This review will reconsider the current paradigm for

Lessons From Sudden Coronary Death A Comprehensive Morphological Classification Scheme for Atherosclerotic Lesions Renu Virmani, Frank D. Kolodgie, Allen P. Burke, Andrew Farb, Stephen M. Schwartz This review will reconsider the current paradigm for understanding the critical, final steps in the progression of atherosclerotic lesions. That scheme, largely an outgrowth of observations of autopsy tissues by Davies and colleagues, 1,2 asserts that the cause of death in atherosclerotic coronary artery disease is rupture of an advanced atherosclerotic lesion. Although this assumption may be partially true, recent autopsy studies suggest that it is incomplete. See p 1177 To reconsider this paradigm, we reexamined the morphological classification scheme for lesions proposed by the American Heart Association (AHA). 3,4 This scheme is difficult to use for 2 reasons. First, it uses a very long list of roman numerals modified by letter codes that are difficult to remember. Second, it implies an orderly, linear pattern of lesion progression. This tends to be ambiguous, because it is not clear whether there is a single sequence of events during the progression of all lesions. We have therefore tried to devise a simpler classification scheme that is consistent with the AHA categories but is easier to use, able to deal with a wide array of morphological variations, and not overly burdened by mechanistic implications. The Current Paradigm The current paradigm is based on the belief that type IV lesions, or atheromas, described by the AHA are stable because the fatty, necrotic core is contained by a smooth muscle cell rich fibrous cap. Virchow s analysis 5 in 1858 pointed out that historically, the term atheroma refers to a dermal cyst ( Grützbalg ), a fatty mass encapsulated within a cap. Extending Virchow s argument, the fibrous cap over the lipid mass of an atherosclerotic plaque is analogous to the capsule containing an abscess, and like an abscess, the plaque can be ruptured. Rupture of the fibrous cap exposes thrombogenic material, initiating platelet aggregation and coagulation in the infiltrating and overlying blood. These thrombotic changes result from activation of the clotting cascade by tissue factor, and further propagation of the thrombosis results from the interaction of platelets with the active thrombogenic matrix. 6 Platelet activation and thrombin formation combined with the evulsion of thrombogenic plaque contents into the lumen then result in sudden occlusion. 7 This widely held concept of atherosclerotic death is based on morphological data from autopsies as well as clinical angiographic studies, in which the presence of surface irregularities has been interpreted as plaque rupture. 8 10 Previous pathological studies of sudden coronary death have demonstrated evidence of plaque rupture associated with thrombosis in 73% of cases. 2 Of the remaining cases, 8% consist of plaque fissure with intraplaque fibrin deposition and hemorrhage, while 19% show no evidence of thrombi. 2 Consequently, recent reviews of atherosclerosis have uniformly accepted plaque rupture as the critical event leading to coronary artery death. 6 Limitations The major limitation of the present paradigm is the lack of a direct, experimental test in a prospective model in humans or animals. For example, lesions in current animal models rarely progress beyond the stage of atheroma (ie, a well-developed fibrous cap overlying a necrotic core). More often, lesions consist of masses of lipid-laden intimal macrophages without a well-developed fibrous cap. Lesions with this histology are rarely clinically significant except in examples of severe hyperlipemia, in which the lumen can become occluded by the sheer plaque burden. 11 As we will review, this situation is quite atypical of human disease. The general failure to observe clinically significant lesions in animals may be a simple function of the relatively short duration of most experimental studies, as suggested by 1 long-term study in a unique strain of hyperlipemic swine. 12 Analysis of human arteries also has its inadequacies. Given the limited ability of current clinical imaging methods to visualize the vessel wall, as opposed to the lumen, 13 we are highly dependent on autopsy material. Those autopsy studies are not entirely consistent with the current paradigm. These inconsistencies, as discussed below, could mean that the Received September 7, 1999; revision accepted February 14, 2000. From the Department of Cardiovascular Pathology (R.V., F.D.K., A.P.B., A.F.), Armed Forces Institute of Pathology, Washington, DC, and the Department of Pathology (S.M.S.), Vascular Biology, University of Washington, Seattle. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or reflecting views of the Departments of the Army, Air Force, or Defense. Reprint requests to Renu Virmani, MD, Department of Cardiovascular Pathology, Armed Forces Institute of Pathology, 6825 16th St NW, Washington, DC 20306-6000. E-mail virmani@afip.osd.mil (Arterioscler Thromb Vasc Biol. 2000;20:1262-1275.) 2000 American Heart Association, Inc. Arterioscler Thromb Vasc Biol. is available at http://www.atvbaha.org 1262

Virmani et al Lesion Classification and Sudden Coronary Death 1263 TABLE 1. Distribution of Culprit Plaques by Sex and Age in 241 Cases of Sudden Coronary Death Acute Thrombi Organized Thrombi No Thrombi: Fibrocalcific Plaque Rupture Erosion Calcified Nodule Totals Men 50 years 45 (46) 17 (17) 2 (2) 15 (15) 20 (20) 99 50 years 19 (23) 8 (10) 3 (4) 27 (33) 26 (31) 83 Women 50 years 1 (3) 14 (42) 0 5 (15) 13 (40) 33 50 years 9 (35) 6 (23) 1 (4) 5 (19) 5 (19) 26 Totals 74 (31) 45 (19) 6 (2) 52 (22)* 64 (26) 241 Values correspond to the number of cases; those in parentheses are percentages. *Organized thrombi with healed myocardial infarct (HMI) 46/52, 89%. No thrombi (stable plaque) with HMI 32/64 (50%); thus, 32/241, or 13%, of sudden death cases have stable plaque with HMI. autopsy population is biased (ie, representing mostly young, sudden coronary death victims and excluding nonfatal clinical events), that the paradigm is incorrect, or, as we will suggest, that the paradigm is incomplete. If the latter is the case, we need to consider other ways, besides rupture, in which an atherosclerotic plaque can produce sudden coronary death. It is very important to realize that the presence of a plaque rupture does not imply a causal association with the thrombus that occluded the lumen. There is ample evidence that nonfatal lesions can contain areas of rupture. For example, Arbustini and collaborators 14 found a 10% incidence of plaque ruptures in lesions of people who died of noncardiovascular causes. These findings suggest that advanced plaques may undergo many nonfatal ruptures without causing death. Moreover, the existence of nonruptured but fatal lesions suggests that the equation of rupture with death is overly simplistic. 14 Exceptions to the Current Paradigm Exceptions to the current paradigm have arisen because of 2 recent articles. 