Antiatherothrombotic Properties of Statins:  Implications for Cardiovascular Event Reduction FREE

CLINICAL TRIALS of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors or statin therapy demonstrate an improvement in cardiovascular end points and coronary stenosis that is incompletely explained by the baseline or treated low-density lipoprotein (LDL) cholesterol level.^1- 5 The beneficial effects of statins on clinical events may involve nonlipid mechanisms that modify endothelial function, inflammatory responses, plaque stability, and thrombus formation (Table 1, Figure 1). This article discusses several nonlipid mechanisms that may contribute to the cardiovascular event reduction observed in clinical trials of statin therapy.

The LDL cholesterol levels serve as the focus of cholesterol treatment guidelines that were established to prevent coronary heart disease (CHD) in patients with hypercholesterolemia who are initially free of CHD and for those with established CHD.^6- 9 Treatment guidelines target LDL cholesterol levels on the basis of epidemiological associations rather than clinical trial outcomes and, thus, promulgate the concept that the magnitude of LDL cholesterol reduction equates directly with a proportionate reduction in atherosclerotic burden and clinical cardiovascular events. The LDL cholesterol level is an important target of cardiovascular prevention; however, elevated LDL cholesterol levels identify less than one half of individuals who will die from CHD.^10- 11 The LDL cholesterol concentrations had a sensitivity of 47% in predicting 10-year CHD death rates in the Lipid Research Clinics Prevalence Study.¹² Similar data are available from the Finnish cohort (n=444) of the Seven Countries Study for which only a modest correlation between serum cholesterol and 30-year CHD mortality was observed (r=0.42).¹¹ The revised National Cholesterol Education Program (NCEP II) guidelines⁷ that stratify risk by LDL cholesterol levels and conventional risk factors are no better in predicting risk.¹² The importance of LDL cholesterol levels on the prediction of CHD can be improved by concomitant measurements of high-density lipoprotein cholesterol levels,¹³ fibrinogen,^14- 15 plasma viscosity,¹⁶ and C-reactive protein.¹⁵

Clinical trials with statin therapy are accompanied by comparable changes in LDL cholesterol levels, but varying reductions in cardiovascular events^{3- 5,17- 30} (Table 2). The relationship between baseline and treated LDL cholesterol levels and cardiovascular end points has been evaluated in several clinical trials.^3- 5 The Regression Growth Evaluation Statin Study (REGRESS) demonstrated that the effect of pravastatin on change in mean coronary artery segment diameter, minimum obstruction diameter, and clinical events was not influenced by baseline LDL cholesterol level.³ Similarly, the Scandinavian Simvastatin Survival Study reported that major coronary events were reduced by a similar amount regardless of the baseline LDL cholesterol level.⁴ The West of Scotland Coronary Prevention Study (WOSCOPS) evaluated the relationship between on-treatment LDL cholesterol levels or total cholesterol change and CHD risk⁵ using the Framingham CHD-risk model.³¹ The CHD event rate in pravastatin-treated patients was not related to the magnitude of LDL cholesterol level lowering when the LDL cholesterol level reduction ranged from 19% to 54%.⁵ The Framingham model accurately predicted the CHD event risk rate in the placebo group but underestimated the CHD risk reduction in the pravastatin therapy group by 35%. These analyses suggest that nonlipid mechanisms^32- 33 may contribute to the cardiovascular event reduction in WOSCOPS²¹ and explain the early clinical benefit in this trial and several others.^{3,17- 18,20- 21,34}

Coronary plaque rupture and erosions are recognized precipitants of thrombosis in acute myocardial infarction, unstable angina pectoris, and sudden death.^35- 40 The vast majority of acute myocardial infarctions arise from atherosclerotic lesions that are minimal to moderate in severity as quantified by arteriography.^38,41 The rapid progression of subclinical coronary stenosis has led to an appreciation of the vulnerable plaque.^35- 36

