Thrombosis |
From the Veterans Affairs Medical Center (S.R.L., R.A.E., D.D.H.), and the Departments of Internal Medicine (S.R.L., D.J.P., D.D.H.) and Pharmacology (D.D.H.), University of Iowa College of Medicine, Iowa City, Iowa; The Scripps Research Institute (J.A.F., J.H.G.), La Jolla, Calif; and the Oregon Regional Primate Research Center (M.R.M.), Beaverton, Ore.
Correspondence to Steven R. Lentz, MD, PhD, Department of Internal Medicine, C303 GH, University of Iowa, Iowa City, IA 52242. E-mail steven-lentz{at}uiowa.edu
| Abstract |
|---|
|
|
|---|
-thrombin or
activated protein C (APC) in cynomolgus monkeys. Two groups of
monkeys were fed either a control diet (n=18) or an atherogenic diet
(n=12) that produces both hypercholesterolemia
and moderate hyperhomocyst(e)inemia. A third group (n=8) was fed an
atherogenic diet for 15 months, and then fed the atherogenic diet
supplemented with B vitamins for 6 months to correct the
hyperhomocyst(e)inemia. The plasma homocyst(e)ine level was higher in
monkeys fed the atherogenic diet (9.6±1.0 µmol/L) than in
monkeys fed the control diet (3.7±0.2 µmol/L) or the
atherogenic diet with B vitamins (3.6±0.2 µmol/L)
(P<0.001). Infusion of thrombin produced a much greater
prolongation of the activated partial thromboplastin time in
monkeys fed the control diet (52±10 seconds) than in monkeys fed the
atherogenic diet either with (24±4 seconds) or without (27±5 seconds)
supplemental B vitamins (P<0.02). Thrombin-dependent
generation of circulating APC was higher in control (294±17 U/mL) than
in atherosclerotic (240±14 U/mL) monkeys (P<0.05),
although levels of fibrinogen, plasminogen, D-dimer, and
thrombin-antithrombin complexes were similar in each group. Injection
of human APC produced a similar prolongation of the activated
partial thromboplastin time in control (31±3 seconds) and
atherosclerotic (29±2 seconds) monkeys. These findings provide
evidence for impaired anticoagulation, due partly to decreased
formation of APC, in atherosclerosis. The blunted
anticoagulant response to thrombin in
hypercholesterolemic monkeys was not corrected by
supplementation with B vitamins.
Key Words: atherosclerosis cholesterol homocysteine protein C thrombomodulin
| Introduction |
|---|
|
|
|---|
The protein C anticoagulant pathway is impaired in patients with inherited deficiencies of protein C or protein S and also in patients with a mutation in factor V (factor V Leiden) that renders it resistant to activated protein C (APC).4 5 The physiological importance of the protein C anticoagulant pathway is underscored by the observation that resistance to APC is found in 20% to 50% of patients with venous thromboembolism.6 Resistance to APC that is independent of factor V Leiden may be associated with increased risk of stroke.7 8
In a previous study, we observed impaired
endothelium-dependent vasodilatation and decreased
protein C activation in aortas from monkeys with moderate
hyperhomocyst(e)inemia [plasma homocyst(e)ine*
10
µmol/L].9 These observations suggested that
endothelial dysfunction, with impaired regulation of
thrombin by thrombomodulin, may contribute to an increased risk of
atherosclerotic and thrombotic vascular disease in
hyperhomocyst(e)inemia.10 The term homocyst(e)ine is used
to indicate that plasma homocysteine assays measure the total
concentration of thiol, disulfide, and mixed disulfide adducts of
homocysteine.
