Atherosclerotic vascular disease is the most common cause of morbidity and mortality in developed countries, and the world-wide importance of acute vascular syndromes is increasing. Acute events are usually triggered by the development of plaque disruption and subsequent thrombus formation. Histological studies have established specific structural features common among unstable plaques. The plaque has to bear remarkably increased mechanical stress at particular regions, and weakening of the extracellular matrix at these sites leads to fibrous cap rupture. The biologic factors that cause weakening of the plaque at these high stress locations are now emerging. Understanding the interplay of plaque architecture, mechanical properties and matrix biology is critical in the future development of therapies to stabilize lesions.
Time for primary review 14 days.
Coronary atherosclerosis constitutes the fifth leading cause of global disease burden and the leading cause in developed societies . We have witnessed a decrease in mortality rate from coronary disease in the past few decades due to therapeutic advances and changes in lifestyle in the general population . However, it is expected that in spite of these advances, over the next two decades ischemic heart disease will become the leading cause of morbidity and mortality in the world, surging past infectious diseases .
Coronary atherosclerotic disease is predominantly an asymptomatic process. Growth of coronary plaque presents clinically as angina when coronary flow is decreased. This clinical syndrome can be stable for many years; and a myriad of medical, interventional, and surgical options are available for its treatment. On the other hand, sudden rupture of a plaque triggers the development of an acute coronary syndrome such as unstable angina, myocardial infarction or sudden death. The sequence of events involved in this pathophysiological process leading from plaque rupture, to intraluminal thrombus formation and finally to myocardial ischemia with or without necrosis, now appears clear [3–6].
A variety of mechanisms have been proposed as the initial mechanical cause of plaque rupture, including: shear stress injury , transient collapse of the stenotic lesion , mechanical shear stress and rupture of the vasa vasorum . Exploration of these mechanisms has enhanced our understanding of the pathophysiology of unstable coronary disease. However, clinical identification of the unstable plaque has been elusive. Cardiologists have relied on coronary angiography as the “gold standard” for defining the degree of severity of coronary disease. Angiography has been a reliable clinical tool for management of chronic ischemia or interventions once instability has developed, but angiography cannot detect the unstable lesion. Little et al. showed that angiographic severity of a lesion poorly predicts the subsequent location of a myocardial infarction . Similarly, Ambrose et al. demonstrated that myocardial infarction frequently developed from ruptured plaques affecting coronary lesions judged to be not severe by angiography .
There are several reasons angiography cannot identify the unstable lesion. Comparative studies between angiographic and postmortem histological findings of lesions found that angiography underestimated the severity of these lesions [13, 14]. In addition, coronary disease is frequently a diffuse process with no clearly normal segments of the coronary tree, thus limiting further the reliability of angiography in estimating percentage of stenosis. Second, angiography cannot see plaque structure, an important component of plaque stability. Third, angiography cannot detect biologic processes causing weakening of the fibrous cap of the atheroma.
Studies on the biological characteristics of the plaque have contributed to the understanding of successful therapies such as the success of lipid-lowering agents. Clinical studies have demonstrated an improvement in the incidence of cardiovascular disease with these drugs but no major changes in lesion severity as shown by angiography . This suggests that the benefits arising from the use of these medications are related to modifications of the structural and biological characteristics of the plaque rather than to absolute regression of the disease process.
Pathological studies by Davies et al. and other groups have established the association between plaque rupture and thrombus formation. These findings have been documented at necropsy in cases of acute myocardial infarction and sudden death but also in individuals dying of noncardiac causes [3–6, 16]. Two primary mechanisms can trigger the development of thrombus formation: frank rupture of the atheromatous fibrous cap and superficial denudation of the endothelium. In addition, morphological and microscopic characteristics of the unstable atheromatous plaque include: (a) thin fibrous cap; (b) eccentric, large lipid core; and (c) infiltrating macrophages . The focus of this review will be to describe our current understanding of the mechanisms involved in plaque rupture, particularly regarding mechanical factors of plaque stability.