15,16 First, in a series of 20 patients with sudden cardiac death, van der Wal et al 15 found plaque rupture in only 60% of lesions with thrombi; the remaining 40% showed only superficial erosion. The term superficial erosion was defined as a thrombus confined to the most luminal portion of a fibrous cap in the absence of fissure or rupture after serial sectioning. Approximately half of the eroded lesions showed a fibrous cap heavily infiltrated by macrophages and T lymphocytes overlying an atheromatous core. The remaining cases were associated with fibrocellular caps without a necrotic core containing mostly smooth muscle cells and a paucity of macrophages and T lymphocytes. The second set of data evolved from examinations in the laboratory performed by 1 of us (R.V.). We studied the coronary vasculature of 200 cases of sudden coronary death (Table 1). 16 22 The definition of sudden coronary death was based on an unexpected death witnessed within 6 hours of the onset of symptoms or death of a person known to be in stable condition 24 hours antemortem. 17 The histological criteria for sudden coronary death included luminal thrombi in 1 or more arteries or in at least 1 major coronary artery with 75% cross-sectional area luminal narrowing. As shown in Table 1, only one third of the lesions in our studies could be described as plaque rupture, and remarkably, 35% of lesions with thrombi failed to show rupture. 16 Many of these lesions, unlike those described by van der Wal et al, 15 did not show significant inflammation. In our series, macrophages and other inflammatory cells were the exception; instead, abundant, often arborized smooth muscle cells embedded in a proteoglycan-rich matrix characterized the eroded tissue. Altogether, these data suggest that coronary thrombi can arise without rupture. Furthermore, because the plaque subjacent to the thrombus in erosion typically does not show inflammatory cells, the data also contradict the prevailing notion that inflammation is a necessary event leading to thrombotic occlusion of coronary arteries. 6,14 Thrombotic occlusion in the absence of rupture also raises a critical question about the central role of rupture in the current AHA paradigm. If thrombi occur without rupture, how can one know whether the presence of rupture in a thrombosed vessel is not an incidental phenomenon? Issues Excluded From the AHA Classification Scheme Table 2 shows the current classification scheme as proposed by the AHA. 3,4 Not addressed in this scheme in direct relation to sudden coronary death are 2 matters that, though potentially important, complicate efforts of classifying lesions. These issues are death without occlusion and the relationship of luminal narrowing to other features of plaque progression. Despite the diagnosis of sudden coronary death, direct implication of death resulting from sheer plaque burden in the absence of thrombi is vexing. In our series, 26% of cases classified as sudden coronary death failed to show direct evidence of thrombi but were diagnosed as undergoing sudden cardiac arrest with severe coronary atherosclerosis (Table 1). In these cases, lesions of AHA category types Va, Vb, and Vc all plaques showing significant stenosis without thrombi were the presumed cause of death because a complete autopsy, including a toxicology screen, failed to indicate other causes. It is conceivable that death associated with severe coronary narrowing resulted from noncardiac causes or a lethal arrhythmia triggered by myocardial ischemia. Alternatively, healed myocardial infarcts were found in 50% of these stable lesions without thrombi at the time of

1264 Arterioscler Thromb Vasc Biol. May 2000 TABLE 2. Current AHA Classification Terms for Atherosclerotic Lesions in Histological Classification Other Terms for the Same Lesions Often Based on Appearance to the Unaided Eye Type I lesion Initial lesion Type II lesion IIa Progression-prone type II lesion Fatty dot or streak Early lesion IIb Progression-resistant type II lesion Type III lesion Intermediate lesion (preatheroma) Type IV lesion Atheroma Atheromatous plaque, fibrolipid plaque, Va Fibroatheroma (type V lesion) fibrousplaque, plaque Vb Calcific lesion (type VII lesion) Calcified plaque Advanced lesions, raised lesions Vc Fibrotic lesion (type VIII) Fibrous plaque Type VI lesion Lesion with surface defect and/or hematoma/hemorrhage and/or thrombotic deposit Complicated lesion, complicated plaque autopsy, indicating that at some point during the life of the patient, a thrombus was most likely present. Thus, the culprit lesion in these patients initially had a thrombus that underwent spontaneous lysis, recanalized, or was nonocclusive. The second topic is the relationship of lesion morphology to lumen area reduction. Some pathologists are of the opinion that severe stenosis is a prerequisite for plaque rupture and luminal thrombosis. 23 However, angiographic studies before and after myocardial infarction frequently show that preexisting lesions at the sites of complete occlusion are not usually accompanied by hemodynamically significant stenosis (ie, 50% diameter reduction). 24 From experience in 1 of our laboratories (R.V.), cross-sectional luminal narrowing of 75% is not a prerequisite for luminal thrombosis, either acute or healed, or for the development of intraplaque hemorrhage. We have shown that in sudden coronary death patients who died of luminal thrombosis, at least 50% of the TABLE 3. Modified AHA Classification Based on Morphological Description Nonatherosclerotic intimal lesions Intimal thickening Intimal xanthoma, or fatty streak Progressive atherosclerotic lesions Pathological intimal thickening Description The normal accumulation of smooth muscle cells (SMCs) in the intima in the absence of lipid or macrophage foam cells Luminal accumulation of foam cells without a necrotic core or fibrous cap. Based on animal and human data, such lesions usually regress. SMCs in a proteoglycan-rich matrix with areas of extracellular lipid accumulation without necrosis thrombi occurred at lesion sites with 75% cross-sectional area stenosis by plaque (corresponding to 50% diameter reduction). 16 Repeated ruptures, however, may be responsible for plaque progression. In a recent report (R.V.), healed plaque ruptures had an overall final stenosis of 80% cross-sectional area luminal narrowing, although the percent stenosis before the final rupture (old lumen) showed only 66% cross-sectional area narrowing ( 50% diameter reduction). 22 We will return to this issue later as part of the discussion of the use of our classification scheme to identify critical mechanisms of sudden coronary death. Modifications of the AHA Classification Our proposed changes to the AHA classification are shown in Table 3 and Figure 1. This modification focuses primarily on the AHA classification of type IV, V, and VI lesions. These Thrombosis Erosion Luminal thrombosis; plaque same as above Thrombus mostly mural and infrequently occlusive Fibrous cap atheroma Well-formed necrotic core with an overlying fibrous cap Absent Erosion Thin fibrous cap atheroma Plaque rupture Luminal thrombosis; plaque same as above; no communication of thrombus with necrotic core A thin fibrous cap infiltrated by macrophages and lymphocytes with rare SMCs and an underlying necrotic core Fibroatheroma with cap disruption; luminal thrombus communicates with the underlying necrotic core Absent Absent Absent Thrombus mostly mural and infrequently occlusive Absent; may contain intraplaque hemorrhage/fibrin Thrombus usually occlusive Calcified nodule Eruptive nodular calcification with underlying fibrocalcific plaque Thrombus usually nonocclusive Fibrocalcific plaque Collagen-rich plaque with significant stenosis usually contains large areas of calcification with few inflammatory cells; a necrotic core may be present. Absent