The characteristics of the typical vulnerable plaque include increased numbers of inflammatory cells, such as macrophages and T lymphocytes at the shoulder region of the plaque, a large lipid pool, few smooth muscle cells and collagen fibers, and a thin fibrous cap^{37- 39,42- 44} (Figure 1). The content of macrophages and T lymphocytes within the plaque is an important determinant of plaque disruption. The macrophages release proteolytic enzymes that weaken a thin overlying fibrous cap and accelerate collagen degradation. T lymphocytes release interferon gamma (IFN-γ) at sites of human plaque disruption.⁴³ Interferon gamma inhibits smooth muscle cells from expressing interstitial collagen genes and provides a molecular basis for impaired collagen synthesis and inhibition of smooth muscle cell proliferation.^45- 47 In addition to impaired synthesis of structural proteins, catabolism of the extracellular matrix by metalloproteinases (interstitial collagenase, gelatinases, stromelysins) can weaken the fibrous cap.³⁷ These matrix-degrading proteinases are expressed by macrophage-derived foam cells and smooth muscle cells that are exposed to inflammatory cytokines (interleukin 1 or tumor necrosis factor). Plaque erosion is characterized by endothelial cell loss, sparsely distributed inflammatory cells, clusters of smooth muscle cells, and proteoglycans at the luminal surface.⁴⁰

Experimental animal models demonstrate that statin therapy can stabilize atherosclerotic plaques.^48- 49 These compositional changes include a reduction in extracellular lipid deposits and the area of macrophages in the intima and media, increase in collagen area and the ratio of collagen area to the area of extracellular lipid deposits, increase in smooth muscle cells, and less calcification and neovascularization in the intima. Lipid-lowering therapies may reduce cardiovascular risk not only through altering the arterial wall (endothelial dysfunction, atherogenesis, plaque stability), but also through their thrombogenic effects and effects on blood flow properties(Table 3).

Elevated LDL cholesterol levels are a major predisposing factor to atherosclerosis. Oxidized LDL impairs endothelial-dependent vasodilation, induces apoptosis of human endothelial cells via activation of a CPP32-like protease, a member of the interleukin 1β–converting enzyme-like protease family,⁵⁰ and generates an inflammatory response. Oxidized LDL modifies the functional response of vascular smooth muscle cells to angiotensin II stimulation.⁵¹

Oxidized LDL inhibits nitric oxide synthase activity of platelets,^52- 53 which promotes thrombus formation through an enhancement of fibrinogen binding to platelets.⁵⁴ Oxidized LDL binds to platelet activation factor, which is an intracellular proinflammatory regulator.⁵⁵ Modified LDL promotes tissue factor expression by monocytes, but LDL inhibits activation of the extrinsic coagulation pathway through binding to the tissue factor pathway inhibitor (TEPI).^56- 57 The LDL levels correlate with vitamin K–dependent coagulation factors (and inhibitors) and fibrinogen levels⁵⁸; however, the significance of these relationships is unresolved.⁵⁹ In subjects with hypobetalipoproteinemia (LDL cholesterol level, <1.81 mmol/L [69.99 mg/dL]), levels of fibrinogen and fibrinolytic markers (plasminogen activator inhibitor 1 [PAI-1] antigen), tissue plasminogen activator inhibitor 1 antigen) are low and this association further supports the relationship between LDL and these hemostatic factors.⁶⁰ In addition, LDL contributes to atherothrombogenesis through an increase in plasma and blood viscosity as shown in in vitro experiments and supported by epidemiological studies.⁵⁹

Endothelial-mediated vasodilatation is impaired in hypercholesterolemia and atherosclerosis.^61- 62 In coronary arteries of patients with atherosclerosis, cholesterol lowering with pravastatin and lovastatin improves endothelial function as evidenced by limiting acetylcholine-induced vasoconstriction (Table 3). ^63- 65 While this improvement in endothelial function was not seen with simvastatin,⁶⁶ the LDL cholesterol–lowering therapy with simvastatin improves peripheral nitric oxide–mediated vascular relaxation.^67- 70 The improved coronary blood flow and vasodilatory response with statin therapy alleviates transient ischemia in patients with stable angina pectoris^71- 72 and improves myocardial perfusion.^73- 74 In patients with mild hypertension, cardiovascular reactivity to angiotensin II and norepinephrine is diminished after 3 weeks of therapy with pravastatin.⁷⁵ Statin therapy can ameliorate endothelial dysfunction and contribute to the observed clinical benefits of those agents, and this improvement may be ascribed to LDL cholesterol lowering and antioxidant properties of statins.