More recently, we examined endothelial function in monkeys fed a diet that produces both hypercholesterolemia and hyperhomocyst(e)inemia.11 In contrast to decreased protein C activation in normocholesterolemic monkeys with isolated hyperhomocyst(e)inemia,9 we observed increased protein C activation in the aorta and carotid artery from atherosclerotic monkeys with the combination of hypercholesterolemia and hyperhomocyst(e)inemia.11 These ex vivo observations suggested that the regulation of thrombin's anticoagulant properties may be altered in atherosclerosis, perhaps through increased expression of thrombomodulin in response to vessel injury.12
Altered protein C activation may influence the susceptibility of large
arteries to chronic or acute thrombotic complications of
atherosclerosis, involving small mural thrombi or
occlusive thrombus, respectively. The major site of activation of
protein C in vivo, however, is thought to be in small vessels, where
the local concentration of thrombomodulin is highest.2
Once generated in the microcirculation, APC is a systemic anticoagulant
with a half-life of
20 minutes.13 It is not known
whether hypercholesterolemia or
hyperhomocyst(e)inemia influences thrombin activity or activation of
protein C in vivo. The goal of this study, therefore, was to test the
hypothesis that anticoagulant responses to
-thrombin are altered in
atherosclerosis. We used an in vivo approach in which
activation of endogenous protein C was measured in response
to an infusion of
-thrombin in cynomolgus monkeys. Our results
indicate that atherosclerosis, with or without moderate
hyperhomocyst(e)inemia, significantly alters activation of the protein
C pathway in vivo.
| Methods |
|---|
|
|
|---|
1.0 µg
vitamin B-12, 0.75 mg vitamin B-6, and <25 µg folic acid
daily).14 This atherogenic diet produces both
hypercholesterolemia and
hyperhomocyst(e)inemia.11 Another group of monkeys was fed
the atherogenic diet for 15 months and then fed the atherogenic diet
supplemented with B vitamins (5 mg folic acid, 400 µg cyanocobalamin,
and 20 mg pyridoxine hydrochloride daily) for 6 months to decrease
plasma homocyst(e)ine concentrations. At the end of the feeding period,
the weights of the monkeys were 5.9±0.4 kg (control diet), 6.7±0.4 kg
(atherogenic diet), and 6.4±0.3 kg (atherogenic diet supplemented with
B vitamins).
Experimental Protocol
Human
-thrombin (3050 U/mg) was purchased from Enzyme
Research Laboratories. Protein C was purified from human plasma by
immunoaffinity chromatography as
described.15 Protein C was activated in vitro with
human thrombin, and APC was purified by anion exchange
chromatography.16 The purified APC
contained no detectable thrombin (<0.05 U of thrombin per mg of
APC).
Before administration of thrombin or APC, animals were sedated with
ketamine hydrochloride (20 mg/kg IM) and anesthetized
with sodium pentobarbital (20 mg/kg IV). A nonobstructive catheter was
inserted into an axillary artery for blood sampling, and the axillary
vein was cannulated for administration of thrombin, APC, and
supplemental anesthesia (sodium pentobarbital 5 mg ·
kg-1 · h-1). Blood
pressure was monitored continuously. Human
-thrombin (1.0 to 2.5
µg · kg-1 ·
min-1) was infused in 10 mL of saline for 10
minutes through the axillary vein catheter. Human APC (50 to 100
µg/kg IV) was injected as a bolus in 5 mL of saline with 5%
dextrose. Blood was collected from the axillary artery catheter
directly into 3.4 mmol/L EDTA [for measurement of blood counts,
homocyst(e)ine, and lipids], a 1/10 volume of 3.8% sodium citrate
with 3.0 mmol/L benzamidine (for determination of protein C and
APC), or a 1/10 volume of 3.8% sodium citrate without benzamidine (for
other hemostatic assays). Blood samples were placed immediately on ice,
and plasma was isolated by centrifugation at
2500g for 30 minutes at 4°C. Aliquots of plasma were
stored at -70°C for up to 20 months before analysis. The
protocol was approved by the University of Iowa and Veterans Affairs
Animal Care and Use Committees.
Hemostatic Assays
Plasma circulating APC and total protein C were measured by
enzyme capture assay using the anti-human protein C light-chain
monoclonal antibody C317 and chromogenic
substrate S-2366 (Chromogenix). This assay has been used previously to
measure APC and protein C in human and baboon
plasma.15 16 18 After determination of immunocaptured APC
amidolytic activity, immobilized protein C zymogen was
activated with 1.0 U/mL Protac (American
Diagnostica Inc), and total protein C amidolytic activity
was determined as described.15 Standard curves were
generated using plasma pooled from 19 monkeys. One unit of APC or
protein C was defined as the amount present in 1.0 mL of pooled
monkey plasma. Compared with pooled human plasma, pooled monkey plasma
contained 1.7-fold higher total protein C activity. This difference in
activity could be due to either a higher concentration of protein C in
monkey plasma or a higher amidolytic activity of monkey APC. The
concentration of APC was <0.1% of total protein C in pooled monkey
plasma.
