| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2000;101:1785.)
© 2000 American Heart Association, Inc.
Clinical Investigation and Reports |
and Lipopolysaccharide Potentiate Monocyte Tissue Factor Induction by C-Reactive Protein
From the School of Pathology, University of New South Wales (A.N., C.L.G.), and the Department of Cardiology, Concord Hospital, University of Sydney, Australia (S.B.F).
Correspondence to Carolyn Geczy, School of Pathology, University New South Wales, NSW 2052, Australia. E-mail c.geczy{at}unsw.edu.au
| Abstract |
|---|
|
|
|---|
Methods and ResultsPeripheral blood mononuclear
cells (PBMCs) from 79 healthy men and women aged 26 to 83 years and 21
healthy postmenopausal women taking hormone replacement therapy (HRT)
were stimulated with CRP, lipopolysaccharide (LPS),
interferon-
(IFN), or their combination. Levels of CRP in the normal
range (1 to 5 µg/mL) increased basal monocyte TF 4- to 6-fold and
40-fold at higher concentrations (25 µg/mL). Coincubation of LPS with
CRP produced a greater-than-additive response. IFN did not induce TF
but synergized with CRP to approximately double activity. There was a
striking positive correlation between age and monocyte TF induction,
with a dramatic rise on monocytes from postmenopausal women that was
not apparent on cells from women taking HRT.
ConclusionsSynergy between CRP and inflammatory mediators may play a direct prothrombotic role in the pathogenesis of coronary atherosclerosis and its acute complications by increasing monocyte/macrophage TF. This may contribute to age and sex differences in coronary events and to the protective effects of HRT.
Key Words: inflammation coagulation immune system atherosclerosis coronary disease
| Introduction |
|---|
|
|
|---|
CRP is secreted by hepatocytes in response to interleukin (IL)-6, a cytokine associated with raised CRP levels in unstable angina.10 CRP accumulates in macrophage-rich regions in developing atherosclerotic lesions11 and upregulates some macrophage proinflammatory cytokines.12 13 CRP also induces monocyte tissue factor (TF),14 15 but it is uncertain whether an elevated CRP level is simply a marker of ongoing inflammation to oxidized lipids or unidentified environmental agents or whether CRP has direct prothrombotic effects in atherosclerosis and its complications.8 16
TF, a potent activator of the extrinsic coagulation cascade, is considered to play an important role in atherosclerosis. It is expressed on macrophages in atherosclerotic plaque17 18 19 20 and may contribute to acute thrombotic events associated with plaque rupture in unstable syndromes.19 20 CD4+ T lymphocytes occur in almost all stages of atherosclerosis and produce interferon (IFN), which may contribute to lesion formation.21 22 IFN does not directly induce TF in blood monocytes23 24 but synergizes strongly with lipopolysaccharide (LPS) to induce TF on inflammatory macrophages.24
Here, we show that IFN and LPS potentiate monocyte TF induction by CRP and report a remarkable relationship between TF induction and both sex and age, as well as a significant protective effect of hormone replacement therapy (HRT) in postmenopausal women. This may represent an important amplification mechanism for triggering coagulation in inflamed plaques that underlie unstable coronary syndromes.
| Methods |
|---|
|
|
|---|
was from Bender & Co.
LPS (Escherichia coli 055: B5) was from Difco; human CRP
(>90% pure) and Hanks balanced salt solution (HBSS) were from
Sigma; RPMI 1640 was from GIBCO; and human factors VII, X, and Xa,
factor Xa chromogenic substrate, murine MAb against human
TF (HTF-1), and Spectrozyme LAL (Limulus amebocyte lysate) were from
American Diagnostica Inc. IgG1
isotype control antibody was from ICN ImmunoBiologicals. Pyrosol LAL
buffer was from Associates of Cape Cod, Inc, and endotoxin-tested
Lymphoprep was from Nycomed Pharma AS. CRP (selected after pretesting
from 3 different sources), cytokines and media tested for
endotoxin by the Limulus assay25 contained <5 pg of
endotoxin per milliliter.