An early step in atherogenesis involves monocyte adhesion to the endothelium and penetration into the subendothelial space. Oxidized LDL binds to the scavenger cell receptor on monocyte-derived macrophages and contributes to foam cell formation. Inflammatory cytokines secreted by macrophages and T lymphocytes can modify endothelial function, smooth muscle cell proliferation, collagen degradation, and thrombosis.^{46- 47,76- 78} Cholesterol level lowering in experimental models is accompanied by a reduction of inflammatory cells within atherosclerotic plaque.^{48- 49,79- 81} Subjects with hypercholesterolemia have increased adhesiveness of isolated monocytes to fixed endothelial cells in vitro, and this response is diminished with lovastatin and simvastatin.⁸¹ Hypercholesterolemic rats treated with fluvastatin have significantly attenuated leukocyte-adherence responses to platelet activation factor and leukotriene B₄.⁸¹

In patients with cardiac transplants, pravastatin may suppress the inflammatory response and inhibit natural killer cell activity in cyclosporin-treated patients.⁸² Although transplant vasculopathy is a distinct pathological entity from atherosclerotic vascular disease, the same inflammatory mediators may determine plaque vulnerability.

The relative content of cholesterol esters in plaque is an important factor influencing plaque stability. The pools of lipid-laden macrophage foam cells are nondistensible and do not absorb transmitted energy. Circumferential shear stress is concentrated on the fibrous cap that separates blood from the thrombogenic lipid core,^83- 84 and plaque disruption exposes the underlying plaque components to blood components that initiate thrombogenesis.^85- 86

Statins inhibit cholesterol ester accumulation in monocyte-derived macrophages either by reducing the availability of free cholesterol toward the enzyme acyl-coenzyme A cholesterol acyltransferase by trapping it in phospholipid-containing pools, or by inhibiting LDL endocytosis related to reduced synthesis of mevalonate or mevalonate by-products required for cholesterol esterification.⁸⁷ Kempen et al⁸⁸ reported a dose-dependent inhibition of cholesterol accumulation in macrophages that was greater with lovastatin and simvastatin than with pravastatin (Table 3). Lowering blood LDL cholesterol levels may facilitate plaque stability either through a reduction in size^89- 90 or by an alteration of the physiochemical properties of lipid cores.^91- 92 Hydrolysis of liquid cholesterol esters to solid cholesterol crystals can yield firmer plaques.

Another factor obscuring the relationship between LDL cholesterol levels and clinical events is the distribution of LDL subspecies within the plaque and the susceptibility of select LDL particles to oxidative modification. Small, dense LDL particles are more atherogenic than larger, buoyant particles, in part, because of enhanced oxidative susceptibility⁹³ and reduced total antioxidant defense.⁹⁴ The small LDL particle diameter is an independent predictor of myocardial infarction,^95- 96 and a predominance of dense LDL particles (density, >1.0378 g/mL) predicted coronary arteriographic benefit in the Stanford Coronary Risk Intervention Project.⁹⁷

In the Kuopio Atherosclerosis Prevention Study (KAPS),²⁶ 3 years of pravastatin therapy prolonged the lag time of LDL lipoproteins (a measure of oxidation resistance), increased plasma and LDL vitamin E levels, and improved overall LDL antioxidant capacity (Table 3). ⁹⁸ Lovastatin and simvastatin have been shown to inhibit LDL oxidation and uptake by macrophages in studies of shorter duration.^99- 105 Simvastatin treatment for 6 months maintains total antioxidant capacity of LDL particles; however, measurements of plasma antioxidants (ubiquinone, dolichol, α-tocopherol, β-carotene, and lycopene) were either reduced^103- 104 or unchanged.¹⁰⁵ Tissue concentrations of ubiquinone may be reduced with certain statins as reported in animal studies.^106- 107 In contrast, plasma ubiquinone levels fall with simvastatin therapy, whereas the concentration of ubiquinone in skeletal muscle biopsy specimens remains unchanged.¹⁰⁵ Overall, these data suggest that statin therapy does not alter tissue antioxidant balance but increases total antioxidant capacity of plasma.