The activated partial thromboplastin time (APTT) was measured in an ACL-300+ coagulometer (Instrumentation Laboratory) with the Platelin L reagent (Organon Tecknika Corp). Fibrinogen was measured in a fibrometer (BBL Fibrosystem, Becton Dickinson) by the Clauss method19 with 100 U/mL bovine thrombin (Dade International, Inc). A standard curve was generated using purified human fibrinogen (Chromogenix).
Plasminogen was measured after activation with
streptokinase (Kabi Pharmacia, Inc) with the use of the
chromogenic substrate S-2251 (Chromogenix).
2-Antiplasmin was detected by measuring the
inhibition of S-2251 hydrolysis by exogenous plasmin (Chromogenix).
Plasma activities of factors V, VIII, IX, XI, and XII were measured in
an ACL-300+ coagulometer with factor V or VIIIdeficient plasma
(George King Bio-Medical). Reference curves for
plasminogen,
2-antiplasmin, and
factors V, VIII, IX, XI, and XII were generated using pooled normal
monkey plasma that contained, by definition, 1.0 U/mL of each plasma
factor. Plasma levels of thrombin-antithrombin complexes and D-dimer
were measured by enzyme immunoassay (Behring Diagnostics
Inc).
In vitro assays for resistance to APC were performed by adding purified human APC (0.6 µg/mL) to the APTT calcium reagent and measuring the APTT.20 Protein C inhibitor (PCI) was measured by enzyme immunoassay using a monoclonal anti-PCI antibody (generously provided by Dr Bonno Bouma, University Hospital, Utrecht, The Netherlands).21 Peroxidase-labeled polyclonal anti-PCI IgG was used for detection, and the assay was standardized against pooled monkey plasma.
Other Assays
Fasting plasma homocyst(e)ine concentration was measured by
high-performance liquid chromatography and
electrochemical detection as described previously.22 23
Total plasma cholesterol, HDL cholesterol, LDL
cholesterol, and triglycerides were measured
using methods established by the Lipid Research Centers and
standardized by the Centers for Disease Control and Prevention,
Atlanta, Ga.24
Statistical Analysis
Comparisons between monkeys fed the control or atherogenic diet
were performed using the unpaired 2-tailed Student's t
test. Comparisons between multiple groups were performed using 1-way
ANOVA. A value of P<0.05 was used to define statistical
significance.
| Results |
|---|
|
|
|---|
|
|
Response to Infusion of Thrombin
In a pilot experiment, a 10-minute infusion of human
-thrombin
produced dose-dependent increases in APTT and circulating APC in a
monkey fed the control diet (Figure 2
). A
dose of thrombin of 2.5 µg · kg-1
· min-1 was chosen to compare anticoagulant
responses in monkeys fed control, atherogenic, and vitamin-supplemented
atherogenic diets. Baseline APTT values before infusion of thrombin did
not differ among the 3 groups (Figure 3A
). Infusion of thrombin produced
greater prolongation of the APTT in monkeys fed the control diet
(prolongation of 52±10 seconds over baseline APTT) than in monkeys fed
the atherogenic diet with (24±4 seconds) or without (27±5 seconds)
supplemental B vitamins (P<0.02; Figure 3A
).
|
|
After infusion of thrombin, peak levels of circulating APC were 18%
lower in monkeys fed the atherogenic diet without supplemental B
vitamins (240±14 U/mL) than in monkeys fed the control diet (294±17
U/mL, P=0.03; Figure 3B
). Levels of circulating APC
did not differ between monkeys fed the atherogenic diet with
supplemental B vitamins and those fed the atherogenic diet without
supplemental B vitamins (Figure 3B
). These findings suggested
that anticoagulant responses to thrombin were impaired in
atherosclerotic monkeys in part because of decreased circulating
APC.