2 Mechanical forces and plaque rupture
To understand the effects that mechanical forces have within the atheroma it is necessary to review briefly concepts of material properties and the physical laws that define the interaction between opposing forces in a cylinder (such as a blood vessel) . Stress is the force acting on a surface divided by the size of the surface; thus, it is defined in units of force per area. Stress can be applied in any direction within the vessel wall and can be radial, circumferential or longitudinal. Normal stress is defined as the perpendicular force applied to the vessel wall (such as by blood pressure); parallel forces to the endothelial surface result in shear stress. For a given radial pressure within a cylinder there is a compensatory circumferential tension. Laplace's law estimates the relationship between these two opposing forces in a cylinder with a thin walled cylinder:
where σ is the circumferential wall stress, P the radial wall stress, r is the radius of the vessel and h the thickness of the wall (Fig. 1). This formula helps explain the tendency to rupture of thin walled vessels like aortic aneurysms. Although in the complex geometries of atherosclerotic vessels many other factors must be considered, the concept remains similar; in the atheromatous plaque, very high circumferential stress can develop in thin fibrous caps, causing the mechanical failure of the plaque.
Strain may be defined as the amount of deformation by a structure as a consequence of an external applied force, normalized to its length. Strain reflects changes in the length (Δl) of the material and is expressed as fraction or percentage of the initial length (l).
The stiffness of a material is characterized by the ratio between the applied stress and the observed strain. This is often expressed by the elastic or Young's modulus (E):
Linear elastic behavior is seen in those structures whose elastic modulus remains constant over a range of stresses; when this modulus remains constant in all directions, we call the material isotropic. An elastic material will have an immediate stress/strain response after a force has been applied to it. Biological structures do not behave in such way; the time response to a given force until a new equilibrium strain has been reached is called viscoelasticity.
Vascular structures are characterized by a wide variability of their material properties: (1) they are distinctly anisotropic, being stiffer in their axial and circumferential directions than radial; (2) their elastic properties are nonlinear, becoming stiffer with higher degrees of strain; (3) in the case of plaques, stiffness also increases with increases in frequency of applied stress. Because of the plaque's complex structure, researchers have used a modern structural computational technique called finite element analysis for the distribution of mechanical forces in the atherosclerotic plaque. This technique allows the evaluation of complex structures by breaking it down into smaller sections in order to determine areas of maximum stress.
Cross sectional area of an artery demonstrating Laplace's relation. The intraarterial pressure (P) is balanced by a circumferential stress (σ). Laplace's law applies to tubes with thin walls (h). In the case of more complex structures like atheromatous arteries, computer methods such as finite element analysis have been used to measure plaque stresses.
Using the finite element approach, Richardson et al. studied the effects of mechanical stress within the fibrous cap in individuals who died of acute coronary thrombosis . Evaluating different geometries of plaques that caused lethal coronary thrombosis, they observed increased levels of stress concentrating at the edges of the fibrous cap near the border with the normal intima. It was also observed that in those cases with very small lipid pools (less than 15% of the vessel circumference), the point of maximum stress was located over the center of the plaque. This concentration of mechanical stress in the fibrous cap regions is possibly due to the inability of the soft lipid core to bear the large mechanical stresses that develop during elevation of blood pressure or repetitive dynamic stress caused by pulsatile blood pressure. Supporting these results, Cheng et al. evaluated the distribution and magnitude of circumferential stress in plaque rupture by studying lesions of patients dying of acute coronary events and comparing them with nonruptured lesions of individuals who died from other causes . Using a finite element model, it was observed that mechanical stress was higher in rupture regions than nonruptured ones. However, it was also found that the location of plaque rupture was not always the area of greatest stress in an individual lesion. This observation suggests that local variations in plaque strength may determine which high stress region actually ruptures.