Smooth muscle cells foster plaque stabilization^39,108 through synthesis of macromolecules that strengthen the fibrous cap³⁷ and absorption of energy transmitted to the vessel wall, which reduces circumferential shear stress along the endothelial surface.^83- 84 Mechanical stress serves as a stimulus for smooth muscle cell synthesis.¹⁰⁹ The smooth muscle cell also is involved in the normal healing process.^39,108 After a plaque ulcerates, the normal reparative process requires proliferation of vascular smooth muscle cells. Vascular smooth muscle cells regulate synthesis of interstitial collagens that are stimulated by transforming growth factor β and platelet-derived growth factor and inhibited by IFN-γ.⁴⁷

The influence of statins on smooth muscle cell proliferation has been evaluated in cell cultures of human femoral and rat myocytes and carotid arteries of rabbits.^110- 114 Balloon-mediated injury of rabbit carotid arteries caused intimal proliferation that was not inhibited in placebo- and pravastatin-treated rabbits, whereas this process was inhibited by treatment with all other statins.^110,112 In cell culture experiments, most statins (atorvastatin, cerivastatin, fluvastatin, simvastatin) except pravastatin inhibit smooth muscle cell proliferation and migration induced by the platelet-derived growth factor and fibrinogen.^111- 112 The permissive action of pravastatin on smooth muscle cell proliferation may be an advantage for the reparative process that follows plaque ulceration.¹⁰⁸ The concentration of statin needed to inhibit smooth muscle cell proliferation in vivo is comparable to plasma levels observed in humans administered conventional doses of these agents.¹¹⁴

Whether plaque disruption leads to an acute ischemic event depends, in part, on the propensity for thrombus to form on the damaged vessel wall. The thrombogenic response may be influenced by the thrombogenicity of the vessel wall components,⁸⁶ interaction of blood components with the lipid pool or smooth muscle cell and proteoglycan complexes, local blood flow properties, and circulating hemostatic factors. The prothrombotic factors that have been evaluated with the statins include tissue factor expression, platelet aggregation, fibrinogen, plasma viscosity, and fibrinolytic factors (Table 3).

Tissue factor and corresponding messenger RNA have been localized in macrophages of human atherosclerotic plaque.⁸⁵ Tissue factor serves as a cofactor for plasma factor VII and cellular receptor for factor VIIa and, thus, plays a central role as the initiator of the extrinsic coagulation pathway.¹¹⁵ Lipophilic statins (fluvastatin, simvastatin) suppress tissue factor expression by cultured human macrophages through inhibition of a geranylgeranylated protein involved in tissue factor biosynthesis.¹¹⁶ This effect on tissue factor was not observed with pravastatin.

The extrinsic coagulation activation pathway is counterbalanced by a serine protease inhibitor known as TEPI. TEPI binds to factor Xa, and this complex inhibits tissue factor–mediated coagulation through binding to the tissue factor—factor VIIa complex.^56- 57 Circulating TEPI is transported by dense subspecies of LDL, lipoprotein(a) (Lp[a]), and high-density lipoprotein.¹¹⁷ TEPI activity is increased in subjects with heterozygous familial hypercholesterolemia,^118- 119 and types IIa (by 70%, P<.001) and IIb (by 36%, P<.001) hyperlipoproteinemias.¹²⁰ Cholesterol level lowering with simvastatin reduces LDL and TEPI without a significant change in factor VIIc.¹¹⁸