Virmani et al Lesion Classification and Sudden Coronary Death 1265 Intimal Xanthomata (Figures 1 and 2) We propose the term intimal xanthoma instead of the type I, fatty streak, or initial lesion in the AHA s scheme. Xanthoma is a general pathological term that describes focal accumulations of fat-laden macrophages. In humans, most of these intimal xanthomata regress, since the distribution of lesions in the third decade of life and beyond is very different from the fatty streaks seen in children. 25 We must emphasize that the presence of an intimal xanthoma is not a basis for categorizing lesions in current animal models as atherosclerotic. This is problematic, based on the fact that the distribution of lesions in animal models is very different from that in the adult human population 26 and the potential for these lesions to regress. 27,28 Figure 1. Our simplified scheme for classifying atherosclerotic lesions modified from the current AHA recommendations. The boxed areas represent the 7 categories of lesion. We have used dashed lines for 2 categories because there is controversy over the role that each of them plays in the initial phase of lesion formation, and both lesions can exist without progressing to a fibrous cap atheroma (ie, AHA type IV lesion). The processes leading to lesion progression are listed between categories. Lines (solid and dotted, the latter representing the leastestablished processes) depict current concepts of how 1 category may progress to another, with the thickness of the line representing the strength of the evidence that these events do occur. comprise the typical intermediate or advanced plaques recognized from the third decade of life onward with the potential to develop thrombi and sudden coronary death. 4 Our scheme is based on the 7 categories shown in Figure 1. These categories include intimal xanthoma, intimal thickening, pathological intimal thickening, fibrous cap atheroma, thin fibrous cap atheroma, calcified nodule, and fibrocalcific plaque. The key features defining these categories are the accretion of lipid in relationship to formation of the fibrous cap, changes over time in the lipid to form a necrotic core, thickening or thinning of the fibrous cap, and thrombosis. Remaining issues such as the culprit lesion associated with the thrombus and specific plaque features representing processes critical to changes in the lesion (eg, angiogenesis, intraplaque hemorrhage, inflammation, calcification, cell death, and proteolysis) are listed as descriptive terms. Intimal Thickening (Figures 1 and 2) While some human lesions may begin as intimal xanthomata, we, along with the authors of the AHA classification scheme, agree that most adult human lesions originate as preexisting intimal masses. Evidence for this tenet comes in part from studies by Kim et al, 29 who showed that atherosclerotic lesions produced in the coronary arteries of hypercholesterolemic swine arise almost exclusively from intimal cell masses. Moreover, the distribution of these normal, developmental intimal masses in children can be correlated with the distribution of characteristic lesions seen in adult humans. 25,30 The origin of fat accumulation subjacent to initially small, very focal, preexisting intimal masses may explain the following paradox. There is very little evidence of cell replication except in early lesions, yet the smooth muscle cells of adult lesion are usually clonal. 30,31 Very few replications over a long time could easily account for quite sizeable atherosclerotic lesions. Thus, the clonality of the lesions may provide a teleological clue, suggesting that the properties of these normal intimal structures may be relevant to the earliest events in the formation of human lesions. Tabas et al, 32 for example, have proposed that the extracellular matrix at these sites may contain enzymes capable of retaining lipids, an initial event in the formation of the necrotic core. Unfortunately, there are very few articles on the evolution of early intimal cell masses in humans, and none of these clarify their precise pathological mechanisms of development. Another reason for considering the role of the intima in giving rise to clinically significant lesions comes from our observation that the majority of erosions occur over areas of intimal thickening, with minimal or no evidence of a lipid core. Figure 1 suggests that erosion may occur in response to the existence of pathological intimal thickening. At present, the criteria for classifying this putative lesion, other than the existence of an overlying region of thrombosis and absence of an endothelium, 16 remain unclear. Fibrous Cap Atheromata (Figures 1 and 3) The AHA classification distinguishes between lesion types IV and V on the basis of the degree of fibrous cap formation, TABLE 4. Morphological Characteristics of Culprit and Rupture-Prone Plaques in Cases of Sudden Coronary Death Plaque Type Necrotic Core, % Cholesterol Clefts, % Macrophages, % Mean No. of Sections With Intraplaque Hemorrhage Rupture 34 17* 12 12 26 20 2.5 1.3 Thin fibrous cap 23 17 8 9 14 10 atheroma Erosion 14 14 2 5 10 12 0 Fibrocalcific 15 20 4 6 6 8 0.05 0.6 P *0.003 vs erosion, 0.01 vs fibrocalcific 0.002 vs erosion, 0.04 vs fibrocalcific 0.001 vs erosion and stable plaque, 0.03 vs thin fibrous cap atheroma 0.01 vs erosion and fibrocalcific plaque