It is crucial to emphasize the importance of fibrous cap thickness in plaque stability. Born and Richardson postulated that plaque rupture was dependent on the structural and physical properties of the fibrous cap . Loree at al. determined that decreasing the thickness of the fibrous cap greatly enhanced the peak circumferential stress, whereas increasing stenosis severity actually decreased stress . These results may, in part, explain the discordance between angiographic findings and clinical events, and emphasize the limitation of current methods of diagnosis in the identification of unstable plaques.
The repetitive deformations caused by the cardiac cycle may play an important role in lesion stability. McCord and Ku evaluated the effect of mechanical tension in arteries by observing morphological and mechanical changes caused by cyclic fatigue on diseased arteries in the areas of maximum stress . Microscopic analysis demonstrated structural damage in the fatigued specimens as compared to normal tissues. This implicates the effect of increasing pulse pressure in plaque rupture in the same way that bending a paper clip repetitively eventually breaks it. In addition, weakening of tissue by compression of the fibrous cap such as by collapse of the artery due to the pressure gradient caused by the stenosis or by arterial spasm remains a possible mechanism. It is important to note that nonbiologic materials may undergo fatigue when subjected to deformation, while biologic materials may compensate under some circumstances to increase tissue strength. Quite possibly, failure to compensate adequately by the remaining cells in the atheroma is a major factor in plaque stability (Fig. 2).
A cascade of events leads to plaque rupture. Accumulation of lipid in the lesion leads to dramatically increased stress on the fibrous cap of the lesion. In addition, lipid acccumulation promotes inflammation through chemotactic factors and upregulation of adhesion molecules. The combination of increased mechanical stress on the fibrous cap and weakening of the fibrous cap extracellular matrix leads to plaque rupture.
3 Biological factors affecting plaque stability
The variable and seemingly unpredictable nature of plaque rupture despite apparent lesion similarity has focused attention on the fibrous cap as an active biological structure. In the absence of inflammation or injury, many connective tissues have little evidence of active extracellular matrix synthesis and degradation; instead, they appear to be dormant. When tissues are injured and the repair process begins, both synthetic and degradative processes greatly accelerate. The predominant component of all fibrous cap is connective tissue matrix proteins, particularly collagen types I and III, but also elastin and proteoglycans. Lendon et al., investigating the mechanical and rupture behavior of an atherosclerotic plaque, observed a correlation between collagen content, increased fracture stress, and decreased vascular extensibility . Diminished collagen synthesis will weaken the fibrous cap strength, and therefore produce a greater tendency to rupture at lower levels of circumferential stress.
Complex interactions among biological factors regulate the synthesis and breakdown of extracellular matrix. Platelet derived growth factor has been found to act as a potent stimulus for smooth muscle cell proliferation [26–28]. In vitro studies have also determined that macrophage derived cytokines like transforming growth factor-β act as potent stimulants of collagen synthesis by vascular smooth muscle cells (VSMCs) . The combined effect of these factors and later organization of the collagen strands by the smooth muscle cells via β1 integrins, influence the strength of the fibrous atheromatous plaque .
Accumulation of T-lymphocytes and macrophages within the atherosclerotic plaque may decrease its stability. This hypothesis is supported by observations of increased densities of T-lymphocytes and macrophage-derived foam cells in the shoulder area of atheromas [31–34]. These inflammatory cells secrete cytokines and proteolytic enzymes into the extracellular matrix, resulting in both decreased synthesis and an enhanced destruction of extracellular matrix. Amento et al. demonstrated suppression of collagen gene expression in smooth muscle cells by interferon-γ (IFN-γ) . The presence of immunostainable IFN-γ in the vicinity of T-lymphocytes in the shoulder of human atheromas is consistent with the role of this cytokine in plaque friability . In addition to the inhibitory effects on collagen synthesis, IFN-γ depresses the proliferation of smooth muscle cells in response to other growth factors and promotes apoptosis in these cells . As a consequence of this, depletion in the number of smooth muscle cells in the plaque results in compromise of the repair process.
