SEARCH

SEARCH BY CITATION

Keywords:

  • antiplasmin;
  • cardiovascular disease;
  • endothelial function;
  • haemostasis function

Abstract.

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Subjects and methods
  5. Methods
  6. Statistics
  7. Results
  8. Discussion
  9. Acknowledgements
  10. References

Objective.  To study whether haemostasis function variables correlate with endothelial function and other vasomotion characteristics of the brachial artery in a randomly selected healthy population of 35-year-old men and women.

Design.  Endothelial function was measured as flow mediated dilatation (FMD) of the brachial artery during reactive hyperaemia and the nonendothelial dependent dilatation after sublingual nitroglycerin (NTG) was administered. Haemostasis and fibrinolysis function were estimated by analysis of von Willebrand factor, plasminogen activator inhibitor-1, antiplasmin and fibrinogen.

Setting.  A general medicine research centre and a university hospital.

Subjects.  Randomly chosen men (= 53) and women (= 56).

Results.  Univariate correlation analysis showed significant correlations between haemostasis factors, conventional risk factors for cardio- vascular disease and indices of vasomotion of the brachial artery. In multivariate analysis, with haemostasis variables and conventional risk factors included, antiplasmin was the strongest explanatory variable for FMD. When antiplasmin was removed from the analysis, the r-value dropped from 0.46 to 0.35. Antiplasmin also correlated with NTG-induced dilatation (positively) and brachial diameter at rest (negatively), albeit less consistently.

Conclusions.  Antiplasmin correlates significantly and independently to FMD, reflecting endothelial function, and also to brachial artery diameter at rest and nitroglycerin-induced dilatation. In multivariate analysis these correlations of antiplasmin to arterial characteristics were stronger than for ‘conventional’ risk factors, such as smoking, blood pressure and serum cholesterol.


Introduction

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Subjects and methods
  5. Methods
  6. Statistics
  7. Results
  8. Discussion
  9. Acknowledgements
  10. References

Endothelial cells situated at the interface between blood and the vessel wall, together with circulating platelets and plasma coagulation proteins, play a pivotal role in regulating haemostasis balance [ 1]. Flow-mediated dilatation (FMD) of the arteries, considered to reflect endothelial function [ 2], is influenced by several risk factors for cardiovascular disease. Thus, FMD is impaired in coronary artery disease patients [ 3], and decreases with age [ 4], in smokers [ 5], in diabetics [ 6] and in children with hypercholesterolaemia [ 2]. FMD increases after cholesterol lowering [ 7] and is preserved in diabetics with good glucose control [ 6]. In line with a modified response to injury theory [ 8], it has been suggested that accelerated fibrin deposition and the impairment of intravascular fibrinolytic activity can initiate and speed up atherosclerosis plaque development, leading to cardiovascular disease (CVD) [ 9]. Endothelial cells synthesize and secrete plasminogen activator (PA) and plasminogen activator inhibitor (PAI), and thus serve as a vital surface for plasmin generation [ 10]. Whether plasma coagulation proteins (and platelets) affect the integrity of the endothelial cell lining, causing disturbed vasomotion and endothelial function in vivo, has not been reported [ 11].

Using the noninvasive postischaemic FMD method for estimating endothelial function clinical studies on healthy subjects can be performed [ 2, 12]. Interestingly, the endothelial function of the brachial artery correlates with that of the coronary arteries [ 13, 14]. Thus, the noninvasively derived FMD of the brachial artery reflects the endothelial function of the brachial artery and also of the more CVD-appropriate coronary arteries.

The main objective of the present work was to study arterial reactivity of the brachial artery and haemostasis and fibrinolysis function variables. Accordingly, von Willebrand factor, plasma fibrinogen, plasminogen activator inhibitor-1 (PAI-1) and antiplasmin were analysed, representing different facets of the haemostasis system, and related to the endothelial function and other vasomotion characteristics of the brachial artery. Furthermore, we were interested to see how haemostasis function variables, together with so called ‘conventional’ risk factors for CVD, influenced arterial reactivity in a healthy randomized population of 35-year-old men and women.