1266 Arterioscler Thromb Vasc Biol. May 2000 the extent to which the lipid core becomes rich in cellular debris, and the development of complicating features. 4 Because there is no clear evidence of a specific sequence of events that relate the extent of changes in the lipid core to the development of a fibrous cap, we suggest the use of descriptive terms for plaque classification. We define a fibrous cap as a distinct layer of connective tissue completely covering the lipid core. The fibrous cap consists purely of smooth muscle cells in a collagenousproteoglycan matrix, with varying degrees of infiltration by macrophages and lymphocytes. Thus, a fibrous cap atheroma may have a thick or thin cap (see below) overlying a lipid-rich core. The lipid core is also part of our classification. As lesions progress, the core of necrotic debris surrounded by macrophages becomes increasingly consolidated into 1 or more masses comprising large amounts of extracellular lipid, cholesterol crystals, and necrotic debris. Thin Fibrous Cap Atheromata (Figures 1, 4, and 5) We have added the thin fibrous cap atheroma as a specific plaque type not recognized by the current AHA classification. 4 We have done this because in our experience, lesions with thin, fibrous caps are those that are most likely to rupture. 19 The AHA s discussion suggests this as well; however, their classification scheme also describes type IV lesions as showing fissuring and hemorrhage, including rupture, of the fibrous cap. 4 Figure 2. Preatherosclerotic coronary lesions. Intimal thickening and intimal xanthoma: Lesions are uniformly present in all populations, although intimal xanthomas are more prevalent with exposure to a Western diet. Intimal xanthomas are commonly produced in animal models; however, they usually do not develop into progressive atherosclerotic lesions. Both lesions occur soon after birth; the intimal xanthoma (otherwise known as a fatty streak) is known to regress. Intimal thickening consists mainly of smooth muscle cells in a proteoglycan-rich matrix, whereas intimal xanthomas primarily contain macrophage-derived foam cells, T lymphocytes, and varying degrees of smooth muscle cells. We define a thin, fibrous cap as 1 that is 65 m thick. This definition was derived from a morphometric series of 41 ruptured plaques, in which 95% of the caps measured 64 m thick; the mean SD plaque thickness was 23 19 m. 19 The thin, fibrous cap is distinguished from the earlier fibrous cap lesions by the loss of smooth muscle cells, extracellular matrix, and inflammatory infiltrate. The necrotic core underlying the thin, fibrous cap is usually large; hemorrhage and/or calcification is often present; and intraplaque vasa vasorum are abundant. 22,33 In our series of 200 sudden death cases, 60% of acute thrombi resulted from rupture of thin fibrous cap atheromata. In these patients, thin fibrous cap atheromas without rupture were found in 70% of cases. On the contrary, these plaques were less frequent (30%) in patients who died of fibrocalcific lesions, with or without a healed myocardial infarct, or plaque erosion. 16 Although the classification scheme in Figure 1 shows the thin fibrous cap atheroma as a separate category, this lesion does not necessarily develop rupture. Other relevant descriptive features of the fibrous cap lesions include the extent of inflammation in the cap, fissuring, calcification, intraplaque vasa vasorum, or intraplaque hemorrhage. Lesions With Thrombi (Figures 1 and 4) Rather than creating separate lesion types for rupture and thrombosis, the simplified scheme proposes to classify le-

Virmani et al Lesion Classification and Sudden Coronary Death 1267 sions with thrombi as being affected principally by 3 distinct processes: rupture, erosion, and, less frequently, the calcified nodule. These processes can occur in the setting of a fibrous cap atheroma or, in the case of erosion, pathological intimal thickening. As we will discuss below, a single lesion may contain morphological evidence of both rupture and erosion. Often, we see a fatal lesion having 1 area of the thrombus in communication with a necrotic core through a ruptured fibrous cap and another area overlying a smooth muscle cell rich plaque representing plaque erosion (Figure 6). Rupture Plaque rupture is defined by an area of fibrous cap disruption whereby the overlying thrombus is in continuity with the underlying necrotic core. Ruptured lesions typically have a large necrotic core and a disrupted fibrous cap infiltrated by macrophages and lymphocytes. The smooth muscle cell content within the fibrous cap at the rupture site may be quite sparse. Plaque ruptures are found in 60% of individuals dying suddenly with luminal thrombi and are the most frequent cause of death in young men ( 50 years) and older women ( 50 years; Table 1). Risk factors most predictive for this type of lesion are hypercholesterolemia, low serum HDL, and a high total cholesterol to HDL cholesterol ratio. 19 In women 50 years old, ruptured plaques compose the vast majority of atherosclerotic lesions associated with acute thrombi, and similar to men, there is an association with increased total cholesterol levels. 21 Erosion Plaque erosion is identified when serial sectioning of a thrombosed arterial segment fails to reveal fibrous cap rupture. Typically, the endothelium is absent at the erosion Figure 3. Pathological intimal thickening versus atheroma. Pathological intimal thickening is a poorly defined entity sometimes refereed to in the literature as an intermediate lesion. True necrosis is not apparent, and there is no evidence of cellular debris; some lipid may be present deep in the lesion, but it is dispersed. The fibrous cap overlying the areas of lipid is rich in smooth muscle cells and proteoglycans. Some scattered macrophages and lymphocytes may also be present, but these are usually sparse. The more definitive lesion, or fibrous cap atheroma, classically shows a true necrotic core (NC) containing cholesterol esters, free cholesterol, phospholipids, and triglycerides. The fibrous cap consists of smooth muscle cells in a proteoglycan-collagen matrix, with a variable number of macrophages and lymphocytes. The media underneath the plaque is often thin. EL indicates extracellular lipid. site. The exposed intima consists predominantly of smooth muscle and proteoglycans, and surprisingly, the eroded site contains minimal inflammation. 15,16 We have chosen to use the term erosion, despite its mechanistic implication, because we are unaware of evidence that such large areas of endothelium are ever absent in nonthrombosed vessels, even over advanced lesions. 34 37 Erosions constitute 40% of cases of thrombotic sudden coronary death. 16 Plaque erosions are more common in young women and men 50 years of age (Table 1) and are associated with smoking, especially in premenopausal women. Figure 1 refers to the intima underlying an area of erosion as pathological intimal thickening, because we assume that some as-yet-identified property leads to this event. It is important to note that this form of thrombotic occlusion does not require a necrotic core, but if present, is usually small. In contrast to plaque rupture, smooth muscle cells are abundant while the presence of macrophage and/or T lymphocytes is variable. 16 Confusion Between Rupture and Erosion as a Primary Event The existence of a spontaneous thrombotic occlusion without rupture complicates the assumption that an occlusive thrombus in rupture is dependent on the rupture. Ruptures in some cases could be incidental events. For example, Figure 6 (A and B) represents a low- and high-power view of a distal coronary section at the level of a rupture site. A more proximal section, however (C and D), is histologically indistinguishable from an eroded plaque. This case exemplifies the morphological diversity of some lesions, in which the origin of thrombus development resulting in the patient s demise is confusing.