Activation of macrophages in the atheromatous plaque leads to the secretion of a variety of proteolytic enzymes capable of degrading the extracellular matrix and consequently weakening the atheroma. Three major families of enzymes participate in extracellular matrix degradation: serine proteases (urokinase and plasmin), cysteine proteases such as cathepsins, which are largely intracellular enzymes, and matrix metalloproteinases (MMPs). Degradation of extracellular matrix requires a two-step process: (1) enzymatic disruption of the extracellular milieu; and (2) clearance via endocytosis of the degraded components and lysosomal degradation. MMPs are an important family of enzymes in the initial stages of degradation and have been the focus of increasing research .
Metalloproteinases are secreted as a proenzyme/inhibitor complex and activated in the extracellular space at neutral pH . Their enzymatic activity is dependent on two Zn2+ molecules for activity and a Ca2+ molecule for stabilization. The proenzyme includes an N-terminal domain which is cleaved when the enzyme is activated, in addition to catalytic and C-terminal domains. The latter two domains are involved in receptor identification and specificity of substrate recognition as well as being the site where the endogenous inhibitor called tissue inhibitor of metalloproteinases binds on a 1:1 stoichiometric basis. MMP activity is substrate-specific, although substantial substrate overlap exists among the various members of this family of enzymes.
Activation of MMPs can occur through several mechanisms that usually involve cleavage of approximately 10 kilodalton from the N-terminus of the proenzyme. In the atheroma, activation can be mediated by reactive oxygen species or plasmin. Lijnen et al., using mice with targeted inactivation of plasminogen, have found that active MMP-9 is not consistently detectable in these mice compared to wild type mice . In addition, activation of some MMPs—in particular MMP-2—can occur through membrane-type metalloproteinases (MT-MMPs), which are bound to the membrane by a C-terminus transmembrane domain. At least three types of MT-MMPs are expressed in vascular smooth muscle, and these proteins may participate in regulation of matrix weakening .
Overexpression of MMPs has been demonstrated in atheromatous lesions both in animals and humans. Nikkari et al. demonstrated increased expression of collagenase (MMP-1) by several cell types in human carotid atherosclerotic plaques and its correlation with histopathological evidence of plaque instability . A potential sequence of events leading to elevated levels of MMPs in the plaque begins with ingestion of lipids by infiltrating macrophages into the atheromatous lesion. Galis et al. noted a difference among macrophages in lesions of cholesterol-fed rabbits and macrophages obtained from broncho–alveolar lavage from the same animals . In addition, production of cytokines such as IFN-γ, interleukin-1 and tumor necrosis factor-α (TNF-α) by activated macrophages and T-lymphocytes will stimulate the production of MMPs by smooth muscles cells .
Another component of atheroma architecture in plaque stability is the lipid core. Its mechanical properties affect the distribution of tensile forces, making stresses concentrate in the fibrous cap, particularly at its edges. Clinical trials using lipid-lowering agents have shown an improvement in morbidity and mortality from cardiovascular diseases . A potential mechanism explaining this effect might be changes in the composition of the lipid core by decreasing cholesterol esters and increasing the proportion of insoluble cholesterol monohydrate. These changes would potentially decrease the degree of stress affecting the fibrous cap .
4 Mechanical forces and extracellular matrix stability
Several investigators have noted an association between overexpression of MMP-1 and the shoulder regions of the atheroma, which are also the regions of increased mechanical stress. We studied the localization of MMP-1 quantitatively by comparing the distribution of stress in human coronary lesions with the expression of MMP-1. Expression of MMP-1 increases severalfold in regions of increased mechanical stress . Since this is where plaque rupture frequently occurs, it is attractive to speculate that the combination of excess matrix degradation and excess mechanical stress at these locations leads to failure of the fibrous cap. This raises the intriguing question that the increased stress in these regions actually promotes weaker extracellular matrix.