Subjects
Peripheral venous citrated blood (30 mL) was
collected from 79 healthy subjects (42 men and 37 women aged 26 to 83
years) and 21 healthy women (aged 50 to 74 years) taking HRT. Subjects
with histories of connective tissue disease, cancer, acute infection,
ischemic heart disease,
hypercholesterolemia, hypertension, or
diabetes, those taking medication (particularly aspirin or
lipid-lowering drugs), and ex-smokers or current smokers were excluded.
Subjects were grouped according to age and sex: group 1 consisted of
men <50 years of age; group 2, premenopausal women <50 years old;
group 3, men >50 years old; and group 4, postmenopausal women >50
years old. Group 5 (postmenopausal women >50 years old taking HRT and
attending a menopause clinic) included 9 women with prior hysterectomy
and 3 with prior hysterectomy plus oophorectomy. Five took oral
conjugated estrogen (Premarin, 1.25 mg/d), and 16 had a 50-mg estradiol
subcutaneous implant. In 15 women, estrogen was unopposed; 6 took
additional medroxyprogesterone (5 to 10 mg/d). In
11 women, testosterone undecanoate (40 mg/d orally in 3 and 100 mg
subcutaneous implant in 8) was also prescribed to improve well-being
and libido.26 27 Duration of HRT was 6 to 48 months (mean
22±12 months). 17ß-Estradiol levels ranged from 99 to 859 pmol/L
(mean±SD 495±233).
Measurement of Procoagulant
PBMCs were obtained by gradient
centrifugation on Lymphoprep as described
previously.25 In some experiments, we enriched monocytes
(
95% pure by nonspecific esterase staining) by incubating PBMCs for
2 hours at 37°C on serum-coated wells and harvesting the lymphocytes.
Monocyte numbers (5% to 10% of PBMCs) did not vary significantly
between groups. PBMCs in serum-free RPMI were incubated at 37°C (5%
CO2 in air) in 96-well plates (Nunc;
5x105 cells/250 µL) with stimulants at the
doses indicated. In agreement with a previous report for
CRP,15 activity induced by CRP±IFN was obvious after 4
hours, was optimal between 12 to 16 hours, and then declined. PBMCs
were routinely cultured for 16 hours, washed twice with RPMI 1640 to
remove lymphocytes, and tested after 3 cycles of freezing at -80°C
and thawing at 37°C. The 1-stage plasma recalcification assay was
performed as described previously,25 and activity was
expressed as milliunits (mU) of TF/106 cells by
comparison with human brain powder as TF standard.25
Factor Xa generated after addition of factors VII and X was measured on viable cells with factor Xaspecific chromogenic substrate as described previously.25 TF procoagulant was confirmed by incubation of PBMCs (106) with control IgG1, or anti-TF MAb (20 µg/0.5 mL) for 1 hour on ice; factor VII was then added (50 µg/50 µL HBSS plus 3 mmol/L CaCl2 plus 0.1% ovalbumin), and factor Xa generation was measured.
Statistical Analysis
Data are expressed as mean±SEM. Statistical
analyses were performed with ANOVA, and the unpaired
t test was used to compare TF levels on PBMCs between
groups. Pearson correlation coefficient analysis was used to
assess associations between values determined to be normally
distributed. The relationship between age and procoagulant was
analyzed by linear or nonlinear regression, and models were
compared with r2 and F values,
inspection of residual plots, and an F ratio test (linear versus
quadratic). Women taking HRT were significantly younger than women not
taking HRT (mean age 57±7 versus 66±9 years, respectively;
P<0.01), and the independent contributions of HRT and age
were determined by stepwise multiple linear regression
analysis. The influence of additional testosterone use was
tested in a model that included age and HRT. Model
R2 was adjusted for sample size. A
value of P<0.05 was considered statistically
significant.