1268 Arterioscler Thromb Vasc Biol. May 2000 Figure 4. Atherosclerotic lesions with luminal thrombi. Ruptured plaques are thin fibrous cap atheromas with luminal thrombi (Th). These lesions usually have an extensive necrotic core (NC) containing large numbers of cholesterol crystals and a thin fibrous cap ( 65 m) infiltrated by foamy macrophages and a paucity of T lymphocytes. The fibrous cap is thinnest at the site of rupture and consists of a few collagen bundles and rare smooth muscle cells. The luminal thrombus is in communication with the lipid-rich necrotic core. Erosions occur over lesions rich in smooth muscle cells and proteoglycans. Luminal thrombi overly areas lacking surface endothelium. The deep intima of the eroded plaque often shows extracellular lipid pools, but necrotic cores are uncommon; when present, the necrotic core does not communicate with the luminal thrombus. Inflammatory infiltrate is usually absent, but if present, is sparse and consists of macrophages and lymphocytes. A high-power view of the eroded surface indicated by a box is shown in Figure 6 for comparison with a similar eroded area of a thrombosed region adjacent to a ruptured plaque. Calcified nodules are plaques with luminal thrombi showing calcific nodules protruding into the lumen through a disrupted thin fibrous cap (FC). There is absence of an endothelium at the site of the thrombus, and inflammatory cells (macrophages, T lymphocytes) are absent. Calcified Nodule A second lesion, albeit an infrequent cause of thrombotic occlusion without rupture, is referred to as a calcified nodule. This term refers to a lesion with fibrous cap disruption and thrombi associated with eruptive, dense, calcific nodules. The origin of this lesion is not precisely known, but it appears to be associated with healed plaques. Interestingly, these lesions are found predominantly in the midright coronary artery, where coronary torsion stress is maximal. It is unclear whether the fibrous cap wears down from physical forces exerted by the nodules themselves, proteases from the surrounding cellular infiltrate, or both. Thin fibrous cap lesions with eruptive, calcified nodules should not to be confused with fibrocalcific lesions that are not associated with thrombi. The latter, as discussed below, appear to be the end result of fibrosis and calcification and are often associated with a narrowed lumen. Moreover, this entity does not appear to be a variant of rupture with calcification, since the nodules themselves appear in the lumen in the absence of an overt intimal tear. Histological Features of Thrombi (Figure 1) Fresh Occlusion Fresh occlusion is identified by a luminal thrombus containing platelet aggregates interspersed with inflammatory cells and a paucity of red blood cells. The thrombus, however, often propagates from its original site, becomes fibrin rich, and contains interspersed red blood cells and leukocytes. In a fresh thrombus, there is no evidence of invasion by endothelial cells and/or smooth muscle cells. Little is known of the mechanism(s) involved in thrombus propagation. Old Occlusion Old occlusions often show the lumen totally occluded by dense collagen and/or proteoglycan with interspersed capillaries, arterioles, smooth muscle cells, and inflammatory cells. These lesions may also demonstrate earlier phases of organizing thrombi containing fibrin, red blood cells, and granulation tissue, especially in the midportion of a long, occluded arterial segment. Lesions Not Necessarily Associated With Thrombi (Figures 1 and 5) Fibrocalcific Lesions Some plaques have thick, fibrous caps overlying extensive accumulations of calcium in the intima close to the media. 38 Because the lipid-laden necrotic core, if present, is usually small, we refer to this category of lesion as fibrocalcific rather than atheroma. Of course, as shown in Figure 1, it is possible that the fibrocalcific lesion is the end stage of a process of atheromatous plaque rupture and/or erosion with healing and calcification.

Virmani et al Lesion Classification and Sudden Coronary Death 1269 Figure 5. Variants of fibrous cap atheromas. Fibroatheromas with intraplaque hemorrhage are lesions with a necrotic core (NC) containing red blood cells and fibrin. The fibrous cap is mature, and deep within the intima are areas of calcification. Thin fibrous cap atheromas, often referred to in the literature as vulnerable plaques, are lesions with large necrotic cores containing numerous cholesterol clefts. The overlying fibrous cap (FC) is thin ( 65 mm) and heavily infiltrated by macrophages; smooth muscle cells are rare and vasa vasorum are present within the adventitia and plaque. Fibrocalcific plaques are referred to as stable lesions in the literature. These lesions are predominantly composed of dense collagen containing a few scattered smooth muscle cells and inflammatory cells; when present, the necrotic core is small. These lesions are often heavily calcified, and it is difficult to determine the previous hemorrhage and/or thrombosis that may be responsible for the plaque size. Intraplaque Hemorrhage (Figure 5) Constantinides 39 originally suggested that hemorrhage into a plaque occurs from cracks or fissures originating from the luminal surface. Davies 1 later defined plaque fissure as an eccentric, intraplaque hemorrhage with fibrin deposition within the necrotic core from an entry into the plaque from the lumen. The fissuring of the fibrous cap occurs at its thinnest portion, typically at the shoulder region, thereby allowing the entry of blood into the necrotic core. As Davies suggested, plaque fissures may represent precursors or subtypes of plaque rupture. 1 Nonetheless, plaque fissures are often incidental findings in advanced plaques in deaths not attributed to cardiovascular causes. 14 Alternatively, Paterson 40 proposed that intraplaque hemorrhage is secondary to rupture of vasa vasorum, a common feature of advanced lesions with plaque rupture and luminal thrombi. In our series of sudden coronary death cases, hemorrhage into a plaque was most frequent in ruptured plaques but was also observed in lesions with only 40% to 50% cross-sectional luminal narrowing. Healed Ruptures/Erosions (Figure 7) Healed ruptures are characterized by a disrupted fibrous cap filled in by smooth muscle cells, proteoglycans, and collagen (Figure 7). Healed ruptures are best identified by picrosirius red staining, 41 whereby newly synthesized type III collagen is seen overlying a ruptured fibrous cap consisting primarily of type I collagen. 20 The matrix within the healed fibrous cap defect may consist of a proteoglycan-rich mass or a collagenrich scar, depending on the phase of healing. Lesions with healed ruptures may exhibit multilayering of lipid and necrotic core, suggestive of previous episodes of thrombosis. 42 Other healed lesions show no evidence of a preexisting rupture of the fibrous cap, and there is usually no necrotic core. Instead, distinct layers of dense collagen interspersed with smooth muscle cells and proteoglycans often containing fibrin and/or platelets are present; we assume these types of lesions are the result of healed erosions (Figure 7). Examples of the Application of the Simplified Classification (Figure 8) The combination of well-defined categories with descriptive adjectives allows us to encompass varying combinations of complex lesions not easily identified by the existing AHA scheme. For example, Figure 8A shows a lesion composed of a healed rupture with a superimposed area of erosion. An example of a complex lesion with multiple, healed plaque ruptures is shown in Figure 8B; we would describe this as a healed rupture of a thin fibrous cap lesion. Other complex