One possible source of MMPs in the fibrous cap shoulders is the VSMCs, although the unstable fibrous cap has a reduced number of these cells. Mechanical signals are powerful regulators of cellular functions, and it has been known for over 20 years that mechanical deformation regulates extracellular matrix synthesis by VSMCs (, and reviewed in Ref. ). Multiple transduction pathways may participate in converting mechanical signals into biochemical signals, including stretch-activated ion channels, paracrine growth factors, G proteins, MAP kinases, integrins, tyrosine kinases, and phospholipid metabolism. No single gene or signaling pathway seems to be responsible for all mechanotransduction; in different experimental conditions, different pathways may predominate [48–50]. We tested the hypothesis that mechanical strains increase MMP-1 secreted by human VSMC. Surprisingly, strains do not induce MMP-1 in VSMCs, but strain is a potent inhibitor of platelet-derived growth factor or TNF-α induced synthesis of MMP-1. Thus, direct induction of MMP-1 by increased mechanical stress on VSMC in the high stress region of the atheroma is not supported by in vitro data .
An alternative possibility is that MMP-1 is increased directly or through paracrine mechanisms involving the macrophage. Currently, little is known about how macrophages respond to mechanical stimuli. Martin et al. hypothesized that changes in morphology are a common feature of macrophage activation and that stretch-activated ion channels may play a role . They identified an outwardly rectifying potassium channel that is inactive at rest but activated by adhesion of cells or stretch of the membrane. This demonstrates that macrophages, like smooth muscle cells , have stretch-activated channels which can transduce mechanical signals. Mastsumoto et al. studied morphology of the monocyte-like cell line U937 and rat peritoneal macrophages with a uniaxial stretch device. They found evidence suggesting that cyclic stretch inhibits the differentiation to vacuolized cells and facilitates the differentiation to spindle cells . Thus, it is possible that macrophages are particularly active in this region of increased mechanical stress.
By understanding factors that regulate extracellular matrix integrity in the atheroma, we may potentially pharmacologically modify lesion stability in the future. One theoretical disadvantage of this approach is that by downregulating matrix degradation, total volume of the atheroma could increase. This would lead to an increase in chronic ischemia; one could argue that chronic ischemia is more easily managed than an acute vascular catastrophe. However, recent animal evidence suggests that both reduction in lesion mass and improved stability may be possible. When Aikawa et al. studied rabbit aortic atherosclerosis in rabbits with high and low cholesterol diets, they found that lesion mass could be reduced at the same time that inflammation decreased and dense collagen in the atheroma increased in the low cholesterol rabbits .
In the past decade, a wealth of insight into the nature of the unstable atheroma has emerged. The success of cholesterol-lowering therapy indicates that plaque rupture is not the inevitable natural history of the atheroma. We now recognize that plaque stability is determined by many factors, including extracellular matrix degradation, mechanical forces caused by lipid deposits, and inflammation. Currently we do not know the cadence of instability, although recent epidemiological studies suggest that inflammation precedes clinical events by many years . This potentially provides us with a prolonged window of opportunity to intervene before lesions cause acute events. As the global importance of acute ischemia is increasing, we must learn more about how to identify unstable lesions and reverse the processes that lead to plaque rupture. Because plaque rupture is a catastrophic mechanical event, understanding the interactions between mechanical forces and extracellular matrix integrity will be essential.
Born GVR, Richardson PD. Mechanical properties of human atherosclerosis. In: Glagov S, Newman W, Schaffer S, editors. Pathobiology of the human atherosclerotic plaques. New York: Springer–Verlag, 1990:413–424.
Libby P, Lee RT. Role of activated macrophages and T-lymphocytes in rupture of coronary plaques. In: Braumwald E, editor. Heart disease: a textbook of cardiovascular medicine (Suppl 2). Philadelphia: Saunders, 1995:1–9.
(1994) Site of intimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an inflammatory process irrespective of the dominant plaque morphology. Circulation 89:36–44.
(1998) Function of the plasminogen/plasmin and matrix metalloproteinase systems after vascular injury in mice with targeted inactivation of fibrinolytic system genes. Arterioscler Thromb Vasc Biol. 18:1035–1045.
(1997) Expression of three membrane-type matrix metalloproteinases in rat vascular smooth muscle cells and characterization of MT3-MMPs with and without transmembrane domain. J Biol Chem. 272:9749–9754.