Summary of the coagulation and fibrinolysis pathways. TF indicates tissue factor; TAT, thrombin-antithrombin complex; PTF, prothrombin fragments; FPA + B, fibrinopeptides A and B; PAP, plasmin-antiplasmin complex; α2AP, α2-antiplasmin; PAI-1, plasminogen activator inhibitor 1; tPA, tissue plasminogen activator; and FDPs, fibrinogen degradation products. Solid arrowhead denotes activation.
Lee KW, Lip GYH. Effects of Lifestyle on Hemostasis, Fibrinolysis, and Platelet ReactivityA Systematic Review. Arch Intern Med. 2003;163(19):2368-2392. doi:10.1001/archinte.163.19.2368
Copyright 2003 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2003
The pathophysiology of atherothrombosis in cardiovascular disease is complex and multifactorial. No doubt, lifestyle habits such as exercise, smoking, diet, and alcohol consumption may have significant influence on cardiovascular disease. As the hemostatic system is assuming an increasingly prominent role in the pathogenesis and progression of atherovascular diseases, this review evaluates the effects of lifestyle habits (or lifestyle modifications) on blood coagulation, fibrinolysis, and platelet reactivity.
The process of initiation, progression, and complication of atherothrombosis in cardiovascular disease is complex and can be influenced by multiple factors. Ischemic coronary syndromes such as unstable angina, myocardial infarction (MI), and sudden ischemic death share common pathophysiologic processes characterized by coronary plaque rupture with superimposed thrombus formation.1,2 Indeed, there is substantial experimental and clinical evidence that blood hypercoagulability or thrombogenicity promotes thrombus formation in the circulation, systemically and locally at the exposed atherogenic surface of the disrupted plaque.3
A wide range of factors has been identified in prospective epidemiologic studies to have a systemic effect on blood thrombogenicity. Certainly, there is increasing evidence of a close relationship between the traditional cardiovascular risk factors such as diabetes mellitus, hypertension or hyperlipidemia, and the increased thrombogenicity, which is characterized by hypercoagulability, hypofibrinolysis, or increased platelet reactivity.4- 6 Conversely, improvements of these cardiovascular risk factors have been associated with a lower prothrombic tendency.7- 10 However, the associations and the effects of exercise or physical activity, psychosocial stress, diet, and other lifestyle habits on plasma indicators of thrombogenesis are less well established.
Further evidence of the influence of lifestyle changes on cardiovascular risk factors and clinical outcomes is illustrated by data from salt restriction and blood pressure reduction11 and improved mortality by diets rich in oily fish. In the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto Miocardico Prevenzione (GISSI Prevenzione)12 trial and the Diet and Reinfarction Trial (DART),13 there was a significant reduction in mortality after MI by increasing dietary n-3 polyunsaturated fatty acids (n-3 PUFA) and fish intake, respectively, and the mortality reduction has been partly attributed to a reduction in sudden cardiac death. Indeed, the recent reanalysis of the course of appearance of the effects of n-3 PUFA has showed an early and highly significant reduction of sudden cardiac death.14 However, many instances of sudden death have a thrombotic basis,15 with evidence of thrombus in the left main coronary artery, and sudden death is not simply an arrhythmogenic phenomenon. The aim of this review is to evaluate the effects of lifestyle habits (or lifestyle modifications) on the plasma indices of the 3 main systems of thrombosis: blood coagulation, fibrinolysis, and platelet reactivity.
We performed a search using electronic databases (MEDLINE, EMBASE, and DARE), and additionally, abstracts from national and international cardiovascular meetings were reviewed to identify unpublished studies. Relevant authors of these studies were contacted to obtain further data.
The process of hemostasis and thrombus formation depends on the fine balance between the coagulation and fibrinolysis systems (Figure 1). The slower intrinsic clotting pathway depends on circulating coagulation factors, such as factors IXa and VIIIa. The more rapid extrinsic pathway is activated when blood is exposed to an extravascular factor such as tissue factor. Factor VII (FVII) plays a key role in the initiation of this coagulation mechanism when it forms complexes with tissue factor from a disrupted atheromatous plaque. Activation of the coagulation system induces the formation of thrombin from prothrombin. Thrombin converts fibrinogen into (insoluble) fibrin and induces platelet activation.
The binding of fibrinogen to platelet glycoprotein IIb/IIIa receptor leads to platelet aggregation. Fibrinogen is also the major determinant of blood and plasma viscosity, explaining 50% of the latter. Hence, increased tendency of hemostasis and thrombosis may be reflected in high levels of plasma fibrinogen, FVII, factor VIII (FVIII), thrombin generation, platelet reactivity, and high plasma viscosity. Increased thrombin generation may be indicated by high activation markers, such as prothrombin activation fragment 1 + 2 (F1 + 2) and thrombin-antithrombin complex (TAT), associated with a decrease in clotting time.
On the other hand, activation of the fibrinolytic system induces the conversion of plasminogen to plasmin by plasminogen activators. Tissue plasminogen activator (tPA) is the main fibrinolytic stimulator. Plasmin promotes the degradation of fibrin within the thrombus, disintegrating clots and hence maintaining vascular patency. Fibrin degrades into soluble fibrin degradation products, including D-dimers. The primary inhibitor of the fibrinolytic process is the plasminogen activator inhibitor type 1 (PAI-1), which inhibits plasminogen activation by binding with tPA to form the PAI/tPA complexes. Therefore, impaired fibrinolytic function may be reflected in high plasma levels of PAI-1 or tPA antigen (which evaluates mainly the inactive PAI-1/tPA complexes) and/or indicated by low plasma levels of tPA activity or activation products such as D-dimer and plasmin 2–antiplasmin complex. Reduced plasmin generation leads to suppression of fibrinolytic activity, thus favoring fibrin persistence and thrombosis.
Most of these variables in the coagulation and fibrinolytic systems can be readily assayed using the enzyme-linked immunosorbent assay (ELISA) technique. It is important to distinguish the difference between the measurement of activity and antigen levels of these molecules. Although the antigen levels refer to the total amount of the circulating proteins (both bound and free), activity levels refer to the functionally active portions of the proteins. Thus, elevated antigen levels of a particular molecule do not necessarily reflect an increase in its functional activity. For example, elevated tPA antigen levels are often a reflection of high levels of circulating PAI-1, resulting in a large portion of tPA antigens being bound to PAI-1 and thereby rendering it inactive.16
Many of these systemic thrombogenic factors may be involved in the initiation of early atherosclerotic lesions and contribute to the progression of coronary thrombosis, plaque growth, and its clinical sequelae. For example, plasma fibrinogen has been shown to stimulate vascular smooth muscle migration and proliferation, promote platelet aggregation, and contribute to blood viscosity and thrombi.17 Many of these indices, including fibrinogen, FVII, von Willebrand factor (vWf, a marker of endothelial damage or dysfunction), D-dimer antigen (a marker of cross-linked fibrin turnover), and tPA antigen, have been identified as independent predictors of subsequent cardiovascular events in prospective studies in both healthy subjects18- 22 and those with cardiovascular risk factors23,24 or established coronary heart disease (CHD).25 In addition, platelet hyperaggregation,26,27 plasma viscosity,28- 30 and PAI-1 levels31- 34 have also been associated with cardiovascular morbidity and mortality in both men and women in prospective studies. Thus, the potential modifications of these hemostatic or thrombogenic factors by simple lifestyle changes as both primary and secondary prevention have attracted considerable interest from the public health perspective.
The recognition that the onset of cardiovascular events is frequently triggered by physical exertion or mental stresses has lead to a possible link between neurohormonal activation and coronary atherothrombogenesis. The increase in sympathoadrenal activation may not only trigger plaque rupture but also directly induce a hypercoagulable state, leading to a rapid propagation of occlusive coronary thrombus and, hence, sudden death. Typically, an intravenous infusion of epinephrine is used to mimic sympathetic activation and to examine the adrenergic effects on thrombogenic markers.35,36 The literature suggests a dose-dependent stimulation of FVIII clotting activity, vWf antigen, tPA activity, and platelets within a 15- to 40-minute infusion of norepinephrine. Although in healthy individuals the increase in coagulability may be counteracted by a rapid rise in fibrinolytic activity, such hemostatic balance between coagulation and fibrinolysis may be impaired in subjects with atherosclerotic disease or risk factors, and hence, catecholamine surge may trigger a hypercoagulable state and enhance the odds of overt thrombosis.37- 39
The precise mechanisms underlying the hemostatic changes with sympathetic activation remain unclear. The lack of inhibition by aspirin in the increase of platelet aggregability and platelet secretory activity during norepinephrine infusion40 or exercise37 suggests that platelets are not being stimulated through the cyclooxygenase-dependent pathway. Furthermore, exercise and mental stress induced in platelet-dependent thrombin generation is suppressed by β-blocker therapy but not by aspirin, further support of the important role of sympathoadrenal activation.41
It should be emphasized, however, that lifestyle modifications rarely involve a single component; for example, an increase in exercise activity may accompany concomitant improvement in diet, which may in turn lead to weight loss and better psychological well-being. In addition, such efforts may also modify other known independent cardiovascular risk factors, such as lipid levels or hypertension. It is therefore difficult to separate the effects of these various factors in clinical studies, especially in observational studies, although coincidental confounding variables can be statistically controlled, at least to some extent, with a large sample size and adequate statistical power. Furthermore, differences in aptitudes toward motivation and compliance between studied subjects are difficult to control and may confound the data. These may account for some of the conflicting results seen among studies that investigate the effects of exercise or dietary changes on thrombogenic factors.
Many long-term epidemiologic studies have demonstrated an unequivocal and strong relationship of increased fitness, exercise, or physical activity during leisure time with reduced cardiovascular risk. Regular exercise is known to lower body weight and blood pressure and improve lipid profile (with a decrease in serum cholesterol levels and an increase in high-density lipoprotein-cholesterol [HDL-C] levels). In addition, regular exercise enhances functional capacity and psychological well-being, as well as quality of life. The underlying biological mechanisms through which these beneficial effects are mediated must be interrelated.
There are many reports on the effects of exercise on coagulation, fibrinolysis, and platelet activation. These mainly consist of intervention studies, prospective randomized controlled trials (Table 1), and numerous large, population-based, cross-sectional studies.
Several cross-sectional studies have consistently shown a positive effect toward an antithrombotic state, especially in lowering plasma levels of fibrinogen and improving fibrinolytic capacity by long-term regular exercise.62,63 However, it should be noted that most intervention or randomized controlled trials lack a comprehensive evaluation of both the hemostatic and fibrinolytic variables and thus provide only fragmentary data on the potential changes in hemostasis attributable to physical exercise. Indeed, some studies have yielded conflicting data, and this may be due to variations in exercise protocol or training programs used, populations studied (age, sex, CHD), seasonal factors, and the lack of standardization in the analytical methods used for the assessment of various hemostatic factors, particularly in the assessment of platelet reactivity.
The available evidence from the intervention or randomized controlled trials would suggest that exercise or physical training evokes multiple effects on blood hemostasis in healthy individuals and in patients with atherovascular disease. For example, patients with atherovascular disease have higher basal levels of PAI-1 and lack a similar degree of increase in tPA activity after exercise when compared with healthy subjects.64 In addition, higher thrombin generation has been found in patients with peripheral vascular disease in response to exercise, whereas no such increase was detected in healthy controls.59
While bearing in mind the considerable inconsistency of the results of various exercise studies due to methodologic variations, there are important differences between the effects of moderate endurance physical training and short-term strenuous exercise on both the hemostatic and fibrinolytic variables (Table 2). By and large, regular physical activities of moderate intensity in training programs enhance blood fibrinolytic capacity and possibly also reduce blood coagulation, although the latter remains disputable. Conversely, short-term strenuous exercise seems to induce a hypercoagulable state simultaneously with an increase in fibrinolytic capacity as evidenced by increased levels of fibrinogen, FVIII coagulant, and platelet activities, higher thrombin generation and hemoconcentration, markedly increased tPA activity, and possibly also decreased PAI-1 and tPA antigen levels. The rise in tPA activity is most apparent and seems to be directly proportional to the level of exercise intensity.52,66 However, the increased level of fibrinolytic activity seems to fall sharply dur ing the recovery period, whereas activation of the coagulation cascade is persistent.50,51
This phenomenon has been thought to possibly precipitate acute coronary thrombosis, leading to sudden cardiac death in susceptible sedentary individuals or patients with preexisting atherovascular disease who may not sustain their fibrinolytic capacity (perhaps due to endothelial dysfunction) when they are exposed to unaccustomed, short-term, strenuous physical exertion. However, recent studies have suggested that functional fibrinolytic activity was similar in physically active men with and without a history of MI67 and in older men with hypertension when compared with normotensive subjects.68
Although the results of the studies that investigated the effects of short-term exercise on fibrinolytic markers are more consistent, studies that investigated plasma fibrinogen concentration have produced conflicting data. Several studies have reported a significant increase in plasma fibrinogen after strenuous exercise,54,69,70 but others using different protocols have shown either no significant effects52,71,72 or even a reduction in plasma fibrinogen level after short-term, intense physical exercise.53 It is possible that the changes in coagulation markers depend on the type of physical exercise to which subjects are subjected. Cerneca et al54 demonstrated that this might be the case, since rowers, marathon runners, and healthy controls revealed a significant increase in plasma fibrinogen levels after near-maximum exercise tests, whereas weightlifters showed no significant change.
Interestingly, genetic factors might also explain the different effects of exercise on hemostatic or fibrinolytic factors. For example, Montgomery et al70 investigated the effects of long-term physical training and short-term, intensive exercise on plasma fibrinogen levels in 156 men in the British Army and found that subjects carrying the A allele of the G453A polymorphism in the β-fibrinogen gene showed a higher increase in plasma fibrinogen than men with the GG genotype.
Various studies have consistently reported a significant improvement in fibrinolytic capacity following regular exercise or physical training. The increase in fibrinolysis is indicated by a decrease in PAI-1 levels and also a rise in tPA activity. Indeed, Szymanski et al73 demonstrated that persons who are habitually active have the lowest basal PAI-1 activity but the highest increase of tPA activity in response to exercise when compared with inactive subjects. This effect has been repeatedly demonstrated using various exercise intensity and duration.46,47,72,74- 76 One study77 reported lower PAI-1 levels in those participating in regular sporting activities than the respective age-matched sedentary individuals or elderly athletes and post-MI patients, but tPA activities were significantly higher after exercise in those with lower pretest PAI-1 level.
Most exercise studies43,50,51,69,78- 80 of varying degrees of intensity and duration have been found to induce a significant increase in FVIII coagulant activity. However, others have demonstrated that regular training exercise does not seem to induce significant effect on resting or postexercise levels of FVIII activity and antigen in normal healthy and sedentary subjects,45,48,55,81 although 4 weeks of physical training has been shown to lower resting levels of FVIII activity and antigen in post-MI patients.82 Although high levels of plasma fibrinogen are usually found in patients with CHD or cardiovascular risk factors, randomized controlled trials have found mixed results in fibrinogen levels in response to regular physical activity.46,76,83- 85 For example, 12 weeks of aerobic exercise training in sedentary hypertensive subjects lowered blood pressure, left ventricular mass, and plasma fibrinogen levels, but changes returned to baseline values (except left ventricular mass) at 2 months after detraining.42 Similarly, 6 months of intensive endurance training reduced plasma fibrinogen in elderly men but not in young men.46 On the other hand, Schuit et al47 reported a significant increase in plasma fibrinogen levels in elderly males after a same duration of intensive training, whereas Ponjee et al48 reported no change in plasma fibrinogen levels in both males and females after 24 weeks of training.
Only a few reports80,83,86- 88 on the effects of exercise on FVII are available and, again, with mixed results. Moderate exercise has no significant influence on FVII or at the most the effect is only relatively short-lived. Studies89- 93 on the effects of exercise on fibrinopeptide A (a marker of thrombin activity and fibrin formation) have again produced conflicting results, although raised plasma markers of thrombin generation (TAT and F 1 + 2) with short-term exercise had been reported. The overall mechanisms underlying these changes in coagulation or fibrinolysis, in response to short-term or long-term endurance exercise, are poorly understood and still remain speculative, but interactions involving neurohormonal pathways are very likely.59,94- 96
The effects of exercise on platelet aggregation and activation have been extensively studied, but the results are highly variable.97,98 It is noteworthy that measurements of platelet reactivity either in vitro or ex vivo aggregability assays or in vivo platelet secretory products (mainly β-thromboglobulin and platelet factor 4) are associated with considerable methodologic difficulties and thus may account for the discrepancies of results reported in the literature.
By and large, short-term, strenuous exercise induces a transient increase in agonist-induced platelet aggregation both in vitro and ex vivo and an increase in platelet counts, adhesiveness, and in vivo platelet secretory activity. Overall, these effects seem to be more pronounced in sedentary than healthy subjects.99,100 In contrast, long-term endurance physical training (preconditioned) in men and women at moderate intensity (50%-55% of peak oxygen consumption) seems to suppress platelet adhesiveness and aggregation both at rest and after short-term, strenuous exercise. However, the effects reversed back to the pretraining state after a period of deconditioning and hence the importance of regular moderate exercise to maintain such potential benefits.56,101
Thus, as with fibrinolytic response, platelet reactivity in response to exercise also seems to be both duration and intensity dependent. However, the underlying mechanisms remain unclear. Increases in catecholamine concentrations and shear stress are probably important.102,103 Perhaps the short-term response may be related to the release of tPA from endothelial cells associated with higher catecholamine release during exercise. Interestingly, the fact that aspirin treatment had no significant influence on platelet activation induced by heavy exercise in patients with stable angina pectoris and matched healthy controls suggests that the response may not be cyclooxygenase-pathway dependent.37,38 This implies that aspirin may have a limited antithrombotic effect during physical exercise and probably also in other situations with increased catecholamine levels such as during acute psychological stress. The chronic platelet response, however, may be related to nitric oxide release as a consequence of regular low-to-moderate exercise training.101
There is increasing evidence that patients with chronic atrial fibrillation are associated with a prothrombic or hypercoagulable state. We demonstrated that short-term exercise to exhaustion significantly increased plasma fibrinogen and lower PAI-1 levels but had no influence on vWf or soluble P-selectin levels in patients with atrial fibrillation when compared with age- and sex-matched patients in sinus rhythm.60 In another similar exercise study in patients with stable congestive heart failure, we also found that plasma viscosity, fibrinogen, and hematocrit levels were significantly increased, both immediately after exercise and at 20 minutes into the recovery period.61
Overweight and obesity, assessed either by body mass index (BMI), a measure of weight in kilograms divided by the square of height in meters, or waist-to-hip circumference ratio (WHR), are associated with increased cardiovascular morbidity and mortality.104,105 Indeed, there is increasing evidence that moderate weight loss could result in regression of coronary arterial lesions and significantly reduces cardiac events and total mortality.106
Most of the present data on thrombogenic profile in overweight or obese persons relate to PAI-1 and tPA antigens. For example, both BMI and WHR correlate strongly and positively with hemostatic factors but negatively with fibrinolytic activity.107 It has been shown that women with high WHR have significantly higher fibrinogen and PAI-1 levels compared with obese women with a low WHR or with lean women.108 Similarly, after adjusting for other lifestyle variables, obese men (BMI >30) had 50% higher PAI-1 activity and 30% higher tPA antigen when compared with men of "ideal" BMI (<25).107 In addition, high fibrinogen and plasma viscosity have also been found to be associated with increasing BMI,107,109,110 although, overall, there is only little evidence that weight reduction reduces plasma fibrinogen or viscosity levels.106,111 On the other hand, there is plenty of evidence that weight reduction by regular exercise and dietary changes reduces PAI-1 and tPA antigen levels,33,112 suggesting a causal relation (Table 3).
Recent evidence has shown that elevated plasma PAI-1 activity seen in obese individuals may be caused by increased PAI-1 release from visceral adipose tissue.122,123 However, a liposuction procedure that removes visceral adipose tissue and achieves a weight reduction of 5% after 3 months without change in lifestyle does not seem to significantly reduce plasma levels of vWf, fibrinogen, or PAI-1.117 In contrast, surgical removal of adipose tissue in 19 patients with morbid obesity with a mean body weight reduction of 50 kg at 6 months and 64 kg at 12 months has led to a significant reduction in FVII, fibrinogen, and PAI-1 activity and a slight increase in tPA activity. Therefore, it seems that a large amount of adipose tissue may need to be removed artificially before an improvement of hemostatic and fibrinolytic profiles could be detected, or it might be that a change in lifestyle, including increased exercise and dietary control leading to weight reduction, is a prerequisite for improvement in coagulation and fibrinolysis. The latter seems more likely the case, since there is evidence that limited weight loss (<7 kg) by lifestyle modifications alone could lead to a reduction of hemostatic factors, FVII, and PAI-1 levels, while increasing the tPA activity.106
Indeed, obesity and syndrome of insulin resistance are inextricably linked with hypertriglyceridemia, hyperinsulinemia, hypo-HDL-cholesterolemia, glucose intolerance, and hypertension. It is known that both triglyceride and insulin resistance correlate strongly and positively with PAI-1.124- 126 In fact, dietary intervention with a low-saturated-fat diet127 or gemfibrozil treatment128 that lowers serum triglyceride levels has been accompanied by improvement and even normalization of the fibrinolytic activity. Thus, it is plausible that weight reduction improves fibrinolytic capacity via modifications in both lipids and insulin resistance profiles. However, data on the interactions between physical activity and diet and hemostasis are scarce, and it is likely that moderation in both efforts would yield a more powerful impact on coagulation and fibrinolysis systems than either lifestyle modification alone. Clearly, more studies are needed to dissect such complex interactions.
The role of dietary changes in modifying CHD risk has been well established.129 The recent Lyon Diet Heart Study130 reported a 50% to 70% lower risk of recurrent heart disease as measured by different combinations of outcome measures, including cardiac death and nonfatal myocardial infarctions in survivors of first-MI patients who received a Mediterranean diet (with more fish, more fiber, but less fat) supplemented with the precursor of n-3 PUFA, α-linolenic acid (18:3n-3, derived mainly from vegetable or seed oil) when compared with controls who received usual care. The study is in parallel with the results of other secondary prevention dietary trials, namely, the DART and GISSI Prevenzione trials,12,13 which similarly used a diet with low intake of total and saturated fats and/or increased intake of oily fish rich in n-3 PUFA. Indeed, at least 2 servings of fish per week, especially fatty fish, equivalent to an intake of n-3 PUFA approaching 1 g/d have been recommended by the American Heart Association.131
The experimental group in the Lyon Diet Heart Study had higher plasma levels of oleic acid, α-linolenic acid, and eicosapentaenoic acid (EPA, 20:5n-3). In the GISSI Prevenzion trial, patients received daily doses of n-3 PUFA as 1 gelatin capsule containing EPA and docosahexaenoic acid (DHA, 22:6n-3) as ethyl esters. In the DART study, at least 3 servings of fatty fish or approximately 15 fish oil capsules per week led to a significant 29% reduction in both cardiac and total mortality within 4 months. The low incidence of cardiovascular disease among Greenland Eskimos and coast land Japanese has been related to high intake of the marine n-3 PUFA: EPA and DHA. However, how particular lipid constituents in these diets contribute to coronary risk is unknown. The rapidity of onset of the beneficial effects seen in these studies suggests that the diet might have anti-inflammatory, antithrombotic, and even membrane stabilizing and hence antiarrhythmic effects besides lowering the rate of progression of atherosclerosis. Indeed, the effects of dietary manipulations or supplementation with individual or complex dietary lipids on thrombogenic variables have attracted considerable interest with a particular emphasis on n-3 PUFA, although the interplay between these lipid constituents and the coagulation system remains largely unclear.
Thus far, only indirect evidence links dietary saturated fatty acids with enhanced thrombogenesis in humans.132,133 When unsaturated fatty acids of the n-9, n-6, or n-3 families replace saturated fatty acids in the diet of experimental animals, the development of atherothrombosis was inhibited, but the doses supplemented tended to be much higher than in human clinical studies.134,135 Data from the Coronary Artery Risk Development in Young Adults136 study showed that usual intake of fish or dietary supplementation with α-linolenic acid, EPA, and DHA was not associated with levels of FVIII, fibrinogen, or vWf and, hence, suggests that usual customary intakes of fish and n-3 PUFA in populations that generally do not consume large amounts of these food items are not associated with these hemostatic factors. Similar results were also found in the PRIME (Prospective Epidemiological Study of Myocardial Infarction) substudy,137 which showed no relationships between fatty acids and fibrinogen, vWf, PAI-1, or FVII levels. By contrast, results from another cross-sectional study138 suggest that increases in dietary n-3 PUFA intake from fish is negatively associated with fibrinogen, FVIII, and vWf and positively associated with protein C levels. Such differences have been attributed to higher EPA and DHA intakes in the latter study.
There are a large number of intervention studies on the effect of n-3 PUFA in the form of fish oil capsules, fishmeals, or its precursor α-linolenic acid on various hemostatic factors (Table 4). We discuss its effects on coagulation, fibrinolysis, and platelet reactivity.
The effect of n-3 PUFA on fibrinogen and blood rheology has been extensively studied. Reductions in fibrinogen (the largest contributor to plasma viscosity) and increased erythrocyte flexibility (a major component of whole-blood viscosity) would be desirable for vascular benefit. By and large, however, many studies have shown no or little improvement in fibrinogen levels after giving n-3 PUFA supplements to different types of patients.140,147,157,158,161- 163 One study164 has reported an increase in erythrocyte flexibility but an unaltered fibrinogen level. Indeed, n-3 PUFA has been linked to improvement in erythrocyte flexibility with lower whole-blood viscosity in few studies,156,165,166 although others have yielded little effects.167- 169 Simply measuring erythrocyte infiltration might be an inadequate method for detecting small but significant differences in erythrocyte flexibility, and this may account for the inconsistency of results among studies.
There is also little agreement on the effects of n-3 PUFA on clotting factors, such as FVII or FVIII. One author170 has reported reduced levels of FVIII with n-3 supplements, but most authors145,171- 176 have found either no influence on or an increase in FVII or FVIII by fish oil ingestion or n-3 PUFA diet. Studies on the intake of n-3 PUFA and vWf concentration have also been similarly conflicting. Most studies136,171,172,174,177,178 have not been able to show an effect with these fatty acids. In addition, few studies have even suggested that n-3 PUFA, including α-linolenic acid, may have antithrombotic effects by enhancing protein C activity,171,174 increasing tissue factor pathway inhibitor,175 or reducing the expression of procoagulant tissue factor activity on monocyte ex vivo.179
The data available on the effects of n-3 PUFA on fibrinolytic activity are also inconsistent. Since the initial study by Barcelli et al,153 which suggested that n-3 PUFA may enhance plasma fibrinolysis, others had found no change in PAI-1 level or in tPA antigen after dietary intervention,145,148,180 and few even reported significant increased in PAI-1 activity.146,160,181,182 Moreover, data available on tPA activity also seem contradictory.145,148,176,183 However, a study has reported that tPA antigen level was inversely related to n-3 PUFA derived mostly from fish oil (EPA, docosapentaenoic acid, and DHA) but not with n-3 PUFA from vegetable origin (α-linolenic acid).137 This is in agreement with a large intervention study184 in diabetic subjects that showed a decrease in tPA antigen after fish supplementation but no effect on PAI-1 activity with high α-linolenic acid diet intake in a double-blind intervention trial.185
Many studies have reported a positive correlation between serum triglycerides and PAI-1 activity. Dietary interventions, such as a low-saturated-fat diet127 or gemfibrozil treatment128 to lower serum triglyceride levels, have been accompanied by improvement and even normalization of the fibrinolytic activity. However, dietary intervention with n-3 PUFA, which is well known for its ability to lower triglyceride levels, has not been shown to be parallel with a decrease in PAI activity,146 indicating that a causal relationship is unlikely between levels of triglycerides and PAI-1 activity during dietary supplementation with n-3 PUFAs. In fact, by pooling data from all these studies, Hansen et al146 were able to calculate that approximately a 17% increase in PAI-1 activity during intervention could be attributed to the fish oil supplement. Notably, there are few data on the effect of n-3 PUFA on D-dimer.184,186
The effects of n-3 PUFA on platelet reactivity have been extensively investigated. Studies of Greenland Eskimos have shown that very high intake of marine n-3 PUFA markedly inhibited platelet reactivity, lowered platelet count, prolonged bleeding time, decreased the ratio of proaggregatory thromboxanes to antiaggregatory prostacyclins, and caused favorable changes in lipid and lipoprotein profiles. These findings are of importance for their low incidence of CHD.187 Although few studies144,174,188 have shown no significant influence on platelet reactivity with n-3 PUFA supplementation, most other intervention studies have demonstrated significant inhibition in platelet reactivity of one sort or another but with conflicting combinations of effects with different agonists in vitro or ex vivo.189- 195 It seems that platelet aggregation induced by low-dose collagen was the most commonly reported index to be influenced. One study152 has demonstrated no significant difference between supplemented α-linolenic acid from vegetable oil and n-3 PUFA from a marine source (EPA and DHA) in their effects on collagen-induced platelet aggregation and thromboxane production, aggregation to the thromboxane A2 mimic, urinary excretion of 11-dehydro-thromboxane B2 and β-thromboglobulin, bleeding time, plasma fibrinogen concentration, antithrombin III activity, FVII coagulant activity, or PAI-1 activity. However, another study has shown that a high α-linolenic acid diet has no significant effect on thromboxane production and platelet aggregation with collagen.185 Notably, data from studies195 on other fatty acids (mainly n-6 PUFA), such as linoleic acid on platelet reactivity, were highly variable, especially in the in vitro assessment of platelet aggregations. Similarly, there was also lack of agreement on the effect of n-3 PUFA on platelet adhesiveness.150,151,196- 198 Novel, well-validated methods for measuring platelet aggregation are desperately needed to solve current controversies.195
Thus far, it seems that there is little evidence to support the hypothesis that changes in coagulation, platelet reactivity, or fibrinolysis systems could account for, at least, some of the beneficial effects afforded by a Mediterranean diet with n-3 PUFA supplementation in the secondary prevention trials mentioned herein. The major effect of n-3 PUFA may be antiarrhythmic rather than antithrombotic.199,200 It remains unsettled whether the diverse effects of n-3 PUFA supplementation on thrombogenic indices are due to different time of supplementation, patient type, or separate effects of EPA and DHA in a mixture of fish oils. It is likely that fish oil and n-3 PUFA have multifaceted actions in the secondary prevention of cardiovascular disease.201
Light-to-moderate alcohol consumption (<30 g/d, ie, 1 to 2 drinks per day) is associated with 10% to 40% lower risks of MI and cardiovascular death compared with abstinence. However, heavy alcohol consumption or binge drinking increases such cardiovascular risks, including stroke.202 The reduction in cardiovascular risks with moderate alcohol intake has mainly been attributed to an increase of HDL-C levels, but this only accounts for 50% of the protective effect.203,204 Increasing evidence has indicated that thrombogenic factors may play an important role in mediating such a complex association independent of HDL-C levels.205,206
The mass of the previously published data on alcohol consumption and hemostasis comes from epidemiologic studies, with only few experimental data reported. Generally, most cross-sectional studies have shown that light-to-moderate alcohol intake is associated with a more favorable coagulation and fibrinolytic profiles as indicated by lower levels of fibrinogen, white blood cell count, plasma viscosity, FVII, and vWf, as well as lower platelet count and activity.107,207 However, heavy or binge alcohol intake is associated with lower fibrinolytic capacity with relatively greater increase in PAI-1 and tPA antigen than tPA activity.205,208 In addition, heavy alcohol consumption also seems to shift the pendulum toward a more procoagulant state, with a rise in the plasma levels of FVII, fibrinogen, and viscosity. Indeed, this may sufficiently predispose individuals to thrombosis, and in the presence of an impaired fibrinolytic state, this may contribute to the increased incidence of ischemic stroke seen in heavier or binge drinkers.20,209 Thus, the results from these studies seem to partly explain the complex relationship between level of alcohol consumption and cardiovascular risk seen in large epidemiologic outcome studies on alcohol.
The results from experimental studies on the effects of moderate alcohol consumption in both healthy subjects and subjects with CHD are contradictory (Table 5). In particular, the alcohol effects on fibrinogen are variable. For example, Pellegrini et al212 found a decrease in fibrinogen level after consumption of 30 g of alcohol that consisted of red wine and alcohol diluted in fruit juice for 4 weeks but found no change in fibrinogen level after consumption of dealcoholized red wine. Because alcohol diluted in fruit juice had an effect similar to that of red wine, it seems that alcohol is the effective mediator in alcoholic drinks. On the other hand, Gorinstein et al210 studied the effect of beer (20 g/d of alcohol) over 30 days and found no change in fibrinogen levels. It could be argued that there may be other substances in beer that inhibit the beneficial effect of alcohol on fibrinogen. However, this may also be due to differences in study design, timing of blood samplings, quantity or regularity of intake (besides type of beverage used), and intrapatient and interpatient variability in alcohol metabolisms. A recent meta-analysis223 of all experimental studies that assessed the effects of moderate alcohol intake on lipid levels and hemostatic factors has concluded that moderate alcohol intake of 30 g/d is causally related to an overall 24.7% lower risk of CHD through favorable changes in lipids (higher HDL-C level) and hemostatic profile (lower plasma fibrinogen levels). However, the precise mechanism(s) by which moderate alcohol intake decreases fibrinogen and increases HDL-C levels is not known.
Most reports seem to indicate that short-term alcohol ingestion leads to inhibition of the fibrinolytic system through a rise in circulating PAI-1 levels.211,212,216,217,220,221 Notably, a recent study217 has shown an acute, dose-dependent rise in PAI-1 antigen level with a parallel prolongation of whole blood clot lysis time after intake of the high dose of red wine. In addition, there was also a tendency for tPA antigen to increase dose dependently, although this was only significant for the high-dose wine group. In the case of binge drinking, this inhibition in fibrinolysis effect persists into the morning following the evening of alcohol consumption,216 and when it coincides with the physiologic morning dip in fibrinolytic activity, this may predispose susceptible individuals to sudden cardiac death. However, the exact reason for decreased fibrinolysis after short-term alcohol intake is still unresolved. Indeed, several studies have discussed whether it is the ethanol component itself or other mediators in red wine (or beer) that induce acute changes in tPA or PAI-1. Veenstra et al221 and Hendriks et al218 have pointed out that the red wine effect on t-PA and PAI-1 antigen levels is probably caused by the effect of ethanol but not the effects of other mediators, such as the phenolic compounds in red wine or port wine, although these constituents might contribute to the effects on platelet function. However, it is difficult to distinguish between the effects of red wine and alcohol per se. Notably, in a study211 with longer-term consumption of beer (4 weeks), the tPA antigen and PAI-1 levels were increased markedly.
Alcohol has also been thought to reduce CHD risk by decreasing platelet reactivity. Indeed, several studies in humans and animals have demonstrated that the immediate effect of light-to-moderate alcohol, either added in vitro to platelets or 10 to 20 minutes after ingestion, can inhibit platelet aggregation to most specific agonists (adenosine diphosphate [ADP], thrombin, collagen, epinephrine) in platelet-rich plasma. This platelet inhibitory effect seems to persist for several hours after alcohol intake.224 However, such beneficial effect is not seen in binge drinkers or in individuals with alcoholism after alcohol withdrawal; instead, a rebound phenomenon of platelet hyperaggregability (especially toward thrombin agonist in vitro)225 and loss of the normal circadian periodicity of the hemostatic system is observed.219 This may explain the increased ischemic strokes or sudden deaths that are known to occur after episodes of binge or heavy drinking.226 Intriguingly, such a rebound phenomenon is not observed after moderate red wine consumption in humans and, in fact, this protection afforded by red wine has been duplicated in rats by alcohol with grade tannins added, which contain the polyphenolic compounds with which red wines are richly endowed. However, it is still unclear how red wine or wine phenolics in particular could significantly inhibit platelet aggregation.
In an interesting study by Lacoste et al,222 rather than evaluating platelet function and platelet inhibition, the authors assessed the effect of alcohol directly on platelet-dependent thrombosis in 12 healthy subjects in an ex vivo model that simulates a deep arterial wall injury exposed to shear forces typical of flow at sites of stenosed arteries, reflecting the in vivo situation of coronary thrombosis. The study demonstrated for the first time that moderate alcohol consumption (24 g of alcohol) in humans had a potent extracorporeal antithrombotic effect both at the time of peak alcohol concentration and 6 hours after alcohol ingestion when blood alcohol level has returned to baseline.
Overall, the balance of anticoagulant, procoagulant, and fibrinolytic effects in any individual in response to alcohol intake may vary, depending on quantity and type of alcoholic beverage ingested and other variables.204,227 The lower level of plasma fibrinogen with moderate alcohol intake may well contribute to the apparent protection alcohol confers against ischemic coronary and cerebral events. On the other hand, consistent evidence suggests that the relatively greater increase in PAI-1 and tPA antigen than tPA activity and the rebound phenomenon of platelet hyperaggregability with short-term alcohol (binge) intake may attenuate this benefit, resulting in a net antifibrinolytic effect of ethanol consumption,205 predisposing individuals to coronary thrombosis and contributing to the increased incidence of ischemic stroke.
The direct effects of smoking on atherothrombogenesis are still unclear. This may be mediated by its many adverse effects on endothelial function, vascular tone, hemostasis, lipid profile, and inflammatory cells. Much of the data regarding the effects of smoking on thrombogenic factors have been derived from epidemiologic and cross-sectional studies. However, intervention studies are accumulating and have reported an increase in blood coagulability but impaired fibrinolysis in habitual smokers when compared with nonsmoking controls (Table 6). It seems that higher plasma levels of fibrinogen and viscosity are the main contributors to higher coagulability found in smokers, whereas the lower fibrinolytic potential is mainly attributed to an increase in PAI-1 activity and possibly also a decrease in tPA activity and lower plasminogen levels.107,228,229,244
In cross-sectional epidemiological studies, lifetime duration of smoking is a strong determinant of initial plasma fibrinogen levels. The effects of smoking on the hemostatic system remain for many years before an exsmoker reverts to a plasma level similar to that of a lifetime nonsmoker,229,245 although a decrease in fibrinogen levels follows quickly after cessation of smoking.246 In prospective data, smoking cessation and the adoption or resumption of smoking are associated with a decrease or an increase, respectively, of approximately 0.15 g/L in plasma fibrinogen. There is evidence to suggest that these changes are not due to concurrent changes in other lifestyle variables.107
Impaired fibrinolytic potential has been found in smokers with CHD247 and peripheral vascular disease,248 with higher levels of PAI-1 activity and tPA antigen than in nonsmokers or light smokers. However, other groups have reported no difference in markers of fibrinolysis and coagulation, although endothelial function has been found to be significantly impaired in healthy smokers compared with nonsmoking controls.231 Interestingly, smokers may have markedly impaired acute substance P–induced endothelial release of active tPA in vivo from coronary and brachial arteries and was closely related to impaired endothelial function in the correspondence arterial beds, which suggests a possible direct link among impaired endogenous fibrinolysis, endothelial dysfunction, and arterial atherothrombosis in smokers.232,233,249 However, although the basal level of PAI-1 activity is higher in long-term smokers, rapid smoking of 2 cigarettes in these patients neither stimulated fibrinolysis nor changed levels of tPA or PAI-1 activities.236 Plasma PAI-1 antigen seems to correlate with cumulative smoking in pack-years,230,236 and on the other hand, other studies75 have suggested that smoking cessation of at least 6 months was associated with a decrease in plasma PAI-1 activity.
Smoking is also known to be associated with increased platelet thrombus formation,250 but studies on the effects of smoking on platelet reactivity have produced conflicting data. For example, rapid smoking of 3 cigarettes in habitual smokers increased ADP-induced platelet aggregation in vitro in one study,235 but others have shown no differences in various agonist-induced platelet aggregations in vitro compared with nonsmokers.237 Certainly, no immediate influence in platelet activity (as indicated by platelet count and soluble P-selectin) occurs after rapid smoking of 2 cigarettes in sequence compared with nonsmoking controls, despite evidence of endothelial damage (indicated by elevated vWf level).241 On the other hand, increased P-selectin expression on platelets has been demonstrated in young, healthy, habitual smokers compared with nonsmokers, underlining the increased platelet activation in nicotine-abusing subjects.239 Importantly, 100 mg/d of aspirin did not reduce platelet activation as measured by unchanged P-selectin expression on platelets and circulating P-selectin plasma levels. This could indicate that enhanced thromboxane A2 production may not be the primary mechanism for increased P-selectin expression in smokers.239 Similarly, dipyridamole alone or in combination with aspirin did not have any significant effect on plasma concentrations of β-thromboglobulin, platelet factor 4, the circulating endothelial cell count (indicates endothelial damage), and the platelet aggregate ratio in habitual, male smokers with CHD.251 Smoking cessation for 6 weeks, however, has resulted in a 29% reduction of circulating P-selectin plasma levels in healthy smokers252 and a decrease in platelet count 2 weeks after cessation.253 Furthermore, aspirin abolished the major cigarette smoke–induced endothelial damage and platelet hyperactivity in the presence of high plasma nicotine levels.240 Notably, passive exposure to tobacco smoke also seems to raise the endothelial cell count and platelet aggregate ratio in a manner similar to that previously observed with active smoking.254
The increase in mental stress, demands at work, anger, or low socioeconomic strain has been associated with an increased risk of atherovascular disease.255- 257 The increased cardiovascular risk may be secondary to excessive cardiovascular reactivity to stress258 but may also involve activation of the coagulation and fibrinolysis systems.
Indeed, an association between psychological factors and several of the coagulation and fibrinolysis variables related to atherosclerosis has provided a plausible psychobiological link to CHD. The characteristic patterns of coagulation and fibrinolysis activation in response to various psychological stressors seem to follow closely that of physical or exercise-induced changes in markers of thrombogenesis. Accordingly, as in short-term, strenuous exercise, acute mental stress simultaneously activates the coagulation system, with increased levels of fibrinogen, total plasma protein, hematocrit, FVII, and FVIII,259,260 and enhances fibrinolysis with increased activity of tPA within a physiological range in healthy subjects.260,261 In patients with atherosclerosis and impaired endothelial anticoagulant function, however, procoagulant responses to acute stressors may outweigh anticoagulant mechanisms and thereby promote a hypercoagulable state.
Similarly, long-term psychosocial stressors, such as prolonged job stress or low socioeconomic strain that provoked a state of vital exhaustion, have been independently associated with hypofibrinolysis, with an increase in PAI-1 and a decrease in tPA activities.262,263 In addition, such long-term mental stress is also independently related to an increase in prothrombic tendency, with increased levels of fibrinogen, FVII antigen, and activity.264 Changes in hemostatic variables in response to psychosocial job stress are particularly interesting. Significant elevation in coagulation FVII and FVIII levels, fibrinogen level, thrombocyte count, thrombin level, and ADP-induced platelet aggregation has been reported during a period of increased workload compared with a calm work period.265 High job demands have also been significantly related to decreases in tPA activity (ie, lower fibrinolytic capacity, independent of other traditional cardiovascular risk factors)266 and hence increase the likelihood of fibrin deposition. The mechanism(s) underlying these changes is unknown, but impaired fibrinolysis in people with long-term psychosocial stress has been linked to insulin resistance, obesity, and triglyceride levels.262
Platelet reactivity also seems to be affected by a variety of psychosocial stressors. Acute mental stresses significantly induce all platelet reactivity variables, such as platelet activation or secretion and in vitro or ex vivo platelet aggregation, in parallel to a concomitant incremental increase of various hemodynamic indices that follow during mental-stress testing. Such a response has been consistently shown in almost every study. Furthermore, these changes seem to be more pronounced in patients with atherovascular disease37,267 compared with healthy controls. Hence, such a response may precipitate acute ischemic coronary events in patients at high risk of cardiovascular events, including individuals with sedentary lifestyle. One group has reported that dipyridamole attenuated the platelet hyperreactivity in post-MI patients but had no effect on stress-induced increase of hemodynamic variables and epinephrine levels.267
Aspirin has only a minimal effect on physical, psychosocial, or norepinephrine stress–induced platelet activation, which suggests that platelets are not being stimulated through the cyclooxygenase-dependent pathway.37,38,41,268,269 However, the mechanism(s) responsible for the increased of prothrombic tendency secondary to psychosocial stress may be related to the sympathoadrenal pathways, but clearly this needs further exploration.
The possible link between coffee or caffeine consumption and the risk of CHD is far from settled, but its effects on various thrombogenic factors might be relevant. However, limited data from intervention study are currently available.
Two well-described randomized controlled trials270 have reported that brewed or boiled coffee, caffeine-containing drinks, and decaffeinated drinks did not have any effects on hemostatic variables, such as fibrinogen level, FVII activity, FVIII antigen, and protein C and S levels. Another experimental study271 seems to support such observations, but another cross-sectional study272 reported an increased in plasma fibrinogen levels with increased coffee consumption.
The available evidence seems to suggest that coffee enhances fibrinolytic potential as whole blood fibrinolysis time is shortened273 and PAI-1 levels are decreased, whereas tPA activity increases274 after consumption of coffee and such effects are blunted during caffeine abstinence. However, one study271 did not find an effect of abstinence from caffeine on blood clot lysis time.
The effects of caffeine intake on platelet activity are more variable. Several in vitro and in vivo studies275,276 have reported increased platelet activation and release after coffee consumption, but others have found the opposite effects.271,272,277 Again, this may be due to the lack of standardization in the analysis methods used to assess platelet reactivity.
Previous epidemiologic studies have suggested that tea consumption is associated with a decreased risk of cardiovascular events, but a recent meta-analysis278 has reported no significant association and, in contrast, the risk may be even increased for CHD in the United Kingdom and for stroke in Australia with increasing tea consumption. Indeed, the antioxidative polyphenolic flavonoids found in tea have been shown to prevent oxidation of low-density lipoproteins both in vitro and in vivo279 and to inhibit platelet aggregation in vitro.280- 282
However, recent randomized controlled trials283- 285 of black or green tea or tea extracts have found no effects on both hemostatic and fibrinolytic variables (eg, fibrinogen, vWf, or FVII and PAI-1, tPA, or urokinase-type plasminogen activator) or on inflammatory markers such as C-reactive protein. Similarly, no significant difference was found from black tea consumption on ex vivo platelet aggregation in patients with CHD286 and on in vitro platelet aggregation in healthy subjects257 when compared with drinking hot water. Interestingly, in the same study, the latter group had found significantly lower (15%) soluble P-selectin levels (but not other adhesion molecules) in those who drank black tea; however, whether such a finding is of any clinical significance is unclear. Thus, it seems that the putative protective effect of tea against development of CHD may not be mediated through effects of tea consumption on hemostasis, fibrinolysis, or platelet activity.
A recent Cochrane Systematic Review has concluded that cardiac rehabilitation with either exercise alone or exercise as part of a comprehensive rehabilitation program in post-MI and postrevascularization patients significantly reduced all-cause or total cardiac mortality by at least 26% to 31%.287 In addition to the reduction of cardiovascular morbidity and mortality, cardiac rehabilitation also significantly improves functional capacity and quality of life and lipid profile and blood pressure.288
Thus, given all the evidence discussed herein, it is highly plausible that lifestyle modifications through a program that incorporates stepwise increment of physical training or exercise, patient education and advice, dietary modifications, and psychosocial stress management would have a significant impact on patients' thrombogenic profile and hence may beneficially influence the overall cardiovascular risk.
However, so far, to our knowledge, no study has been reported on the overall effects of such a comprehensive cardiac rehabilitation program on changes of the various variables of hemostasis, fibrinolysis, and platelet reactivity. Previous studies have mainly focused on the effects of short-term or regular physical activities on fibrinolytic responses in post-MI or post–coronary artery bypass grafting patients who participated in cardiac rehabilitation exercise programs (Table 7).
In keeping with epidemiologic data, patients with CHD have higher basal levels of PAI-1 and tPA antigen, suggesting impaired fibrinolytic activity compared with healthy subjects. Although healthy subjects tended to have a marked fibrinolytic response to exercise, patients with CHD have a lower increase in the fibrinolytic potential as evidenced by changes in tPA activity and PAI-1 levels after regular physical training.292 Although the increase of fibrinolytic capacity may be counterbalanced by an increase in blood coagulability and platelet activity during short-term exercise,91 lower plasma fibrinogen level has in fact been found in both post-MI and post–coronary bypass grafting patients who engage in regular aerobic exercise during cardiac rehabilitation.82,83
The hemostatic system is assuming an increasingly prominent role in the pathogenesis and progression of atherovascular diseases. Human lifestyle or physical activities have diverse effects on coagulation, fibrinolysis, and platelet reactivity. There have been abundant studies of the effects of exercise, weight loss, dietary lipids (especially n-3 PUFA), smoking, alcohol, and psychosocial stress on the 3 main systems of thrombogenesis. The data from intervention and randomized clinical trials are largely fragmented, rarely complete, and inconsistent, mainly due to the differences in study design and the inherent complexity of subjects' confounders and the lack of standardization of the various analytical methods used in the assessment of coagulation, fibrinolysis, and platelet function. The in vivo significance of examining one portion of the complex overall system is unclear. How much could one correlate the in vitro or ex vivo findings to the true in vivo biological activities in many of the human biological systems is largely unknown. Nevertheless, these data have provided us with important preliminary explanations for the relative contribution of the various thrombogenic markers in relation to lifestyle habits to clinical outcomes reported in epidemiologic studies. Available evidence from these studies support lifestyles that adopt strategies to lose weight, stop cigarette smoking, engage in regular moderate exercise and relaxation, and regularly consume light-to-moderate alcohol and fatty fish should significantly lower coagulability, promote fibrinolysis, and reduce platelet reactivity. The overall effects ought to translate into an improved cardiovascular or other beneficial clinical outcome in healthy individuals, those with cardiovascular risk factors, or those with established CHD. It follows that a cardiac rehabilitation program that incorporates a stepwise increment of physical training or exercise, patient education and advice, dietary and personal habit modifications, and psychosocial stress management would have a significant impact on patients' hemostatic profiles and hence beneficially influence the overall cardiovascular risk.
Corresponding author: Gregory Y. H. Lip, MD, FRCP, Haemostasis, Thrombosis and Vascular Biology Unit, University Department of Medicine, City Hospital, Dudley Road, Birmingham B18 7QH, England (e-mail: email@example.com).
Accepted for publication December 17, 2002.
This study was funded by a National Health Service Research and Development Health Technology assessment project grant (Dr Lee).
We acknowledge the support of the City Hospital Research and Development Programme for the Haemostasis, Thrombosis, and Vascular Biology Unit, Birmingham.