Subjects and methods

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Subjects and methods
  5. Methods
  6. Statistics
  7. Results
  8. Discussion
  9. Acknowledgements
  10. References

Subjects

The study population has been described in detail [ 12]. In short, 100 men and 100 women, all 35 years old, were invited by letter to participate in a cardiovascular risk factor study. They were selected from the official population register of Stockholm County, Sweden, and constituted a random sample of the inhabitants in Sollentuna municipality. Sixty-six of the 92 men (72%) living in the municipality and 74 of the 88 (84%) nonpregnant women residents came for control. None of the participants had clinical signs or symptoms of infectious disease. They were asked to avoid aspirin and other anti-inflammatory drugs for 2 weeks prior to the blood sampling. Furthermore, the participants had a physical examination and answered standardized questions about their medical history, smoking habits and family history of disease. This was followed by donation of blood for analysis. The study was approved by the local ethics committee.

Methods

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Subjects and methods
  5. Methods
  6. Statistics
  7. Results
  8. Discussion
  9. Acknowledgements
  10. References

Blood chemistry

Blood was drawn after fasting overnight (since 21.00 hours) between 08.00 and 10.00 after a 15-minute rest. Analysis included haemostasis variables, routine chemistry, blood glucose and serum lipoproteins.

The inhibitor of t-PA plasminogen activator inhibitor (PAI-1) was determined by a functional spectrophotometric method [ 15], using kits from Biopool AB (Umeå, Sweden). p-Fibrinogen was analysed immunochemically by means of quantitative electroimmunoassay ('rocket' electrophoresis). Antiplasmin was determined by a spectrophotometric method as described by Tegner-Nilsson et al. [ 16].

Analysis of the von Willebrand factor was performed immunochemically by means of ELISA (Diagnostica Stago, Asniér, France).

Lipoproteins were analysed for serum cholesterol and serum triglycerides by enzymatic techniques. High density lipoprotein (HDL) cholesterol was measured in serum after precipitation of the apolipoproteinB containing lipoproteins. Low density lipoprotein cholesterol (LDL) was calculated by the Friedewald formula. Serum was analysed for the concentration of apolipoproteinA-I, apolipoproteinB and lipolipoprotein (a) by ELISA techniques.

Endothelial and nonendothelial function

The method has been reported before [ 12], and basically follows the method described by Celermajer and colleagues [ 2, 17]. In short, ultrasound scans of the brachial artery were made with a duplex scanner (Acuson 128XP), with a 7.0 MHz ART linear array transducer. Subjects were asked not to smoke, and not to drink coffee or tea for at least 2 h before the study. Scans were videotaped on super VHS cassettes for off-line analysis. A resting B-mode scan was recorded, flow velocity was measured with a 5 MHz pulsed Doppler with a 1.5 mm gate width in the centre of the vessel at a 70° angle. Flow-increase was induced by instantaneous inflation of a small pneumatic tourniquet placed around the forearm of the patient to 300 mmHg for 4.5 min. A second scan was recorded from 30 s before and 90 s after cuff deflation. Following a 10-minute-rest, a third scan was recorded. Then 0.40 mg sublingual NTG spray was administered and 3–4 min later the last scan was recorded. Vessel diameter was measured directly from videotape recordings by two observers unaware of the subjects' smoking habits, blood chemistry or the stage of the experiment. The between-observer coefficient of variation for measuring vessel diameter at baseline, after hyperaemia and after NTG, was 1.3%.

Statistics

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Subjects and methods
  5. Methods
  6. Statistics
  7. Results
  8. Discussion
  9. Acknowledgements
  10. References

Results are given as mean ± SD. Univariate or stepwise multiple linear regression analyses were used. Groups were compared with Student's two-tailed unpaired t-test. Probability levels < 0.05 were considered significant.

Results

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Subjects and methods
  5. Methods
  6. Statistics
  7. Results
  8. Discussion
  9. Acknowledgements
  10. References

The physiological endothelial function test was performed within a fortnight from the blood sampling. Of the 140 subjects participating in the screening, 109 had scans of adequate quality. The ultrasound B-mode picture was of inadequate quality in 26 cases and two scans were lost due to technical problems and three subjects did not show up. Mean values for haemostasis function variables and the other major CVD risk factors were not significantly different between the subjects with adequate and failed ultrasound scans.

Vasomotion of the brachial artery

Women had lower blood flow at rest and smaller diameter compared to men ( Table 1). Flow-mediated dilatation was higher in women compared to men (all < 0.05). Nitroglycerin-induced dilatation was not significantly different between the sexes.