| Results |
|---|
|
|
|---|
4-fold (233±44 mU), and basal procoagulant
activity was enhanced 13-fold with 10 µg/mL CRP (670±130 mU), levels
greater than those induced by 1 ng/mL LPS (440±58 mU). When LPS and
CRP were combined, activity was somewhat greater than with each
stimulant alone (Table 1
|
IFN did not induce procoagulant activity and did not amplify the
LPS-provoked response, but when combined with increasing concentrations
of CRP, synergy was observed (Table 1
and Figure 2
),
particularly with
10 µg/mL of CRP (P<0.001 compared
with unstimulated peripheral blood mononuclear cells
[PBMCs]). Tumor necrosis factor provoked a weak response but did not
synergize with CRP (not shown). Activity on monocytes depleted of
lymphocytes and stimulated with CRP with or without LPS was only half
that expressed by total PBMCs, and after CRP+IFN stimulation, it was
only 25% of the unfractionated population.
|
Factor Xa generation was dependent on factors VII and X, and the magnitude of responses corresponded to those measured by plasma recalcification. Factor Xa generated by viable cells after stimulation with CRP more than doubled when coincubated with IFN (from 89±31 to 176±86 ng FXa/106 PBMC, n=6). Activity was inhibited by >90% with a neutralizing anti-TF monoclonal antibody (MAb), confirming the major contribution of surface-associated TF to the procoagulant response. TF antigen, localized by flow cytometry, was only expressed on monocytes (CD14+), whereas lymphocytes (CD14-) were negative (not shown).
The requirement for new RNA and protein synthesis for TF induction was confirmed by reduction of activity to basal levels when PBMCs were stimulated with CRP±LPS/IFN in the presence of 5 µg/mL cycloheximide or actinomycin D, respectively.
TF Induction Is Influenced by Sex and Age
Basal activity on monocytes from individuals >50 years old (Table 1
, groups 3 and 4, and Figure 1
, A and B) was low (71 to 75 mU)
but was higher than on monocytes from young subjects; young women
(group 2) expressed the lowest levels. Responses to 10 pg/mL LPS were
significantly different between groups, and activity of cells from men
<50 years old was about half that of men >50 years old (Table 1
and Figure 1C
). Dramatic increases were observed with
PBMCs from postmenopausal women (group 5): net stimulation by LPS was
significantly greater (690±113 mU) than TF induced on cells from
groups 1, 2, or 3, particularly from women <50 years old (87±20 mU).
Figure 1A
and 1B
shows the rises in basal TF with
increasing age, particularly on monocytes from women
(r=0.83, P<0.001 for women; r=0.62,
P<0.001 for men). Age-related responses to LPS
(r=0.44, P=0.007 for men; r=0.64,
P<0.001 for women; not shown) and CRP (r=0.44,
P<0.007 for men; r=0.76, P<0.001 for
women; not shown) followed a similar trend.
|
TF induced by CRP plus either LPS or IFN was also greatest on cells
from older individuals (Table 1
), particularly postmenopausal
women (>5000 mU), and showed high correlation with increasing age
(r=0.83 for CRP+LPS; r=0.84 for CRP+IFN; Figure 2
, D and F). Six of 13 women >60 years
old showed activity >7000 mU; responses of PBMCs from men in this age
range never exceeded this level (Figure 1
, C and E). Although
significant, correlations between age and activity on monocytes from
men (r=0.54 for CRP+LPS, Figure 1C
; r=0.50
for CRP+IFN, Figure 1E
) were consistently less than for
women. There were no significant differences between linear and
nonlinear fits in results from men and women, although in women,
residual plots revealed signs of curvature consistent with a
quadratic or exponential fit.
HRT Reduces Monocyte Responsiveness
Basal or induced TF on monocytes from women treated with estrogen
did not differ significantly from that on monocytes from women also
taking progestogen or additional testosterone, and therefore results in
group 5 were pooled. No associations between reduced TF induction and
duration of treatment (up to 4 years) or plasma levels of
17ß-estradiol were obvious. Basal TF on monocytes from the HRT group
was significantly less (P=0.016) than on cells from
postmenopausal women, and activity induced by CRP (1556±106 mU) was
approximately half that of untreated women (group 4, 3070±318 mU;
Table 2
). Multiple linear regression
analysis confirmed that differences in mean ages did not
account for the reduced TF in the HRT group. The partial
R2 for HRT in the CRP+LPS/IFN
stimulation groups was of the same order of magnitude as for
age, with P<0.001 and coefficients >2000 mU, which
indicates a large effect of HRT after adjustment for age (Table 2
). Interestingly, mean responses of PBMCs to CRP±LPS/IFN from
women taking HRT approached levels in men aged <50 years (group 1,
Table 1
). Monocytes from 5 of 21 HRT subjects exhibited <2000
mU of TF (Figure 1
, D and F), well within the range of activity
induced by the combination stimulus on monocytes from women aged <50
years (Table 1
). Testosterone use in women taking HRT did not
influence these findings: when added to the model that included age and
HRT, testosterone use was not an independent predictor of TF activity
after any stimulation (all P>0.4), and a separate
multivariate analysis that excluded women
taking testosterone yielded almost identical results to the original
analysis, with the same patterns of significance.
