Platelet aggregation and thrombosis play a central role in the pathogenesis of atherosclerosis, since platelet aggregates and thrombus formation has been found in the coronary arteries at sites of plaque rupture [1]. Platelets, upon their activation, further stimulate thrombus formation and recruit additional platelets by releasing thromboxane A₂ (TXA₂), adenosine diphosphate (ADP), and cytokines, which produce and promote surface thrombin generation and release vasoconstrictor substances [2]. Arterial hypertension, apart from being a major risk factor for atherosclerotic coronary artery disease, is characterized by an increased thrombotic tendency (the thrombotic paradox of hypertension or 'Birmingham paradox') with evidence of platelet activation, hypercoagulation and endothelial dysfunction [3]. Platelet dysfunction has attired attention as a potential cause of increased cardiovascular morbidity and mortality in essential hypertension [4]. All the above suggest the importance of platelet function in coronary artery disease and hypertension.

Platelet function is regulated by several factors that exhibit opposing effects, and the dynamic balance between prothrombotic and anti-antithrombotic factors determines health or thrombosis. Platelet activation is modulated by the function of normal endothelium that exerts a protective antithrombotic effect through the production of nitric oxide (NO) and prostacyclin (PGI₂), substances that cause vasodilatation and inhibit platelet aggregation [2]. Endothelial dysfunction leads to nitric oxide (NO) deficiency which has been implicated in the underlying pathobiology of many of these disorders (NO insufficiency states) [5]. Moreover, endothelin-1 (ET-1, a potent vasoconscrictive substance released by the endothelium) and plasma catecholamines play a major role in the pathogenesis of acute ischemic events, and stimulate platelet aggregation, thus exerting prothrombotic actions.

In contrast to regular physical exercise that beneficially affects cardiovascular mortality [6], acute vigorous exercise that is characteristically accompanied by sympathetic nervous system (SNS) activation may trigger the occurrence of acute ischemic events [7]. Furthermore, epidemiological studies show that acute ischemic events exhibit a circadian rhythm coinciding with increased activity of the SNS and increased platelet aggregation during early morning hours [8]. SNS activation and the subsequent rise of plasma catecholamines result in increased platelet aggregation (possibly by activating platelet α₂-adrenergic receptors), thus offering a possible explanation for the association between SNS activation and acute ischemic events [9].

Although several lines of evidence elucidate the function of each one of these vasoactive substances (catecholamines, TXA₂, prostacyclin, and ET-1) in patients with coronary artery disease, their dynamic interaction remains largely unclarified. The aim of our study was to investigate platelet aggregation during acute submaximal exercise (representing SNS activation) in patients with coronary artery disease and hypertension compared to normal volunteers. Furthermore, we aimed at clarifying the potential contribution of vasoactive substances (such as catecholamines, thromboxane A₂, prostacyclin and ET-1) on platelet aggregation by the simultaneous determination of these factors during acute exercise.

The baseline characteristics of the three groups are presented in Table 1. Normal volunteers were younger, leaner and had lower blood pressure than hypertensives and patients with coronary artery disease.

Systolic blood pressure increased significantly in all three groups during treadmill exercise, while diastolic remained unchanged in patients with coronary artery disease or hypertension and fell, though not significantly, in healthy individuals. Blood pressure at all stages remained higher in the patients with coronary artery disease and hypertension (groups A and B) than in the healthy individuals (Figure 1). The Double Product was determined in all patients at the peak of exercise with no significant differences between the three groups (Table 2). Heart rate at the end of exercise as percentage of the maximum heart rate (predicted by an age based formula; maximum heart rate = 220-age) was determined and there were no significant differences between the three groups (Table 2).

Plasma catecholamines increased significantly (p < 0.01) in all groups during exercise. Plasma noradrenaline at rest was higher in hypertensive patients (p < 0.01) and patients with coronary artery disease (p < 0.05) compared to healthy individuals. At exercise peak plasma noradrenaline increased about 10-fold, and hypertensives had higher noradrenaline levels than normal subjects (p < 0.01) and coronary patients (p = non significant, Figure 2A). Plasma adrenaline levels showed a similar fluctuation. At rest, hypertensive patients had higher plasma adrenaline levels but the difference was not significant. Plasma adrenaline increased during exercise in all three groups, and at exercise peak was higher in hypertensive individuals compared to healthy subjects (p < 0.01) and patients with coronary artery disease (p = non significant, Figure 2B).

The 6-keto-prostaglandin-F1a levels increased significantly (p < 0.05) during exercise in all three groups. Patients with coronary artery disease had lower plasma levels of 6-keto-prostaglandin-F1a compared to subjects from the other two groups and the difference was significant only at the peak of exercise compared with healthy volunteers (p < 0.05, Figure 3A).