Table 1.   Clinical characteristics, blood and physiological variables of 35-year-old men and women Thumbnail image of

Clinical characteristics and blood chemistry

All participants gave self-report answers indicating that they were free from infectious diseases or symptomatic inflammation. By analogy, C-reactive protein was normal in all subject, i.e. below 5 mg L−1 in 105 subjects and between 5 and 22 in four subjects.

There were no significant differences between men and women in concentrations of haemostasis variables. Women had lower blood pressure and lower concentrations of B-glucose, apolipoprotein B and triglycerides and higher HDL cholesterol and apolipoprotein A-I concentration (all < 0.05, see Table 1).

Univariate correlations between brachial indices and various characteristics

Univariate correlation analysis between the analysed variables are shown in Table 2 for men and for women. In the text we mainly outline brachial indices and haemostasis function correlations.

Table 2a.   Univariate correlations of brachial artery flow-mediated dilatation (FMD), nitroglycerin-mediated dilatation (NTG), bloodflow at rest and diameter at rest versus other characteristics in 35-year-old men Thumbnail image of
Table 2b.   Univariate correlations of brachial artery flow-mediated dilatation (FMD), nitroglycerin-mediated dilatation (NTG), bloodflow at rest and diameter at rest versus other characteristics in 35-year-old women Thumbnail image of

Significant correlations were seen between antiplasmin and brachial artery indices. Thus in men antiplasmin levels correlated inversely significantly with blood flow at rest, diameter at rest and nitroglycerin-mediated dilatation. In women antiplasmin levels correlated significantly with FMD. Tertiles of antiplasmin concentration for the whole population were plotted against brachial indices in the Figures. With increasing tertile FMD increased and baseline diameter decreased. Brachial diameter in women correlated signifi- cantly to von Willebrand factor concentration. Flow-mediated dilatation in men correlated significantly with fibrinogen levels.

The PAI-1 activity correlated significantly to antiplasmin and body mass index in both sexes, and to fibrinogen in women. Furthermore, PAI-1 correlated significantly to B-glucose and total triglyceride.

A gender difference was found in that women had stronger correlations between PAI-1 and B-glucose and PAI-1 and triglyceride than men and that PAI-1 and HDL cholesterol correlated (inversely) in women but not in men. No significant univariate correlations were seen between PAI-1 and brachial artery indices.

Antiplasmin correlated significantly with PAI-1 and P-fibrinogen in both sexes and, furthermore, to BMI in women. It also correlated with S-cholesterol in men and with triglyceride in both genders.

Multivariate analysis of vasomotion

Forward stepwise multiple regression analysis was performed, including both haemostasis function variables and other risk-factor parameters ( Table 3). Independent variables included were gender, smoking habits, diastolic blood pressure, Body Mass Index, leucocyte particle count, B-glucose, S-cholesterol, triglyceride, lipoprotein(a), von Willebrand factor, PAI-1, antiplasmin and P-fibrinogen. The dependent variables, one at a time, were FMD, NTG-induced dilatation, blood flow at rest and diameter of the brachial artery at rest.

Table 3a.   Summary of *stepwise multiple regression analysis for flow-mediated-dilatation in 35-year-old men and women (= 109) Thumbnail image of

The total r-value for FMD was 0.46. Antiplasmin was the strongest explanatory variable. Thus, removing antiplasmin from the equation resulted in a reduction of the r-value to 0.35 and a loss of B-glucose and P-fibrinogen as contributors. In the analysis without antiplasmin, contributors to FMD variation were sex, S-cholesterol, Lp(a) and von Willebrand factor. Multivariate analysis was also performed for women and men separately. For both genders antiplasmin levels came out as a significant variable explaining FMD function.

Antiplasmin also came out as a significant contributor influencing NTG-induced dilatation, blood flow at rest and vessel diameter at rest. However, removing antiplasmin from the analysis had minor effects on the total r-values, indicating a less pronounced role of antiplasmin in these contexts, as compared to that of the FMD response.

Discussion

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Subjects and methods
  5. Methods
  6. Statistics
  7. Results
  8. Discussion
  9. Acknowledgements
  10. References

The novel finding of the present study in healthy 35-year-old subjects is that the fast inhibitor in plasma of plasmin, antiplasmin, correlates significantly and independently with flow-mediated dilatation of the brachial artery, an index of endothelial function. The correlation is found in univariate analysis and in multivariate analysis with inclusion of the main CVD risk-factor variables where it comes out as the strongest explanatory factor for flow-mediated dilatation. Significant correlations, although weaker and less consistent, were also found between antiplasmin and brachial diameter at rest and NTG-induced dilatation.