|
PBMCs from women aged >50 years were sensitive to as little as 5
µg/mL CRP (Figure 2A
); TF activity induced by 10 µg/mL was
some 3-fold more than that expressed by monocytes from younger women or
those taking HRT. The trend was more obvious when increasing doses of
CRP were combined with IFN (Figure 2B
). PBMCs from men generated
less obvious age-related responses with low concentrations of CRP (5
µg/mL), although coincubation of this concentration with IFN induced
3-fold more TF on cells from men aged >50 years. High CRP
concentrations generated greater differences on monocytes from men of
different ages (Figure 2C
), although activity never reached the
magnitude seen in women (Figure 2
, A and B).
| Discussion |
|---|
|
|
|---|
TF is proposed as a key mediator of thrombosis in
atherosclerosis.28 29 Circulating TF is
detected in patients with unstable angina,30 31 and high
levels are expressed on macrophages in unstable
plaques.32 33 Here, we report responses of monocytes from
healthy subjects to levels of CRP in the normal
physiological range (1 µg/mL), with TF activity
4-fold above basal levels and even greater at higher concentrations
(Figure 2
). Considering the significantly enhanced responses in
the elderly, particularly when LPS or IFN was added (Table 1
,
Figures 1
and 2
), this mechanism may contribute to the
increased risk of coronary events seen in population studies at
plasma CRP concentrations >2 to 7 mg/L6 34 and in studies
of unstable coronary disease at levels >3 to 15.5
mg/L.35 36
Synergy with IFN is important because the development of
atherosclerosis has a cell-mediated immune
component21 that includes IFN production by
CD4+ T lymphocytes, and because T-cell activation
is present in unstable angina.37 IFN doubled TF
expression induced by CRP to 50 to 100 times basal levels (Table 1
), and a critical role for lymphocytes for optimal
responsiveness to CRP+IFN was indicated. Amplification of monocyte TF
by contact with T lymphocytes is mediated, at least in part, by
ligation of macrophage CD40 with CD40 ligand, a
CD4+ T-cell surface protein implicated in the
pathogenesis of atherosclerosis and the triggering of
acute coronary events.38
Coronary artery disease incidence increases with age and is
greater in men than in age-matched women, although sex differences
decline with age, and elevated CRP levels predict cardiac events in
older men and women.4 5 6 7 34 TF activity on monocytes from
young women stimulated by CRP±LPS/IFN was significantly less than on
cells from men of the same age (Table 1
). In contrast, activity
on monocytes from postmenopausal women, particularly those aged >60
years, increased exponentially to levels higher than those induced on
monocytes from males of the same age (Figure 1
). This mechanism
may contribute in part to the age- and sex-related increases in risk
and the predictive role of CRP. Postmenopausal women taking estrogen
apparently have fewer cardiovascular events than
untreated women,39 40 although in the HERS trial (Heart
and Estrogen/progestin Replacement Study),41 HRT was not
associated with reduced cardiovascular events; patients
taking HRT had more events in the first year of treatment and fewer
events in the fourth and fifth years, which indicates a biphasic
response. Hypertension, insulin resistance, and
hyperlipidemia are all beneficially affected by
estrogen,42 and estrogen retards plaque progression
in animal studies.43 44 Moreover, plasminogen
activator inhibitor-1 levels are reduced in
women taking oral HRT, which suggests an altered balance between
coagulation and fibrinolysis.45 Although
the effects of HRT on plasma levels of components of coagulation do not
indicate a protective effect,42 our findings strongly
support the view that HRT may be cardioprotective by affecting cellular
procoagulants. TF activity on monocytes from women taking HRT was
dramatically less than that of untreated women (Table 2
), even
with relatively low levels of CRP (Figure 2
), and it approached
activity exhibited by monocytes from young men (Table 1
).