Platelets from patients with elevated LDL levels are more sensitive to aggregating agents than are platelets of normocholesterolemic subjects.¹²¹ The LDL causes intracellular acidification through inhibition of Na⁺/H⁺ antiport in human platelets, which mobilizes intracellular calcium in the resting state and after stimulation with agonists.¹²² The adenosine diphosphate–induced fibrinogen binding to platelets is increased in a dose-dependent manner by LDL (0.5-2.0 g of protein per liter).¹²³ Simvastatin reduces platelet aggregation and thromboxane production after 4 to 24 weeks of therapy, whereas lipid lowering was observed by 2 weeks of treatment.^124- 125 Lovastatin therapy has been accompanied by both an increase and a decrease in platelet count and adenosine diphosphate–induced platelet aggregation.^126- 127 Pravastatin normalizes platelet-dependent thrombin generation in hypercholesterolemic subjects, but this effect is unaccompanied by a change in prostaglandin production.¹²⁸ Pravastatin also has been shown to reduce cytosolic calcium and platelet aggregation.¹²⁹

Statins may reduce platelet aggregation by changing the cholesterol content of platelet membranes, which alters membrane fluidity.¹²⁹ In an experimental model that simulates primary hemostasis, hypercholesterolemic patients treated with aspirin (325 mg/d) deposit more platelet aggregates on damaged porcine aorta than normocholesterolemic subjects, and this response was reduced by pravastatin therapy.¹³⁰ In hypercholesterolemic subjects treated with aspirin (325 mg/d) and randomized to either pravastatin or simvastatin at equivalent LDL cholesterol-lowering doses, pravastatin inhibited platelet-thrombus formation on an injured artery. This was not observed in simvastatin-treated patients.¹³¹ These latter studies suggest a differential effect of statins on tissue-dependent platelet aggregation.

Fibrinogen levels and plasma viscosity may be used to stratify cardiovascular risk in hypercholesterolemic patients with and without established CHD.^14- 16 Elevated plasma viscosity may contribute to atherothrombosis through impaired microcirculatory flow, shear stress damage at the blood-endothelial interface, facilitation of plasma protein interaction with the endothelium in poststenotic recirculation zones, and increased propensity for thrombosis.¹³²

Several studies examining the influence of statins on fibrinogen levels in hypercholesterolemic patients have shown mixed results.¹³³ Lovastatin has been accompanied by 19% to 24% increases in fibrinogen levels in two 6-month studies of 26 and 49 patients, respectively,^126,134 and by 5% in a 12-month study of 260 patients.¹³⁵ Two other small, short-term studies of 15 and 35 patients reported no change in fibrinogen concentration.^136- 137 In contrast, Mayer et al¹²⁷ found a significant 10% reduction in fibrinogen levels in 20 hypercholesterolemic patients treated with lovastatin after 16 weeks of therapy, with further reductions throughout the 12-month interval of the study. This study was limited by fibrinogen measurements made at monthly intervals without adjustment for multiple observations and selection of a second fibrinogen value at baseline that was 4% higher than the initial value.¹³³

