Serrano S, 106. Blann AD, Amiral J, McCollum CN,
Platelet polyphosphates are proinflam- Balada E, Nicolau I, Vilardell M. Protein Z Lip GY. Differences in free and total tis-
matory and procoagulant mediators in levels and anti-protein Z antibodies in pa- sue factor pathway inhibitor, and tissue
vivo. Cell 2009;139:1143-56. tients with arterial and venous thrombo- factor in peripheral artery disease com-
81. Rcken C, Tautenhahn J, Bhling F, et sis. Thromb Res 2008;121:727-34. pared to healthy controls. Atherosclerosis
al. Prevalence and pathology of amyloid 95. Sofi F, Cesari F, Tu Y, et al. Protein Z- 2000;152:29-34.
in atherosclerotic arteries. Arterioscler dependent protease inhibitor and protein 107. Massberg S, Grahl L, von Bruehl ML,
Thromb Vasc Biol 2006;26:676-7. Z in peripheral arterial disease patients. et al. Reciprocal coupling of coagulation
82. Maas C, Govers-Riemslag JW, Bouma J Thromb Haemost 2009;7:731-5. and innate immunity via neutrophil ser-
B, et al. Misfolded proteins activate factor 96. Caplice NM, Mueske CS, Kleppe LS, ine proteases. Nat Med 2010;16:887-96.
XII in humans, leading to kallikrein for- Simari RD. Presence of tissue factor path- 108. Mosnier LO, Zlokovic BV, Griffin JH.
mation without initiating coagulation. way inhibitor in human atherosclerotic The cytoprotective protein C pathway.
J Clin Invest 2008;118:3208-18. plaques is associated with reduced tissue Blood 2007;109:3161-72.
83. Merlo C, Wuillemin WA, Redondo M, factor activity. Circulation 1998;98:1051-7. 109. Waugh JM, Li-Hawkins J, Yuksel E, et
et al. Elevated levels of plasma prekalli- 97. Badimon JJ, Lettino M, Toschi V, et al. al. Thrombomodulin overexpression to
krein, high molecular weight kininogen Local inhibition of tissue factor reduces limit neointima formation. Circulation
and factor XI in coronary heart disease. the thrombogenicity of disrupted human 2000;102:332-7.
Atherosclerosis 2002;161:261-7. atherosclerotic plaques: effects of tissue 110. Gerdes VE, Kremer Hovinga JA, ten
84. Porcu P, Emanueli C, Desortes E, et al. factor pathway inhibitor on plaque throm- Cate H, Brandjes DP, Bller HR. Soluble
Circulating tissue kallikrein levels corre- bogenicity under flow conditions. Circu- thrombomodulin in patients with estab-
late with severity of carotid atherosclero- lation 1999;99:1780-7. lished atherosclerosis. J Thromb Haemost
sis. Arterioscler Thromb Vasc Biol 2004; 98. Jang Y, Guzman LA, Lincoff AM, et al. 2004;2:200-1.
24:1104-10. Influence of blockade at specific levels of 111. Peter K, Nawroth P, Conradt C, et al.
85. Stone OA, Richer C, Emanueli C, et al. the coagulation cascade on restenosis in a Circulating vascular cell adhesion mole-
Critical role of tissue kallikrein in vessel rabbit atherosclerotic femoral artery inju- cule-1 correlates with the extent of human
formation and maturation: implications ry model. Circulation 1995;92:3041-50. atherosclerosis in contrast to circulating
for therapeutic revascularization. Arterio- 99. Oltrona L, Speidel CM, Recchia D, intercellular adhesion molecule-1, E-selec
scler Thromb Vasc Biol 2009;29:657-64. Wickline SA, Eisenberg PR, Abendschein tin, P-selectin, and thrombomodulin.
86. LaRusch GA, Mahdi F, Shariat-Madar DR. Inhibition of tissue factor-mediated Arterioscler Thromb Vasc Biol 1997;17:
Z, et al. Factor XII stimulates ERK1/2 and coagulation markedly attenuates stenosis 505-12.
Akt through uPAR, integrins, and the after balloon-induced arterial injury in 112. Salomaa V, Matei C, Aleksic N, et al.
EGFR to initiate angiogenesis. Blood 2010; minipigs. Circulation 1997;96:646-52. Soluble thrombomodulin as a predictor of
115:5111-20. 100. Zoldhelyi P, Chen ZQ, Shelat HS, Mc- incident coronary heart disease and