1270 Arterioscler Thromb Vasc Biol. May 2000 lesions may show multiple areas of extensive calcification and a ruptured thin, fibrous cap containing an acute or resolving luminal thrombus or scar. Comparison of AHA Types With the Simplified Classification One test of the simplified scheme is the ability of our categories, with appropriate descriptive adjectives, to replace the AHA s distinctions of type III from types IV, V, and VI lesions and their alphabetic subsets. We would equate the AHA s roman-numbered terms to the following. Type III We would classify this type of lesion as a pathological intimal thickening (Figure 3) with a poorly formed fibrous cap (because of the absence of a necrotic core). Typically, these lesions show incompletely coalesced extracellular lipid, most of which is located deep within the plaque, underneath a layer of macrophages and smooth muscle cells. The smooth muscle cells may contain lipid droplets. Electron microscopy or histochemical assays for apoptosis may show evidence of ongoing cell death. Type IV This category is identical to our fibrous cap atheroma with a well-formed cellular cap overlying a confluent, necrotic, fatty core (Figure 3). Type V The AHA classifies lesions in which prominent new, fibrous tissue has formed in a plaque containing a lipid core as type Va. Again, rather than a separate category, we would describe type V lesions simply as fibrous cap atheromata with thick cellular caps overlying a largely necrotic, fatty mass. If the Figure 6. Complex lesions showing areas of rupture and erosion. A, Coronary section with plaque rupture containing a luminal thrombus (Th), disrupted thin fibrous cap, and underlying necrotic core (NC). B, A high-power view of the region outlined by the box in A, showing macrophage infiltration of the thin fibrous cap (arrows). C, Section proximal to rupture site (A) showing the propagated thrombus over a lesion resembling plaque erosion. D, A high-power view of the region outlined by the box in C showing a smooth muscle cell and proteoglycan-rich plaque. Without serial sectioning, the rupture site may not have been identified. This case exemplifies the histological diversity of some lesions that confound the diagnosis underlying the patient s demise. plaque contains a lipid core that is calcified, the AHA classifies this as type Vb. When type V lesions show marked fibrosis and little lipid, they are referred to as type Vc. Type V lesions may also show fissures, hemorrhage, and/or thrombi. Type Va lesions often show patterns of multilayering (ie, multiple layers of lipid core separated by fibrous tissue) suggestive of repetitive disruption, thrombosis, and healing. Types Va, Vb, and Vc are easily and clearly defined with our simplified scheme by applying appropriate adjectives to our fibrous cap atheroma category, without recourse to confusing and restrictive numbering schemes. The AHA scheme claims that type V lesions are more severely narrowed than are those of type IV. 4 The basis for this statement is unclear, and several studies have shown that narrowing does not appear to be directly correlated with plaque mass. 43,44 Because 20% of sudden coronary deaths occur in the absence of luminal thrombi or a healed myocardial infarct, it is essential that culprit lesions with 75% cross-sectional area luminal narrowing be classified. In addition, patients with stable angina have infrequent ( 20%) luminal thrombi but do show 50% diameter reduction by angiography. 45 Therefore, we cannot ignore these severely narrowed lesions, but instead, the pathophysiological processes leading to blood flow reduction must be sought. Type VI The AHA defines the most advanced lesions as type VI. Type VI, in our opinion, is a poorly defined category that includes lesions of both types IV and V. Those plaques with lesions showing disruption of the luminal surface are referred to as type VIa, with hematoma or hemorrhage as type VIb, and with luminal thrombi as type VIc. Type VIabc lesions are those containing features of all 3 types. 4

Virmani et al Lesion Classification and Sudden Coronary Death 1271 Figure 7. Healed plaque erosion and rupture. Healed erosion shows deep multilayering of collagen separated by elastin layers and a paucity of smooth muscle cells. The superficial plaque is rich in smooth muscle cells, collagen, and proteoglycans. Healed ruptures are lesions with rupture of the fibrous cap (arrowhead) over a necrotic core. The plaque overlying the disrupted cap (healed site) consists of smooth muscle cells, proteoglycans, matrix, and collagen (arrow). Plaques with total occlusion from prior thrombi contain mostly smooth muscle cells in a collagen- and proteoglycan-rich matrix, with capillaries and an inflammatory cell infiltrate. This section shows a necrotic core (NC), though not always present, and the lumen is filled with an organized thrombus (orgth) with multiple capillary channels. We suggest that this category is confusing and unnecessarily implies a clear progression from type IV and V lesions. Again, it is easier to refer to a well-defined category such as thin fibrous cap atheroma with the appropriate modifiers. For example, in our experience, most type VI lesions would be described as ruptured thin fibrous cap atheromas. Mechanistic Targets in Advanced Lesions The proposed scheme, including the processes and descriptive adjectives in Figure 1, suggests specific morphological features that represent the critical events and potential mechanistic targets for animal model development or clinical intervention. These include arterial narrowing, erosion, fibrous cap thinning, plaque rupture, and intramural coagulation (Table 4). The varying level of clinical importance among these targets will not be considered, since there are no data to support stratification. Narrowing Stenosis of atherosclerotic vessels is the most common morphological target for therapeutic intervention. Surprisingly, this is the change with the least understood histological basis. Since publication of the article by Glagov et al 43 in 1987, we have known that plaque mass (at least in lesions with 40% cross-sectional luminal narrowing) is not correlated with lumen size in humans. Despite innumerable studies targeting intimal mass or plaque surface area in response to balloon angioplasty, lipid feeding, or genetic alterations, it is unclear whether manipulation of any specific histological feature, other than plaque mass itself, correlates well with vessel size. There are, however, some mechanistic clues from experimental studies in the apoe-knockout mouse. Clinical studies of restenosis after angioplasty, especially those with serial examinations by intravascular ultrasound, have shown a correlation between luminal narrowing and a decrease in the external elastic lamellar area, rather than any increase in intimal or medial mass. 45 Intriguingly, studies involving 1 of us (S.M.S.) showed a similar phenomenon at certain highly consistent sites of lesion formation in the apoe-knockout mouse. 