After infusion of thrombin, platelet count decreased <15%, plasma
fibrinogen decreased
35%, factor V activity decreased 40% to 50%,
and factor VIII activity decreased 60% to 70% in each group of
monkeys (Table 2
). Plasma activities of
factors IX, XI, and XII decreased <30%, and total plasma protein C
activity decreased <15% after infusion of thrombin in each group
(Table 2
). No significant differences in the generation of
thrombin-antithrombin complexes or D-dimer or in the consumption of
fibrinogen, plasminogen, or
2-antiplasmin were observed in the 3 groups
(Table 2
).
|
Response to APC In Vitro
One potential mechanism for impaired anticoagulant response to
thrombin is resistance to the anticoagulant effects of APC. To
determine whether resistance to APC could be detected in plasma from
atherosclerotic monkeys in vitro, the APTT of monkey plasma was
measured in the absence or presence of 0.6 µg/mL of human APC in an
APC resistance assay. Addition of APC prolonged the APTT by 44.2±3.6
seconds in monkeys fed the control diet and by 36.0±2.7 seconds
in monkeys fed the atherogenic diet (P>0.05; Figure 4
).
|
Response to Injection of APC
To determine whether monkeys fed the atherogenic diet exhibited
resistance to APC in vivo, we measured anticoagulant responses to
injection of purified human APC. Injection of APC into a control monkey
produced prolongation of the APTT and elevation of plasma APC that were
dose dependent (Figure 5
). A bolus (75
µg/kg) of APC produced a similar prolongation of the APTT in monkeys
fed control (31±3 seconds) and atherogenic (29±2 seconds) diets
(P>0.05; Figure 6A
). Recovery
and elimination of injected human APC were similar in the 2 groups,
with peak levels of 407±21 and 430±13 U/mL and plasma half-lives of
23.9±1.3 and 24.3±1.3 minutes, respectively (Figure 6B
). Thus,
the half-life of human APC was not altered by the atherogenic diet.
|
|
Because plasma PCI inhibits both activation of protein C and APC activity,21 25 we also measured levels of PCI in monkey plasma. Levels of PCI were similar in monkeys fed the control diet (0.96±0.07 U/mL), the atherogenic diet (1.00±0.08 U/mL), and the atherogenic diet supplemented with B vitamins (0.94±0.08 U/mL).
| Discussion |
|---|
|
|
|---|
-thrombin in low doses into control and atherosclerotic cynomolgus
monkeys. Our major findings are that anticoagulant responses to
thrombin are impaired in monkeys with diet-induced
atherosclerosis and that supplementation with B
vitamins does not restore anticoagulant responses to normal in
atherosclerotic monkeys with persistent
hypercholesterolemia. Decreased responses to
thrombin in atherosclerotic monkeys were associated with decreased
generation of APC in vivo. These findings suggest that impairment of
the protein C anticoagulant pathway may contribute to thrombogenicity
in atherosclerosis. After its intravenous infusion, thrombin is rapidly cleared from the circulation, and some of it binds to thrombomodulin and activates protein C.2 3 18 26 Although thrombomodulin is expressed on the endothelial surface of both large and small vessels, most protein C is activated in vivo in small vessels, where the ratio of endothelial surface area to plasma volume is highest.2 In baboons, infusion of thrombin causes extensive activation of protein C and is associated with a systemic antithrombotic response that can be demonstrated by a decreased accumulation of fibrin and platelets on a thrombogenic vascular graft.15 16 18 This antithrombotic response can be completely abated by pretreatment with a monoclonal antibody that blocks thrombin-dependent protein C activation.18 Infusion of thrombin also prolongs the APTT and generates anticoagulant APC in cynomolgus monkeys.27
The monkeys in this study were fed an atherogenic diet for
20
months. In previous studies, we demonstrated that this diet produces
both hyperhomocyst(e)inemia and
hypercholesterolemia, with moderately severe
atherosclerotic lesions in large vessels.11 28 We
anticipated that infusion of thrombin might generate higher levels of
APC in atherosclerotic monkeys than in monkeys fed the control diet,
because we previously observed increased protein C activation in
explants of aorta and carotid artery from atherosclerotic
monkeys.11 In contrast to our expectations, however,
levels of circulating APC were almost 20% lower after infusion of
thrombin in atherosclerotic monkeys than in control monkeys. These
observations suggest that although local activation of protein C may be
increased in large vessels that contain atherosclerotic lesions,
systemic generation of APC, which occurs predominantly in small
vessels, may be decreased in atherosclerotic animals.