Plasma TXB₂ levels showed a significant increase during exercise in all three groups (p < 0.05), reaching their peak at maximum exercise. There was a trend towards higher levels at the end of exercise in patients with coronary artery disease, but there were no significant differences between the three groups (Figure 3B). Plasma TXB₂ levels remained elevated compared to baseline after 10 min of rest.

The ratio of 6-keto-prostaglandin-F1a to thromboxane B₂ decreased during exercise in all groups, but the decrease was statistically significant only for the CAD group (p = 0.01). There was no significant difference between the three groups at rest but at the peak of exercise patients with CAD had lower 6-keto-prostaglandin-F1a/thromboxane B₂ ratio compared with healthy volunteers that almost reached statistical significance (p = 0.053) (Figure 3C). A negative correlation between the PGI₂/TXB₂ ratio and platelet aggregation at the end of exercise test with collagen (p < 0.05) and ADP (p = 0.07) was found in the group with CAD compared with controls, indicating that endothelial dysfunction (reduced ratio PGI₂/TXB₂) contributes to increased platelet aggregation in CAD patients.

Plasma ET-1 levels were higher at baseline in patients with coronary artery disease compared to healthy volunteers (p < 0.05), while during exercise ET-1 increased significantly (p < 0.05) only in hypertensive patients (Figure 3D).

Platelet aggregation decreased significantly during exercise (p < 0.05) in all groups with the use of ADP and collagen as inducing agents. At exercise peak, normal volunteers showed significantly greater inhibition of plasma aggregation when compared with hypertensives (p < 0.05) and patients with coronary artery disease (p < 0.05) with the use of both ADP and collagen (Figure 4A and 4B). After 10 min of rest platelet aggregation remained reduced in healthy volunteers compared to baseline (p < 0.05), while in hypertensives and patients with CAD increased and was higher than in healthy volunteers (p < 0.05).

Thrombotic events represent the most common and important complication of atherosclerotic disease. The clarification of pathophysiologic mechanisms underlying the thrombotic process seems of utmost importance for the prevention of such events and the improved management of these patients. Our study findings indicate that patients with coronary artery disease exhibit a thrombotic tendency compared to healthy subjects during exercise (increased platelet aggregation compared to healthy individuals), while hypertensive patients are lying in between these two groups.

Exercise may induce platelet activation in vivo and enhanced platelet reactivity in vitro, but results concerning exercise-induced platelet aggregation are far from unanimous. Much controversy exists regarding platelet function during SNS activation; studies examining the results of physical and mental stress on platelet aggregation in hypertensives and in patients with ischemic heart disease have shown conflicting results [9‐11]. Platelet responses to exercise depend on several factors, such as exercise intensity, the exercise protocol used, and physical fitness. Moreover, measurements of platelet reactivity either in vitro or ex vivo (aggregability assays) or in vivo (platelet secretory products, mainly beta thromboglobulin and platelet factor 4) are associated with considerable methodologic difficulties and thus may account for the discrepancies of results reported in the literature.

Our findings of decreased platelet aggregation during exercise despite the increase in plasma catecholamines seem paradoxical. Increased shear-stress in states of increased SNS activity results in increased platelet aggregation, probably mediated by the action of adrenaline, as its infusion results in increased platelet aggregation in normotensives and hypertensives [12‐14]. There is evidence that the increased platelet aggregation in the early morning hours is due to increased sensitivity to the action of adrenaline [15]. The stimulatory effect of epinephrine on platelet functions has been linked to activation of α₂-adrenoceptor in the membrane of the platelets, since the effect can be blocked by yohimbine, a specific antagonist of α₂-adrenoceptor [16]; moreover, catecholamines during strenuous exercise have been shown to enhance platelet activation, facilitate the ability of ADP-activated fibrinogen receptors, promote fibrinogen binding to platelet receptors, and more importantly increase the number of α₂-adrenoceptors despite a decrease in their affinity [17]. Thus, the decreased platelet aggregation found in our study might be explained by two hypotheses: a) either noradrenaline that increases much more than adrenaline (five fold more) during exercise exerts different effects on platelet aggregation, or b) other factors induced by physical exercise counter-regulate the negative effects of adrenaline on platelet aggregation.

Regarding the first hypothesis, experimental studies have shown increased intra-arterial thrombosis during adrenaline infusion but decreased thrombosis during noradrenaline infusion [18]. In addition, prolonged exposure of platelets to increased adrenaline levels results in decreased aggregation to adrenaline, probably due to adrenergic receptor downregulation [19]. There is also evidence that β₂ adrenergic receptor stimulation causes inhibition of adhesion and aggregation in platelets through increased NOS activity in platelets [20]. β₂ adrenergic receptor-mediated inhibition of adhesion and aggregation in platelets may partially offset the proaggregatory effects of α₂ adrenergic receptor stimulation in response to the endogenous catecholamines (noradrenaline and adrenaline).