Antiplasmin forms complexes with plasmin, causing inactivation of plasmin and lower plasma levels of both reactants [ 18, 19]. A high antiplasmin level may thus reflect low consumption due to low plasmin generation, in turn indicating a nonaroused fibrinolytic system. Indeed, Marongiu et al. [ 20] showed a partial consumption of antiplasmin in patients with elevated fibrinopeptide A levels, the latter reflecting fibrinolysis activation secondary to increased thrombin activation. A central role for thrombin/thrombin receptor interaction in fibrinolysis activation is indicated by the finding of increased thrombin receptor expression during vascular lesion formation [ 21].

Antiplasmin has been measured extensively since its characterization in the mid 70s but prospective studies linking its concentration to cardiovascular disease development have not been published. However, a recent study has indicated a role for antiplasmin in the development of atherosclerosis as it colocalizes with PAI-1 at plaque sites in the vessel wall [ 22]. Coagulation factor XIII crosslinks fibrin and antiplasmin and thereby protects fibrin from plasmin degradation [ 23, 24]. The pathophysiological effect of the complex formation is not clear, but it has been suggested that it modulates plasmin activity in the clot [ 23]. Furthermore, a fibrin protection from plasmin in plasma is likely to influence rheological properties by raised plasma viscosity. A high viscosity causes slower blood flow, which in turn is associated with less vasodilated arteries [ 12]. In our study, gender, smoking and B-glucose followed by antiplasmin contributed to explain the blood flow at rest in multivariate analysis. The more pronounced flow-mediated dilatation in subjects with high antiplasmin could accordingly be due to a smaller baseline diameter. The concomitant correlations between antiplasmin and brachial artery diameter at baseline (negative) as well as NTG-induced dilatation (positive) in combination with the FMD correlation support this argument.

Compromised liver function could possibly be an antecedent of both low antiplasmin levels and endothelial dysfunction. Synthesis of antiplasmin mainly occurs in the liver and concentrations are lower in patients with liver cirrhosis [ 25, 26]. Self-report information on alcohol intake (data not presented) which is known to correlate to liver function, was however, not correlated to antiplasmin concentration.

In conclusion, in a representative sample of healthy 35-year-old men and women, the study shows a positive correlation between antiplasmin and flow-mediated dilatation which is independent of and possibly stronger than that of so called conventional risk factors like smoking, blood pressure and cholesterol.

Acknowledgements

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Subjects and methods
  5. Methods
  6. Statistics
  7. Results
  8. Discussion
  9. Acknowledgements
  10. References

This study was funded by grants from the Karolinska Institute Foundation, the Foundation for Old Servants, Swedish Match AB and the Swedish Society for Medical Sciences.