n engl j med 364;18 may 5, 2011 1759

The New England Journal of Medicine
Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.
mechanisms of disease

symptomless carotid artery atherosclero- 124. Randi AM, Biguzzi E, Falciani F, et effects of simvastatin and low-dose aspi-
sis in the Atherosclerosis Risk in Com- al. Identification of differentially ex- rin. Circulation 2003;108:1673-5.
munities (ARIC) Study: a case-cohort pressed genes in coronary atherosclerotic 137. Chiang N, Hurwitz S, Ridker PM,
study. Lancet 1999;353:1729-34. plaques from patients with stable or un- Serhan CN. Aspirin has a gender-depen-
113. Wu KK, Aleksic N, Ballantyne CM, stable angina by cDNA array analysis. dent impact on antiinflammatory 15-epi-
Ahn C, Juneja H, Boerwinkle E. Interac- J Thromb Haemost 2003;1:829-35. lipoxin A4 formation: a randomized hu-
tion between soluble thrombomodulin 125. Mahmoodi BK, Brouwer JL, Veeger man trial. Arterioscler Thromb Vasc Biol
and intercellular adhesion molecule-1 in NJ, van der Meer J. Hereditary deficiency 2006;26(2):e14-e17.
predicting risk of coronary heart disease. of protein C or protein S confers in- 138. Morris T, Stables M, Hobbs A, et al.
Circulation 2003;107:1729-32. creased risk of arterial thromboembolic Effects of low-dose aspirin on acute in-
114. Aleksic N, Wang YW, Ahn C, Juneja events at a young age: results from a large flammatory responses in humans. J Im-
HS, Folsom AR, Wu KK. Assessment of family cohort study. Circulation 2008;118: munol 2009;183:2089-96.
coronary heart disease risk by combined 1659-67. 139. Muhlestein JB. Effect of antiplatelet
analysis of coagulation factors. Athero- 126. Cho YP, Kwon TW, Ahn JH, et al. therapy on inflammatory markers in ath-
sclerosis 2008;198:294-300. Protein C and/or S deficiency presenting erothrombotic patients. Thromb Hae-
115. Lentz SR, Miller FJ Jr, Piegors DJ, et as peripheral arterial insufficiency. Br J most 2010;103:71-82.
al. Anticoagulant responses to thrombin Radiol 2005;78:601-5. 140. Dieker HJ, French JK, Joziasse IC, et
are enhanced during regression of athero- 127. Mann J, Davies MJ. Mechanisms of al. Antiplatelet therapy and progression
sclerosis in monkeys. Circulation 2002; progression in native coronary artery dis- of coronary artery disease: a placebo-con-
106:842-6. ease: role of healed plaque disruption. trolled trial with angiographic and clini-
116. Castellino FJ, Ganopolsky JG, Noria Heart 1999;82:265-8. cal follow-up after myocardial infarction.
F, Sandoval-Cooper MJ, Ploplis VA. Focal 128. Burke AP, Kolodgie FD, Farb A, et al. Am Heart J 2007;153(1):66.e1-66.e8.
arterial inflammation is augmented in Healed plaque ruptures and sudden coro- 141. The Post Coronary Artery Bypass
mice with a deficiency of the protein C nary death: evidence that subclinical rup- Graft Trial Investigators. The effect of ag-
gene. Thromb Haemost 2006;96:794-801. ture has a role in plaque progression. gressive lowering of low-density lipopro-
117. Salomaa V, Matei C, Aleksic N, et al. Circulation 2001;103:934-40. tein cholesterol levels and low-dose anti-
Cross-sectional association of soluble 129. Finn AV, Nakano M, Narula J, Kolod- coagulation on obstructive changes in
thrombomodulin with mild peripheral gie FD, Virmani R. Concept of vulnerable/ saphenous-vein coronary-artery bypass
artery disease: the ARIC study. Athero- unstable plaque. Arterioscler Thromb Vasc grafts. N Engl J Med 1997;336:153-62.
sclerosis 2001;157:309-14. Biol 2010;30:1282-92. [Erratum, N Engl J Med 1997;337:1859.]
118. Zorio E, Navarro S, Medina P, et al. 130. Holmes DR Jr, Kereiakes DJ, Klei- 142. Byington RP, Evans GW, Espeland