Atherosclerotic monkeys did not exhibit significant resistance to human
APC in vitro (Figure 4
), and injection of human APC produced a
similar prolongation of the APTT in monkeys fed control and atherogenic
diets (Figure 6
). The plasma half-life of human APC was
24
minutes in both control and atherosclerotic monkeys. This calculated
half-life is similar to that estimated for endogenous APC
in baboons (
16 minutes) in our previous studies18 and
in cynomolgus monkeys (
15 minutes) in a previous
report.27 Thus, differences in clearance or inactivation
of APC do not account for differences in anticoagulant response to
thrombin between control and atherosclerotic monkeys.
After infusion of thrombin into monkeys, there was a significantly
decreased activation of protein C associated with
atherosclerosis and
hypercholesterolemia, since peak levels of
circulating APC were
20% lower in atherosclerotic monkeys compared
with control monkeys (Figure 3B
). To help determine whether this
modest difference in plasma APC could account for the large difference
in APTT prolongation observed in response to thrombin (Figure 3A
), we examined the relationship between circulating APC levels
and APTT by using data obtained after administration of either APC or
thrombin (Figure 7
). After injection of
APC, the APTT was similarly and directly related to the level of
circulating APC in monkeys fed either the control or the atherogenic
diets (Figure 7A
). During the first 5 to 10 minutes after
beginning the infusion of thrombin, the relationship between APTT and
circulating APC was very similar to that observed after injection of
APC (Figure 7B
). At later times after thrombin infusion,
however, there was a leftward shift in the APTT versus APC
dose-response relationship, with a much greater prolongation of the
APTT than can be accounted for by plasma APC alone (Figure 7B
).
Furthermore, this shift in response to APC was significantly blunted in
atherosclerotic monkeys compared with control monkeys (Figure 7B
). These findings suggest that the impaired anticoagulant
response to thrombin in atherosclerotic monkeys is caused partly by
diminished generation of APC and partly by another mechanism. Because
the baseline APTT values were the same, this second mechanism is
thrombin dependent, and it may or may not require the presence of APC.
This second mechanism may indicate a greater procoagulant response to
thrombin in atherosclerotic monkeys, which is consistent with
the greater thrombin-induced fall in platelets observed in
hypercholesterolemic monkeys than in control
monkeys.
|
Several other potential mechanisms could contribute to an impaired anticoagulant response to thrombin in atherosclerotic monkeys. It is conceivable but unlikely that infusion of thrombin could produce a lower plasma concentration of thrombin in atherosclerotic monkeys than in control monkeys, perhaps because of differences in plasma volume induced by the atherogenic diet. This possibility is unlikely because other hemostatic effects of thrombin, including consumption of fibrinogen and generation of thrombin-antithrombin complexes and D-dimer, were similar in the 3 groups of monkeys. For the same reason, it is unlikely that inactivation or clearance of thrombin differed significantly in monkeys fed control and atherogenic diets.
Another potential mechanism for altered protein C activation in different vascular beds is differential expression of the endothelial protein C receptor (EPCR), which potentiates thrombomodulin-dependent activation of protein C by thrombin.29 EPCR is present mainly on large blood vessels, whereas thrombomodulin is expressed on both large and small vessels.30 It is not known whether the expression or activity of EPCR is altered in hypercholesterolemia or atherosclerosis.
Finally, it is possible that protein C activation may be altered by
effects of the atherogenic diet on plasma lipoproteins. Despite large
increases in total cholesterol and LDL
cholesterol in monkeys fed the atherogenic diet, plasma
levels of HDL cholesterol decreased by >50% (Table 1
). In reactions with purified lipoproteins in vitro, HDL
enhanced the anticoagulant effects of APC and protein S,31
and atherogenic lipoproteins support prothrombin activation by factors
Xa and Va.32 The effects of HDL or other lipoproteins on
these coagulation reactions have not been studied in vivo. Whatever in
vivo mechanisms may ultimately be identified, the studies described
here indicate that the coagulation system's response to thrombin is
markedly abnormal and, on balance, shifted toward a procoagulant
potential in atherosclerotic monkeys.