References

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Subjects and methods
  5. Methods
  6. Statistics
  7. Results
  8. Discussion
  9. Acknowledgements
  10. References
  • 1
    Vaughan DE, Schafer AI, Loscalzo J. Normal mechanisms of haemostasis and fibrinolysis. In: Loscalzo J, Creager MMA, Dzau VJ, eds. Vascular Medicine. Boston, Toronto, London: Little, Brown and Company, 1992; 233 50.
  • 2
    Celermajer DS, Sorensen KE, Gooch VM, Spiegelhalter DJ, Miller OI, Sullivan I et al Non-invasive detection of endothelial dysfunction in children and adults at risk of atherosclerosis. Lancet 1992; 340: 1111 15.
  • 3
    Liberman EH, Gerhard MD, Uehata et al Flow-induced vasodilatation of the human brachial artery is impaired in patients <40 years of age with coronary artery disease. Am J Cardiol 1996; 78: 1210 14.
  • 4
    Celermajer DS, Sorensen KE, Spiegelhalter DJ, Georgakopolous D, Robinson J, Deanfield JE. Aging is associated with endothelial dysfunction in healthy men years before the age-related decline in women. J Am Coll Cardiol 1994; 24: 471 6.
  • 5
    Celermajer D, Sorensen K, Georgakopoulos D, Bull C, Thomas O, Robinson et al Cigarette smoking is associated with dose-related and potentially reversible impairment of endothelium-dependent dilation in healthy young adults. Circulation 1993; 88: 2149 55.
  • 6
    Jensen-Urstad KJ, Reichard PG, Rosfors JS, Lindblad LEL, Jensen-Urstad MT. Early atherosclerosis is retarded by improved long-term blood glucose control in patients with IDDM. Diabetes 1996; 45: 1253 8.
  • 7
    Vogel RA, Coretti MC, Plotnick GD. Changes in flow-mediated brachial artery vasoactivity with lowering of desirable cholesterol levels in healthy middle-aged men. Am J Cardiol 1996; 77: 37 40.
  • 8
    Ross R. The pathogenesis of atherosclerosis – an update. N Engl J Med 1986; 314: 488 500.
  • 9
    Kadish J. Endothelium, fibrinolysis, cardiac risk factors, and prostaglandins: a unified model of atherogensis. Med Hypothesis 1995; 45: 295 313.
  • 10
    Lijnen HR & Collen D. Endothelium in haemostasis and thrombosis. Prog Cardiovasc Dis 1997; 39: 343 50.
  • 11
    Vogel RA. Coronary risk factors, endothelial function and atherosclerosis; a review. Clin Cardiol 1997; 20: 226 32.
  • 12
    Jensen-Urstad K, Johansson J, Jensen-Urstad M. Vascular function correlates with risk factors for cardiovascular disease in a healthy population of 35 year old subjects. J Intern Med 1997; 241: 507 13.
  • 13
    Anderson TGM, Meredith I, Charbonneau F, Delagrange D, Craeger M, Selwyn et alSystemic nature of endothelial dysfunction in atherosclerosis. Am J Cardiol 1995; 75: 71B 4B.
  • 14
    Anderson TJ, Uehata A, Gerhard MD, Meredith IT, Knab S, Delagrange et al Close relation of endothelial function in the human coronary and peripheral circulations. J Am Coll Cardiol 1995; 26: 1235 41.
  • 15
    Chmielewska J & Wiman B. Determination of tissue plaminogen activator and its ‘fast’ inhibitor in plasma. Clin Chim Acta 1986; 32: 482 5.
  • 16
    Tegner-Nilsson A-C, Friberger P, Gyzander E. Determination of a new rapid plasmin inhibitor in human blood by means of a plasmin specific tripeptide substrate. Scand J Clin Lab Invest 1977; 37: 403 9.
  • 17
    Sorensen KE, Celermajer DS, Spiegelhalter DJ, Georgakopoulos D, Robinson J, Thomas et al Non-invasive measurement of human endothelium dependent arterial responses: accuracy and reproducibility. Br Heart J 1995; 74: 247 53.
  • 18
    Wiman B & Collen D. On the kinetics of the reaction between human antiplasmin and plasmin. Eur J Biochem 1978; 84: 573 8.
  • 19
    Levi M, Roem D, Kamp AM, de Boer JP, Hack CE, ten Cate JW. Assessment of the relative contribution of different protease inhibitors to the inhibition of plasmin in vivo. Thromb Haemost 1993; 69: 141 6.
  • 20
    Marongiu F, Mulas G, Mamusa AM, Mameli G, Marassi PP, Tronci M et al Relationship between elevated fibrinopeptide A levels and alpha2-antiplasmin. Haemostasis 1987; 17: 301 4.
  • 21
    Wilcox JN, Rodriguez J, Subramanian R, Ollerenshaw J, Zhong C, Hayzer D et al Characterizaion of thrombin receptor expression during vascular lesion formation. Circ Res 1994; 75: 1029 38.
  • 22
    Robbie LA, Booth NA, Brown AJ, Bennet B. Inhibitors of fibrinolysis are elevated in atherosclerosis plaque. Arterioscler Thromb Vasc Biol 1996; 16: 539 45.
  • 23
    Sukata Y & Aoki N. Cross-linking of alpha2-plasmin inhibitor to fibrin by fibrin-stabilizing factor. J Clin Invest 1980; 65: 290 7.
  • 24
    Sobel JH & Gawinowicz MA. Identification of the alpha chain lysine donor sites involved in factor XIIIa fibrin cross-linking. J Biol Chem 1996; 271: 19288 97.
  • 25
    Fair DS & Plow EF. Synthesis and secretion of the fibrinolytic components, including alpha2 antiplasmin, by a human hepatoma cell line. J Lab Clin Med 1983; 101: 372 84.
  • 26
    Knot EA, Drijfhout HR, ten Cate J et al Alpha2-Plasmin inhibitor metabolism in patients with liver cirrhosis. J Lab Clin Med 1985; 105: 353 8.

Received 21 January 1998; accepted 16 June 1998.