46 Analysis of our data in these susceptible areas in the mouse showed 2 possible clues as to an underlying mechanism. First, lumen loss was correlated with medial atrophy. This suggests that accommodation of the vessel lumen to an increasing mass of intima may be an active process requiring functional, viable medial smooth muscle. This seems counterintuitive, because medial atrophy is usually associated with aneurysm formation 47 ; however, the mechanisms responsible for vessel wall remodeling are essentially unexplored. Inflammation of the adventitia was the second feature associated with narrowing in the apoe-knockout studies. 46 Similarly, arterial balloon injury models in the rabbit and pig have demonstrated that adventitial fibrosis is a more impor-

1272 Arterioscler Thromb Vasc Biol. May 2000 Figure 8. Complex lesions showing multiple ruptures and erosions. Complex coronary lesions from sudden death patients are illustrated. Rupture/erosion: The previous rupture site overlying a necrotic core (NC) is identified by the arrowhead; an acute thrombus (Th) from plaque erosion occupies the lumen. Interestingly, the fibrous cap appears to be thick, and the rupture has occurred in the center of the cap. Repetitive plaque rupture: Multiple rupture sites are shown by the arrowheads. NC* corresponds to the second necrotic core, and an organizing thrombus (orgth) resulting from the second rupture is identified by the arrow. tant determinant of lumen loss than is intimal mass. 48,49 In 1 report, Katsumata and colleagues 50 demonstrated that chronic application of interleukin-1 to the adventitia alone caused luminal narrowing in pig coronary arteries. Although few autopsy studies have examined the adventitia, our study of adventitial fibrosis and thickness in human atherosclerotic coronary arteries after angioplasty (R.V.) has not been found to be correlated with lumen loss and restenosis. 51 Finally, the possible role of healing in narrowing is intriguing. There is an extensive literature on wound contracture, and there is no reason to assume that wound healing after arterial thrombosis would not lead to a similar phenomenon. Contraction of intimal wound tissue as it heals could overcome the normal capability of the media to dilate in response to an increased intimal mass. In an experiment from 1 of us (S.M.S.), narrowing of vessels after angioplasty depended on the formation of intramural fibrin, raising the intriguing possibility that fibrin could mediate wound contracture in injured vessels. 52 An extreme example of the possible role of wound healing could be the fibrocalcific lesions described by Kragel and colleagues. 38 Given the extensive matrix formation in these vessels, it seem likely that a loss of adaptive remodeling would result from highly collagenized intima restricting the ability of the vessel to dilate. This could explain the association of severe narrowing with the fibrocalcific plaque. Erosion Currently, we have little understanding of the mechanism(s) of erosion. Besides the thrombus, the most striking aspects of this lesion are the absence of an endothelium and the activated appearance of the underlying smooth muscle cells. One might assume that the endothelium has somehow been dislodged. However, animal models of atherosclerosis, contrary to some early speculations about the effects of hyperlipidemia 53 or smoking, 54,55 have not shown large areas of spontaneous denudation. Moreover, unless active tissue factor is present, it is not obvious that the simple desquamation of endothelium in a high-shear artery would lead to a coagulative and thrombotic process resulting in occlusion. Thus, erosion is a mystery in much need of a model. One possibility is that erosion is a manifestation of vasospasm. Unfortunately, little is known of the sequelae of vasospasm in atherosclerotic vessels. Fibrous Cap Thinning The mechanism of fibrous cap thinning is not known. There is, however, evidence for extensive apoptosis of smooth muscle cells within the cap of advanced atherosclerosis, as well as those cultured from plaques. 56,57 Virchow defined apoptosis in the plaque over a century ago: Thus we have here an active process which really produces new tissues, but then hurries on to destruction in consequence of its own development. 5 Consistent with Virchow s prescient point of view, plaque smooth muscle cells, both in vivo and in vitro, show limited ability to replicate, even after angioplasty. 58 60 Moreover, plaque smooth muscle cells show elevated levels of spontaneous apoptosis, as has been demonstrated both in vivo and in vitro. 58 60 Intriguingly, Pollman et al 61 found that antisense to Bax, an antiapoptotic gene, promoted cell death and caused thinning of the neointima formed by injury. A similar process of ongoing programmed cell death combined with replicative senescence occurring in the fibrous cap could be operative in the lesion we call fibrous cap thinning. A

Virmani et al Lesion Classification and Sudden Coronary Death 1273 thin fibrous cap would rupture because of its inability to maintain the cap in the face of, for example, macrophage-derived proteolysis, as discussed in the next section. Intriguingly, we have recently found that the normal intima expresses c-flip, another antiapoptotic gene, but that this gene is lost in areas of apoptosis of the fibrous cap. 62 Plaque Rupture Plaque rupture is the suspected cause of death in 60% of patients with sudden coronary death and thrombosis, and of these patients, 75% show previous sites of plaque rupture. 20 Despite the clinical importance of erosion, rupture remains a critical target for investigation. The most extensive hypothesis to explain rupture is that proposed by Libby and colleagues, an expansion of the original work described by Henney et al. 63 This hypothesis proposes that the critical effects of inflammation are the cytokines that drive the expression of proteases and obstruct the actions of proteolytic inhibitors. Particularly intriguing is the hypothesis of Hansson and colleagues, 64 67 who suggested that specific antigens elicit a T-cell response and that disease progression may be stimulated by autoimmune responses to oxidized lipoproteins. The principle limitation of this inflammation/protease hypothesis is that we do not know when the inflammatory or immune process becomes critical. That is, in experimental animals, proteolytic activity may be elevated even in early lesions that do not rupture. 