Regarding the second scenario, noradrenaline has been reported to be able to stimulate endothelial cells to release prostacyclin and nitric oxide, both of which are known to be potent inhibitors of platelet aggregation [21]. In addition, the increased blood flow during exercise causes increased shear stress and vasodilatation through the production of NO and PGI₂[22]. The increased prostacyclin production, plasma NO metabolites and platelet cGMP may suppress platelet reactivity [23‐25]. All our patients exhibited increased PGI₂ production during exercise, a potent inhibitor of platelet aggregation. The difference was significant between healthy volunteers and patients with coronary artery disease.

Accumulating evidence suggests that platelets of hypertensive patients show signs of activation²⁵ and an increased tendency to aggregate [27]. Hypertensive patients showed higher urinary excretion rate of 11-dehydro-TXB₂ compared with normotensive controls [28]. Furthermore, exercise is associated with increased TXA₂ production [29‐31]. On the other hand, products of platelet activation (serotonin, TXA₂, ADP) directly increase platelet aggregation. At the same time, these agents act on endothelial cells and cause further release of prostacyclin and NO and inhibit platelet aggregation². During exercise TXB2 production was increased in all groups, more in patients with coronary artery disease and less in healthy volunteers but the difference was not statistically significant. The balance between PGI₂ and TXA₂, as expressed by the ratio of 6-keto-prostaglandin-F1a to thromboxane B₂ was changed in favour of the prothrombotic TXA₂ in patients with CAD, that exhibit the higher degree of endothelial dysfunction.

Increased endothelin-1 levels during exercise may contribute to the pathogenesis of acute myocardial infarction that sometimes follows vigorous physical exercise. Even small increments of ET-1 may facilitate vasoconstrictive responses of coronary vessels to other stimuli. Although ET-1 generation de novo requires several hours there is evidence that ET-1 is stored within internal storage vesicles of endothelial cells, and its secretion is a rapid process with the formation and transit time of a vesicle to the plasma taking approximately 10 min. This could explain the abnormal ET-1 synthesis and release during exercise in our patients with hypertension [32]. The fact that only hypertensive patients exhibited an increase in ET-1 is probably explained by the fact that known mechanisms of ET-1 secretion, as increased SNS activity, pressure and shear stress [33] were higher during exercise in hypertensive patients. Experimental data concerning ET-1 effects on platelet function are contradictory showing decreased [34], increased [35] or unchanged [36] platelet aggregation. The increased levels of ET-1 in hypertensives and patients with coronary artery disease are consistent with the endothelial dysfunction present in these patients and the role of ET-1 in cardiovascular disease.

Endothelial dysfunction is characterized by a reduction in prostacyclin and NO, probably resulting in a thrombogenic vascular environment that allows thrombus formation induced by exposure of highly thrombogenic substances from ruptured or erosive plaques. Even though the differences were not statistically significant, healthy individuals had higher PGI₂ and lower TXA₂ levels, while patients with coronary heart disease had lower PGI₂ and higher TXA₂. The negative correlation between the PGI₂/TXA₂ ratio and platelet aggregation in CAD patients indicates that endothelial dysfunction contributes to the thrombotic tendency in these patients.

A major limitation of our study is that normotensives are younger and leaner than the other two groups, and despite their abstinence of regular exercise this may have influenced favourably their response to exercise. Although evidence regarding the relationship between obesity and platelet aggregation are contradictory [37], data from the Northwick study suggest that obesity (body weight or body mass index) is not associated with platelet aggregation in a large sample of about 1000 subjects [38]. Another important limitation of our study is the lack of oxygen consumption estimation during exercise in order to evaluate exercise intensity; however, the use of the double product and the percentage of maximum heart rate achieved at exercise peak as indirect indices of oxygen consumption (revealing a comparable exercise intensity in the three groups) overcomes this limitation. Finally, in our study, NO metabolites and platelet aggregation after adrenaline induction were not determined; we believe that the evaluation of these parameters, in a clinical setting similar to ours, will significantly contribute to the clarification of the mechanisms underlying thrombosis in CAD, thus representing an excellent idea for further research in this area.

In summary, platelet aggregation was inhibited during exercise in healthy individuals, hypertensives and patients with coronary artery disease. This inhibition of platelet aggregation probably represents a protective endothelial mechanism through the production of PGI₂ and NO. Endothelial dysfunction is probably responsible for the smaller inhibition of platelet aggregation during exercise in patients with coronary artery disease and hypertensives compared to healthy subjects. The negative correlation between the PGI₂/TXA₂ ratio and platelet aggregation in CAD patients indicates that endothelial dysfunction contributes to the thrombotic tendency in these patients. Other causes as defective intraplatelet NO production or resistance of platelets to the action of NO cannot be refuted.