A previous study of LPS-stimulated monocytes from women taking estrogen indicated lower amounts of tumor necrosis factor, thromboxane B2, and TF than produced by cells taken before treatment; 12 months was required for stabilization of reduced responsiveness.46 We failed to demonstrate an effect of estradiol in vitro on TF induced by CRP on monocytes from untreated women (not shown), which suggests an acquired response. Although mechanisms are unclear, they may exert their effects via an indirect pathway that regulates the synthetic capacity of monocytes, possibly through an effect on stem cells. If a period of some months is required to regulate monocyte responsiveness, this may explain the biphasic response in HERS.41
Results reported here suggest an important amplification loop linking some well-established risk factors for cardiovascular disease and inflammation. Deposits of CRP around collections of foam cells,11 together with CD4+ T cells producing IFN, could collectively induce high levels of TF, which might initiate thrombus formation. Moreover, elevated CRP produced as a result of infection (eg, by Chlamydia) may prime blood monocytes and elevate responses to LPS, thereby exacerbating existing inflammatory lesions and provoking thrombosis. The balance of these mediators, and their interaction with age, sex, circulating levels of CRP, estrogen replacement therapy, levels of anticoagulants,31 and fibrinolytic factors, may contribute to the variable outcome of plaque rupture and thrombosis in coronary syndromes.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received August 30, 1999; revision received October 28, 1999; accepted November 15, 1999.
| References |
|---|
|
|
|---|
interferon and
interferon on thromboplastin
response in monocytes and endothelial cells. Clin
Exp Immunol. 1987;70:471478.[Medline]
[Order article via Infotrieve]
synergizes
with lipopolysaccharide to induce macrophage membrane
procoagulants. J Immunol. 1988;141:15361542.[Abstract]This article has been cited by other articles:
![]() |
J. A Bastarache, L. Wang, T. Geiser, Z. Wang, K. H Albertine, M. A Matthay, and L. B Ware The alveolar epithelium can initiate the extrinsic coagulation cascade through expression of tissue factor Thorax, July 1, 2007; 62(7): 608 - 616. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Cai, C. Song, I. Endoh, J. Goyette, W. Jessup, S. B. Freedman, H. P. McNeil, and C. L. Geczy Serum Amyloid A Induces Monocyte Tissue Factor J. Immunol., February 1, 2007; 178(3): 1852 - 1860. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Paffen and M. P.M. deMaat C-reactive protein in atherosclerosis: A causal factor? Cardiovasc Res, July 1, 2006; 71(1): 30 - 39. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Biolo, A. Amoroso, S. Savoldi, A. Bosutti, M. Martone, D. Pirulli, F. Bianco, S. Ulivi, S. Bertok, M. Artero, et al. Association of interferon-{gamma} +874A polymorphism with reduced long-term inflammatory response in haemodialysis patients Nephrol. Dial. Transplant., May 1, 2006; 21(5): 1317 - 1322. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. F. Bonvini, T. Hendiri, and E. Camenzind Inflammatory response post-myocardial infarction and reperfusion: a new therapeutic target? Eur. Heart J. Suppl., October 1, 2005; 7(suppl_I): I27 - I36. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Ikonomidis, J. Lekakis, I. Revela, F. Andreotti, and P. Nihoyannopoulos Increased circulating C-reactive protein and macrophage-colony stimulating factor are complementary predictors of long-term outcome in patients with chronic coronary artery disease Eur. Heart J., August 2, 2005; 26(16): 1618 - 1624. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. J. Harvey and D. P. Ramji Interferon-{gamma} and atherosclerosis: Pro- or anti-atherogenic? Cardiovasc Res, July 1, 2005; 67(1): 11 - 20. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Napoleone, A. Di Santo, G. Peri, A. Mantovani, G. de Gaetano, M. B. Donati, and R. Lorenzet The long pentraxin PTX3 up-regulates tissue factor in activated monocytes: another link between inflammation and clotting activation J. Leukoc. Biol., July 1, 2004; 76(1): 203 - 209. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Hashimoto, K. Kitagawa, H. Hougaku, H. Etani, and M. Hori Relationship Between C-Reactive Protein and Progression of Early Carotid Atherosclerosis in Hypertensive Subjects Stroke, July 1, 2004; 35(7): 1625 - 1630. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. T. Willerson and P. M. Ridker Inflammation as a Cardiovascular Risk Factor Circulation, June 1, 2004; 109(21_suppl_1): II-2 - II-10. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Paffen, H. L. Vos, and R. M. Bertina C-Reactive Protein Does Not Directly Induce Tissue Factor in Human Monocytes Arterioscler. Thromb. Vasc. Biol., May 1, 2004; 24(5): 975 - 981. [Abstract] [Full Text] |
||||
![]() |
S.-H. Kim, S. M. Lessner, Y. Sakurai, and Z. S. Galis Cyclophilin A as a Novel Biphasic Mediator of Endothelial Activation and Dysfunction Am. J. Pathol., May 1, 2004; 164(5): 1567 - 1574. [Abstract] [Full Text] [PDF] |
||||
![]() |
X. Gao, O. I. Bermudez, and K. L. Tucker Plasma C-Reactive Protein and Homocysteine Concentrations Are Related to Frequent Fruit and Vegetable Intake in Hispanic and Non-Hispanic White Elders J. Nutr., April 1, 2004; 134(4): 913 - 918. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Wakatsuki, N. Ikenoue, K. Shinohara, K. Watanabe, and T. Fukaya Effect of Lower Dosage of Oral Conjugated Equine Estrogen on Inflammatory Markers and Endothelial Function in Healthy Postmenopausal Women Arterioscler. Thromb. Vasc. Biol., March 1, 2004; 24(3): 571 - 576. [Abstract] [Full Text] |
||||
![]() |
G.M. Hirschfield and M.B. Pepys C-reactive protein and cardiovascular disease: new insights from an old molecule QJM, November 1, 2003; 96(11): 793 - 807. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. M. Rahel, F. L.J. Visseren, M.-J. Suttorp, T. H.W. Plokker, J. C. Kelder, B. M. de Jongh, K.P. Bouter, and R. J.A. Diepersloot Preprocedural serum levels of acute-phase reactants and prognosis after percutaneous coronary intervention Cardiovasc Res, October 15, 2003; 60(1): 136 - 140. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Silvestri, O. Gebara, C. Vitale, M. Wajngarten, F. Leonardo, J. A.F. Ramires, M. Fini, G. Mercuro, and G. M.C. Rosano Increased Levels of C-Reactive Protein After Oral Hormone Replacement Therapy May Not Be Related to an Increased Inflammatory Response Circulation, July 1, 2003; 107(25): 3165 - 3169. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. T.H. Yeh and J. T. Willerson Coming of Age of C-Reactive Protein: Using Inflammation Markers in Cardiology Circulation, January 28, 2003; 107(3): 370 - 371. [Full Text] [PDF] |
||||
![]() |
A. D. Pradhan, J. E. Manson, J. E. Rossouw, D. S. Siscovick, C. P. Mouton, N. Rifai, R. B. Wallace, R. D. Jackson, M. B. Pettinger, and P. M Ridker Inflammatory Biomarkers, Hormone Replacement Therapy, and Incident Coronary Heart Disease: Prospective Analysis From the Women's Health Initiative Observational Study JAMA, August 28, 2002; 288(8): 980 - 987. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. L. Bhatt and E. J. Topol Need to Test the Arterial Inflammation Hypothesis Circulation, July 2, 2002; 106(1): 136 - 140. [Full Text] [PDF] |
||||
![]() |
E. T.H. Yeh, H. V. Anderson, V. Pasceri, and J. T. Willerson C-Reactive Protein: Linking Inflammation to Cardiovascular Complications Circulation, August 28, 2001; 104(9): 974 - 975. [Full Text] [PDF] |