Supplementation of the atherogenic diet with B vitamins for 6 months did not restore normal anticoagulant responses to thrombin. Because these monkeys were exposed to combined hypercholesterolemia and hyperhomocyst(e)inemia for 15 months before correction of their hyperhomocyst(e)inemia with B vitamins, we cannot exclude the possibility that elevated plasma homocyst(e)ine may have contributed to impairment of protein C activation or to another process that shifts the in vivo balance of procoagulant and anticoagulant forces. We have observed decreased thrombomodulin-dependent protein C activation in homocysteine-treated endothelial cells33 and in aortas from nonatherosclerotic monkeys with diet-induced hyperhomocyst(e)inemia.9 Further studies will be necessary to define the relative effects of hypercholesterolemia and hyperhomocyst(e)inemia on anticoagulant responses to thrombin in the absence of preexisting atherosclerosis.
| Acknowledgments |
|---|
Received August 25, 1998; accepted December 15, 1998.
| References |
|---|
|
|
|---|
glutamine mutation. Stroke. 1996;27:11631166.This article has been cited by other articles:
![]() |
D. D. Heistad Oxidative Stress and Vascular Disease: 2005 Duff Lecture Arterioscler. Thromb. Vasc. Biol., April 1, 2006; 26(4): 689 - 695. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. R. Lentz Another Lesson From the Factor V Leiden Mouse: Thrombin Generation Drives Arterial Disease Circulation, April 12, 2005; 111(14): 1733 - 1734. [Full Text] [PDF] |
||||
![]() |
A. Slungaard, J. A. Fernandez, J. H. Griffin, N. S. Key, J. R. Long, D. J. Piegors, and S. R. Lentz Platelet factor 4 enhances generation of activated protein C in vitro and in vivo Blood, July 1, 2003; 102(1): 146 - 151. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. R. Lentz Thrombosis of Vein Grafts: Wall Tension Restrains Thrombomodulin Expression Circ. Res., January 10, 2003; 92(1): 12 - 13. [Full Text] [PDF] |
||||
![]() |
G. Sandusky, D. T. Berg, M. A. Richardson, L. Myers, and B. W. Grinnell Modulation of Thrombomodulin-dependent Activation of Human Protein C through Differential Expression of Endothelial Smads J. Biol. Chem., December 13, 2002; 277(51): 49815 - 49819. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. R. Lentz, D. J. Piegors, J. A. Fernandez, R. A. Erger, E. Arning, M. R. Malinow, J. H. Griffin, T. Bottiglieri, W. G. Haynes, and D. D. Heistad Effect of hyperhomocysteinemia on protein C activation and activity Blood, August 28, 2002; 100(6): 2108 - 2112. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. R. Lentz, F. J. Miller Jr, D. J. Piegors, R. A. Erger, J. A. Fernandez, J. H. Griffin, and D. D. Heistad Anticoagulant Responses to Thrombin Are Enhanced During Regression of Atherosclerosis in Monkeys Circulation, August 13, 2002; 106(7): 842 - 846. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. G. Laszik, X. J. Zhou, G. L. Ferrell, F. G. Silva, and C. T. Esmon Down-Regulation of Endothelial Expression of Endothelial Cell Protein C Receptor and Thrombomodulin in Coronary Atherosclerosis Am. J. Pathol., September 1, 2001; 159(3): 797 - 802. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. R. Lentz, D. J. Piegors, M. R. Malinow, and D. D. Heistad Supplementation of Atherogenic Diet With B Vitamins Does Not Prevent Atherosclerosis or Vascular Dysfunction in Monkeys Circulation, February 20, 2001; 103(7): 1006 - 1011. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. B. Taylor Jr, D. J. Stearns-Kurosawa, S. Kurosawa, G. Ferrell, A. C. K. Chang, Z. Laszik, S. Kosanke, G. Peer, and C. T. Esmon The endothelial cell protein C receptor aids in host defense against Escherichia coli sepsis Blood, March 1, 2000; 95(5): 1680 - 1686. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1999 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |