Understanding of the pathogenesis of congestive heart failure (CHF) has improved remarkably in recent years. However, despite better knowledge and novel pharmaceutical strategies, this disease is still one of the most brutal killers in the Western world. The pathophysiology of CHF is complex, and much of our comprehension revolves strictly around the neurohormonal and mechanical mechanisms involved. It has been suggested that CHF is associated with altered hemostasis, but whether a prothrombotic state and platelet activation contributes to the pathogenesis and progression of the disease is still not well known. Current knowledge is updated with regard to platelet activation biomarkers in CHF patients, and the potential role of current therapeutic regiments in preventing these abnormalities.
Introduction
In terms of incidence, prevalence, morbidity, mortality, and economic costs, heart failure represents a major and growing public health concern. Despite significant progress in the prevention and treatment of cardiovascular disease in the past two decades, the incidence and prevalence of CHF have been increasing steadily in recent years.1-3 In the US alone, there are currently four to five million patients suffering from CHF, with approximately 400,000 new cases diagnosed, resulting in nearly one million hospitalizations, and responsible for over 250,000 deaths per year.4-6 The number of hospitalizations of patients with a principal diagnosis of CHF has increased over the past two decades, and is expected to double as a result of progressive aging of the Western population over the next 40 years.2,7,8 At present, the economic impact of treating CHF is exceeding US$11 billion each year. Considering all the above-mentioned facts, it is pertinent that we continue to strive to improve our understanding of the mechanisms, etiologies, and treatments of this prevalent disease.
Thromboembolism is a critical and relatively common complication of CHF. Indeed, a variety of factors associated with CHF predispose to thrombosis. For instance, the existence of CHF increases the risk of thromboembolism, whether or not concomitant atrial fibrillation (AF) is present.9 Small observational studies have suggested that pulmonary thromboembolism and stroke are common in patients with dilated cardiomyopathy.10 The annual risk of stroke is increased to 4% in patients with severe CHF.11
Several studies have shown that patients with CHF have increased plasma concentrations of beta-thromboglobulin (╬▓-TG),12-14 a marker of platelet activation. Increased plasma concentrations of fibrinopeptide A and thrombin-antithrombin III (TAT) complexes, both measures of thrombin activity, have also been demonstrated.12 This article summarizes the data on platelet activation in heart failure, the effects of commonly prescribed medications for CHF on platelet function, and the potential role of the platelet ADP-receptor antagonists.
CHF and Platelets
In the late 1970s, investigators noted that patients with CHF had significantly more circulating platelet aggregates when compared with healthy controls.15 Subsequently, several markers of platelet activity have been found to be increased in CHF patients including beta thromboglobulin, platelet factor four, and cellular adhesion molecules such as P-selectin, platelet/endothelial cell adhesion molecule, and osteonectin. The interaction between endothelial cells, leukocytes, and platelets involving various adhesion proteins may be of particular importance due to possible immunologic and inflammatory responses in the pathogenesis and natural progression of CHF.16,17 Although the origin of platelet activation in CHF remains to be established, and is probably multifactorial, increased platelet activity might be related to elevation of cytosolic-free calcium concentration.18 Platelets in CHF may also be affected by enhanced sympathoadrenal activation and catecholamine release.19 It seems possible that reduced kidney and liver blood flow results in decreased clearance of platelet-activating substances, which may attribute to the higher incidence of clinically manifested thrombotic events in CHF patients.
Beta-thromboglobulin is an alpha-granule constituent, which is a very sensitive but not very specific marker of platelet activation. The increased level of beta-thromboglobulin and platelet factor four was observed in patients with ischemic and idiopathic cardiomyopathy.20 P-selectin (CD62-P), originally described as PADGEM/GMP 140, is an alpha granule membrane protein that is also expressed on the platelet surface upon activation.21 Elevated soluble P-selectin levels often coincide with other abnormal markers of platelet function, deteriorated platelet morphology, and enhanced platelet aggregation in CHF patients.23,24 Modulation of P-selectin, von Willebrand factor (vWF), and other hemorheological indices may contribute to a hypercoagulable state in CHF, especially in female patients and in those with more severe New York Heart Association (NYHA) class.25
Statistically, CHF is more common in men than women, although the effect of gender on prognosis is not clear. In general, the studies evaluating CHF have been conducted in predominantly male populations. The Studies of Left Ventricular Dysfunction (SOLVD) patient registry, of which 26% were female patients, found that women had a significantly higher risk of morbidity and mortality, with more hospitalizations (33% versus 25%) and deaths (22% versus 17%).26 In this study, female patients with CHF had greater abnormalities of hemorheology and vWF than male patients, which could contribute in part to the higher thrombotic event rates in female CHF patients. Indeed, by multiple regression analyses, female gender was an independent predictor for elevated plasma viscosity, fibrinogen, and vWF levels. CHF is known to be associated with impaired endothelium-dependent vasodilatation and diminished release of endothelium-derived nitric oxide in response to stimuli, which contributes to the peripheral vasoconstriction, which is well documented in heart failure.27 Consequently, the elevated baseline vWF levels reflect the presence of pre-existing endothelial dysfunction in CHF. As a procoagulant product of the endothelium, vWF may further enhance the prothrombotic state through its effects on platelet aggregation and adhesion to the endothelium.28 Indeed, elevated vWF levels are associated with an increased risk of reinfarction and mortality in patients after acute myocardial infarction (MI) or left ventricular (LV) aneurysm.29, 30
Effects of CHF Therapy on Platelets
Incidence of clinical manifestation with thromboemboli in patients with CHF is estimated at more than 2% per year.31 The role of antithrombotic agents to prevent these events has been studied for over 50 years.32 However, there have been very limited studies evaluating the use of antiplatelet agents to reduce acute coronary events and death in patients with heart failure. The drugs, mostly commonly prescribed for CHF, already provide modest antiplatelet effect (with the exception of digoxin, which substantially increases the concentration of intracellular Ca) and, as a result, enhance platelet aggregation, and up-regulate platelet serotonin receptors.33,34
Diuretics such as hydrochlorothiazide, furosemide, spironolactone and indapamide, augment the synthesis of prostaglandins D2, E2, and I2, probably through facilitated reorientation of endoperoxide biotransformation. With the exception of hydrochlorothiazide, these drugs can also suppress thromboxane A2 production. Lipoxygenase formation of hydroxyeicosatetraenoic acid can be enhanced by spironolactone and indapamide.35
Indapamide inhibits the second wave of platelet aggregation induced by adenosine diphosphate and collagen in platelet-rich plasma by 50%. In the model of isolated platelets, indapamide inhibits aggregation induced by low doses of thrombin by 70%, and diminishes the thrombin-induced release of serotonin from dense granules by up to 80%. Hydrochlorothiazide at the same concentrations has no effect on platelet aggregation, and the inhibitory effect on the secretion was inconsistent and never exceeded 30%. By contrast, when aggregation was induced by arachidonic acid, indapamide had no effect either on aggregation or on thromboxane formation, indicating that it was not acting via arachidonic acid passway. Instead, indapamide inhibits platelet responses by blocking calcium mobilization.36
The antiplatelet properties of beta-blockers are less potent and probably much less clinically meaningful. In vitro, propranolol and carvedilol reduce platelet aggregation induced by epinephrine and adenosine diphosphate (ADP).37 But in the clinical setting, beta-blockers failed to affect very important biomarkers of platelet activity (P-selectin and vWF) in patients with CHF.25