Circulating activated protein C is reduced man NS, Moliterno DJ, Patti G, Grines CL. MA, et al. Effects of lovastatin and warfa-
in young survivors of myocardial infarc- Combining antiplatelet and anticoagulant rin on early carotid atherosclerosis: sex-
tion and inversely correlates with the se- therapies. J Am Coll Cardiol 2009;54:95- specific analyses. Circulation 1999;100(3):
verity of coronary lesions. J Thromb Hae- 109. e14-e17.
most 2006;4:1530-6. 131. Bhatt DL, Fox KA, Hacke W, et al. 143. Knatterud GL, Rosenberg Y,
119. Matsumoto K, Yano Y, Gabazza EC, Clopidogrel and aspirin versus aspirin Campeau L, et al. Long-term effects on
et al. Inverse correlation between activat- alone for the prevention of atherothrom- clinical outcomes of aggressive lowering
ed protein C generation and carotid ath- botic events. N Engl J Med 2006;354:1706- of low-density lipoprotein cholesterol lev-
erosclerosis in Type 2 diabetic patients. 17. els and low-dose anticoagulation in the
Diabet Med 2007;24:1322-8. 132. The Warfarin Antiplatelet Vascular post coronary artery bypass graft trial.
120. Webb JH, Blom AM, Dahlbck B. Vi- Evaluation Trial Investigators. Oral anti- Circulation 2000;102:157-65.
tamin K-dependent protein S localizing coagulant and antiplatelet therapy and 144. Spronk HM, Soute BA, Schurgers LJ,
complement regulator C4b-binding pro- peripheral arterial disease. N Engl J Med Thijssen HH, De Mey JG, Vermeer C. Tis-
tein to the surface of apoptotic cells. J Im- 2007;357:217-27. sue-specific utilization of menaquinone-4
munol 2002;169:2580-6. 133. Schulman S. Care of patients receiv- results in the prevention of arterial calci-
121. Anderson HA, Maylock CA, Wil- ing long-term anticoagulant therapy. fication in warfarin-treated rats. J Vasc
liams JA, Paweletz CP, Shu H, Shacter E. N Engl J Med 2003;349:675-83. Res 2003;40:531-7.
Serum-derived protein S binds to phos- 134. Antithrombotic Trialists Collabora- 145. Rennenberg RJ, van Varik BJ, Schur
phatidylserine and stimulates the phago- tion. Collaborative meta-analysis of ran- gers LJ, et al. Chronic coumarin treatment
cytosis of apoptotic cells. Nat Immunol domised trials of antiplatelet therapy for is associated with increased extracoro-
2003;4:87-91. prevention of death, myocardial infarc- nary arterial calcification in humans.
122. Liao D, Wang X, Li M, Lin PH, Yao Q, tion, and stroke in high risk patients. BMJ Blood 2010;115:5121-3.
Chen C. Human protein S inhibits the up- 2002;324:71-86. [Erratum, BMJ 2002;324: 146. Schulman S, Kearon C, Kakkar AK,
take of AcLDL and expression of SR-A 141.] et al. Dabigatran versus warfarin in the
through Mer receptor tyrosine kinase in hu- 135. Patrono C, Garca Rodrguez LA, treatment of acute venous thromboembo-
man macrophages. Blood 2009;113:165-74. Landolfi R, Baigent C. Low-dose aspirin lism. N Engl J Med 2009;361:2342-52.
123. Zhu D, Wang Y, Singh I, et al. Protein for the prevention of atherothrombosis. 147. Connolly SJ, Ezekowitz MD, Yusuf S,
S controls hypoxic/ischemic blood-brain N Engl J Med 2005;353:2373-83. et al. Dabigatran versus warfarin in pa-
barrier disruption through the TAM re- 136. Ferroni P, Martini F, Cardarello CM, tients with atrial fibrillation. N Engl J
ceptor Tyro3 and sphingosine 1-phos- Gazzaniga PP, Davi G, Basili S. Enhanced Med 2009;361:1139-51.
phate receptor. Blood 2010;115:4963-72. interleukin-1beta in hypercholesterolemia: Copyright 2011 Massachusetts Medical Society.

1760 n engl j med 364;18 may 5, 2011

The New England Journal of Medicine

Downloaded from on June 7, 2011. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.