Lovastatin has been accompanied by an increase in whole blood viscosity,¹²⁶ while plasma viscosity is either reduced¹³⁶ or does not change.^126,137 Pravastatin has been accompanied by a 7% to 9% lowering of fibrinogen levels in 3 studies of 16 to 24 patients treated from 10 to 24 weeks.^138- 140 Three studies reported no change in fibrinogen concentration with pravastatin.^26,141- 142 However, KAPS reported a nonsignificant 7% elevation in fibrinogen in patients treated with either pravastatin and placebo for 3 years, but the increase in fibrinogen was 20% lower in the pravastatin group than in the placebo group.²⁶ Unlike the Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial,¹⁴³ KAPS did not age adjust fibrinogen levels.²⁶ In addition, 2 pravastatin studies reported a reduction in whole blood viscosity at shear rates ranging from 75 to 375 seconds⁻¹, corrected blood viscosity at 112.5 seconds⁻¹ and 225 seconds⁻¹, and plasma viscosity.^138,140 Simvastatin does not change fibrinogen as shown in 4 studies of 12 to 111 subjects followed up for 10 weeks to 2 years.^{140,142,144- 145} Additionally, blood, plasma and serum viscosity measurements were unchanged with simvastatin therapy.¹⁴⁴ Two comparative studies that randomized subjects to either pravastatin or simvastatin reported a favorable or neutral change in fibrinogen with pravastatin and a neutral change with simvastatin.^140,142 No published studies have evaluated the effect of fluvastatin on fibrinogen levels. Atorvastatin (80 mg daily) was accompanied by a 46% increase in fibrinogen levels in 22 heterozygous familial hypercholesterolemic subjects treated for 6 weeks.¹⁴⁶ The fibrinogen elevation was greater in subjects randomized to atorvastatin administered as a twice-daily dose compared to a single dose (P<.05). In 789 subjects, a lower dose of atorvastatin (10 mg daily) was accompanied by a smaller (4%) elevation in plasma fibrinogen levels.¹³⁵ Other investigators evaluated multiple dosages of atorvastatin (10-80 mg daily) on fibrinogen levels in 95 subjects and reported an increase in fibrinogen levels that ranged from 19% to 24%.¹⁴⁷

The variable effect of statins on fibrinogen may result from different actions on fibrinogen-regulating cytokines,⁷⁸ study populations with genetic variation at the fibrinogen gene locus,¹⁴⁸ or measurement variability.¹⁴⁹ Another factor to consider is age adjustment of fibrinogen levels in long-term studies.¹⁴³

Fibrinolytic mechanisms evaluated with statins include measurements of Lp(a) and PAI-1, the principal inhibitor of the fibrinolytic system. Impaired fibrinolysis as measured by an elevation in PAI-1 is predictive of ischemic heart disease in free-living subjects¹⁵⁰ and survivors of myocardial infarction.^150- 152 Pravastatin reduces PAI-1 antigen levels by 26% to 56%,^139,141 and this appears to facilitate fibrinolysis. Lovastatin has been shown to both decrease PAI-1 by 22%¹³⁴ and elevate PAI-1 by 34% in 260 subjects.¹³⁵ Atorvastatin therapy increased PAI-1 by 36% after 12 months.¹³⁵ Simvastatin increased PAI-1 by 18% in 111 patients treated for 2 years.¹⁴⁵ Fluvastatin has a neutral effect on PAI-1 antigen.¹⁵³

Lipoprotein(a) interferes with fibrinolysis by competing with plasminogen binding to plasminogen receptors, fibrinogen, and fibrin.¹⁵⁴ The net effect is impaired plasminogen activation and plasmin generation at the thrombus surface.¹⁵⁵ The Lp(a) levels can increase by as much as 34% with statin therapy^156- 160 and potentially impair clot lysis. Since the importance of Lp(a) on cardiovascular risk diminishes with LDL cholesterol–lowering therapy,¹⁵⁵ the importance of this modest evaluation in Lp(a) levels on overall cardiovascular risk is unclear. The opposing effects of statins on the different mediators of fibrinolysis may offset each other.

Statins influence critical pathways that regulate plaque stability and thrombosis, and these properties extend beyond LDL lowering. The spectrum of direct antiatherogenic properties of statins includes maintenance of endothelial function, anti-inflammatory actions, and a permissive action on smooth muscle cell proliferation that allows for synthesis of extracellular matrix proteins involved in the reparative response. Following plaque disruption, statins influence thrombosis through variable inhibitory actions on platelet deposition and aggregation, coagulation factors, rheology, and fibrinolysis. The qualitative differences among statins may influence the effectiveness in the prevention of cardiovascular events and atherosclerosis disease progression.

Knowledge about the intricacies of atherosclerosis and thrombosis continues to expand rapidly, and these mechanisms should be used to support the evidence from randomized clinical trials. Since the nonlipid properties of statins differ despite comparable LDL cholesterol level lowering, the net clinical efficacy of these agents requires validation by randomized clinical trials.