Angiotensin converting enzyme (ACE) inhibitors have become the 'cornerstoneÔÇÖ of therapy for CHF. It seems reasonable to expect that ACE-inhibitors might be beneficial, in part due to their antiplatelet activity. Indeed, angiotensin II receptors are also present at the platelet surface,38,39 although their role is not yet identified. Angiotensin II induces dose-dependent elevations of intraplatelet-free calcium, which was in turn dependent on the extracellular calcium levels.40 Angiotensin II per se potentates agonist-induced platelet aggregation41 and causes mild activation of the coagulation cascade with increases in plasma levels of thrombin-antithrombin complex and prothrombin fragment F1+2, established markers of thrombin generation. Other reports suggest that angiotensin II exhibit mild platelet-activating effects, and also enhance coagulation in vivo.42 Angiotensin II increases platelet aggregation through activation of the G-protein-linked pathway present in platelets. On the other hand, in vitro platelet aggregation induced by thrombin or ADP can modulate the angiotensin II release.43 The decrease in plasma angiotensin II levels can lead to an increase in platelet angiotensin II receptor expression, or can lead to angiotensin II production in platelets through turning on a feedback mechanism. Most likely, platelets have alternative enzymatic pathways leading to angiotensin II production which could not be inhibited by captopril.48 The prothrombotic effect of angiontensin II could be dependent on the release of PAI-1 from endothelial cells.44, 45
Therapy with ACE inhibitors is associated with the decreased levels of beta-thromboglobulin, at least in hypertensive patients.46 Captopril and fosinopril decreased TxB2 concentration 27.5% to 67.5%, with no significant changes in adenosine diphosphate (ADP)-, epinephrine- or thrombin-stimulated platelet aggregation when compared with baseline levels, or between different ACE inhibitor.47 Antiplatelet activity is likely a common feature of all ACE inhibitors. At least in a high concentration enalapril reduced platelet aggregation48 and fosinopril-reduced platelet aggregation induced by TRAP by 50% in an animal study,49 and in male patients with hypertension.50 Treatment with captopril in patients suffering from recent MI, reduced the expression of glycoprotein IIb/IIIa receptors by 30%.51
Briefly increased bradykinin levels with the consequent enhanced synthesis of vasodilatory prostaglandins, appear to mediate a significant benefit of ACE inhibitor therapy in patients with CHF. In contrast, aspirin inhibits cyclo-oxygenase, and thereby suppresses prostaglandin production. Thus, these counteracting effects on prostaglandins may result in antagonism between ACE inhibitor and aspirin therapy in heart failure patients.52 Since the results of the Co-operative North Scandinavian Enalapril Survival Study II (CONSENSUS-II) trial were published in 1997,53 the question of interaction between aspirin and ACE inhibitors was extensively discussed. Moreover, the most recent meta-analysis of 31,622 patents enrolled in the Global Utilization of Streptokinase and t-PA for Occluded Coronary Arteries (GUSTO-I) trial and 2,619 subjects enrolled in the Evaluation in PTCA to Improve Long-Term Outcome with Abciximab GPIIb/IIIa Blockade (EPILOG) trial has shown that the mortality rate was increased among the patients who received both aspirin and ACE inhibitors. Thus, the unadjusted mortality was greater among patients treated with both ACE inhibitors and aspirin than among patients treated with aspirin alone (3.3% versus 1.6%, P <0.001 for GUSTO-I, and 3.7% versus 1.2%, P <0.001 for EPILOG).54 In contrast, the analysis of mortality data of 11,575 patients with coronary artery disease (CAD) screened for the Bezafibrate Infarction Prevention trial has demonstrated the significant death reduction among patients treated with ACE inhibitor-aspirin combination than ACE inhibitor alone (19% versus 27%).55 This very important clinical question is far from being completely elucidated and further study of this issue is warranted.
The potential role of angiotensin II receptor blocker in the treatment of CHF is extensively investigated. Val-HeFT trial has shown moderate but highly significant effect of valsartan on the primary end-point of all-cause mortality and morbidity (32.1% on placebo, 28.8% on valsartan, p=0.009) in patients with CHF.56 The current trial, VALsartan In Acute myocardial iNfarcTion (VALIANT) is going to determine the effect of valsartan on mortality and morbidity in post-MI patients with heart failure or impaired LV function.57
Interestingly, some of the novel angiotensin receptor blockers also exhibit anti-platelet properties presumambly via modulating of eucosanoid pathway. Thus, losartan competitively antagonists with thromboxane A2 receptor, inhibits thromboxane A2-induced platelet aggregation in rats,58 and decreases platelet aggregation in patients with hypertension, but has no effect on plasma concentrations of plasminogen activator inhibitor-1 and von Willebrand factor in hypertensive subjects.59 Irbesartran has a similar effect on platelet function in both animals and humans.60
In summary, it has been suggested that CHF is associated with altered hemostasis, but whether this prothrombotic state contributes to the pathogenesis and progression of the disease is still unclear. ACE inhibitors, diuretics, and beta-blockers provide substantial, but modest anti-thrombotic properties, while glycosides may activate platelets. The justification for more aggressive antiplatelet regimens in patients with CHF is far from obvious. Aspirin is an established medication for prevention and treatment of CHD, but it could decelerate the beneficial effect of ACE inhibitors. Clopidogrel is a relatively safe and established medication that reduces the incidence of stroke, myocardial ischemia, or vascular death. It is currently the drug of choice in the prophylaxis of subacute stent thrombosis. Future clinical investigations must consider better strategies to avoid activation of blood coagulation and platelets. However, the role of pure antiplatelet strategies, especially the aggressive ones, should be considered with caution. Ôûá