68 Similarly, although atherosclerotic plaques do show clear evidence of collagenase and elastase activity, 69,70 the time course of net activity in relationship to rupture will probably remain unknown until we have an animal model for this advanced stage. Libby s work, however, offers intriguing possibilities. The first is their recent observation that stromelysin-3, a protease that itself is able to digest proteolytic inhibitors, is present only in advanced human lesions, as opposed to xanthomata. 71 It is intriguing to imagine that expression of stromelysin-3 or an accumulation of oxidation products that inactivate protease serpins 72,73 could tip the proteolytic balance. This could be especially significant when macrophage-produced protease activity is present in a fibrous cap that has lost its smooth muscle cells through apoptosis and senescence, as suggested above. Intramural Coagulation Although we know that occlusion of the vessel in sudden coronary death in most cases depends on coagulation, surprisingly little is known about the mechanisms promoting procoagulant conditions. Advanced lesions contain smooth muscle cells and macrophages that express tissue factor, and this can be shown to activate the extrinsic pathway of coagulation. 74,75 However, to our knowledge, there have not been careful studies of when this critical initiating factor for the extrinsic pathway is first found during the progress of atherosclerotic lesion formation. Even less is known about the natural expression of tissue factor pathway inhibitor or the presence of annexin V, a cytoplasmic molecule released during cell death that binds phosphatidylserine, a critical cofactor for tissue factor. Studies of the time course of tissue factor activation and appearance of procoagulant activity during plaque progression are needed in human tissue as well as in animal models. Summary In our analysis of lesions from patients after sudden coronary death, we identified plaques that were difficult to categorize by the current AHA classification scheme. Therefore, we have devised a simplified classification that relies on descriptive morphology, with minimal implication of the mechanisms involved. This classification scheme highlights specific morphological events that are appropriate targets for development of animal models or human diagnostic procedures, which will permit us to test hypotheses as to the final stages of the disease. The most important events include erosion, rupture, thinning of the fibrous cap, and development of a procoagulation and thrombotic environment. Our categorization of the lesions largely depends on the status of the fibrous cap. The critical modulators of these changes in the fibrous cap are, at best, poorly known. The most tenable hypothesis, however, is that fibrous cap thinning results from the proinflammatory activities of macrophages and lymphocytes residing in the plaque. The final test of whether this classification scheme is useful seems to us to have 3 parts. First, can we describe the range of lesions? Second, can the scheme be used to categorize and stage lesions in nonhuman species? Third, and most important to us, does this scheme highlight specific features of the plaque that require mechanistic study? This last feature is extremely important if we are to understand how plaques progress from clinically benign xanthomata or fibrous cap atheromata to lesions that kill. Acknowledgment This work was supported in part by a research grant from the National Institutes of Health (RO1 HL61799-02) (to R.V.). References 1. Davies MJ, Thomas AC. Plaque fissuring: the cause of acute myocardial infarction, sudden ischaemic death, and crescendo angina. Br Heart J. 1985;53:363 373. 2. Davies MJ. Anatomic features in victims of sudden coronary death: coronary artery pathology. Circulation. 1992;85:119 124. 3. Stary HC, Chandler AB, Glagov S, Guyton JR, Insull W Jr, Rosenfeld ME, Schaffer SA, Schwartz CJ, Wagner WD, Wissler RW. A definition of initial, fatty streak, and intermediate lesions of atherosclerosis: a report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Arterioscler Thromb. 1994;14: 840 856. 4. Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insull W Jr, Rosenfeld ME, Schwartz CJ, Wagner WD, Wissler RW. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis: a report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Arterioscler Thromb Vasc Biol. 1995;15:1512 1531. 5. Virchow R. Cellular Pathology Based on Physiological and Pathological Histology. Birmingham, Ala: Classics of Medicine Library; 1858. 6. Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med. 1999; 340:115 126. 7. Falk E, Shah PK, Fuster V. Coronary plaque disruption. Circulation. 1995;92:657 671. 8. Nobuyoshi M, Tanaka M, Nosaka H, Kimura T, Yokoi H, Hamasaki N, Kim K, Shindo T, Kimura K. Progression of coronary atherosclerosis: is coronary spasm related to progression? J Am Coll Cardiol. 1991;18: 904 910. 9. Giroud D, Li JM, Urban P, Meier B, Rutishauer W. Relation of the site of acute myocardial infarction to the most severe coronary arterial stenosis at prior angiography. Am J Cardiol. 1992;69:729 732.

1274 Arterioscler Thromb Vasc Biol. May 2000 10. Ambrose JA, Winters SL, Arora RR, Eng A, Riccio A, Gorlin R, Fuster V. Angiographic evolution of coronary artery morphology in unstable angina. J Am Coll Cardiol. 1986;7:472 478. 11. Caligiuri G, Levy B, Pernow J, Thoren P, Hansson GK. Myocardial infarction mediated by endothelin receptor signaling in hypercholesterolemic mice. Proc Natl Acad Sci U S A. 1999;96:6920 6924. 12. Prescott MF, Hasler-Rapacz J, von Linden-Reed J, Rapacz J. Familial hypercholesterolemia associated with coronary atherosclerosis in swine bearing different alleles for apolipoprotein B. Ann N Y Acad Sci. 1995; 748:283 292. 13. Fishbein MC, Siegel RJ. Pathology of coronary atherosclerosis: implications for intravascular ultrasound imaging. In: Siegel RJ, ed. Intravascular Ultrasound Imaging in Coronary Artery Disease. New York, NY: Marcel Dekker Inc; 1998:1 17. 14. Arbustini E, Grasso M, Diegoli M, Pucci A, Bramerio M, Ardissino D, Angoli L, de Servi S, Bramucci E, Mussini A, et al. Coronary atherosclerotic plaques with and without thrombus in ischemic heart syndromes: a morphologic, immunohistochemical, and biochemical study. Am J Cardiol. 1991;68:36B 50B. 15. van der Wal AC, Becker AE, van der Loos CM, Das PK. Site of intimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an inflammatory process irrespective of the dominant plaque morphology. Circulation. 1994;89:36 44. 16. Farb A, Burke AP, Tang AL, Liang TY, Mannan P, Smialek J, Virmani R. Coronary plaque erosion without rupture into a lipid core: a frequent cause of coronary thrombosis in sudden coronary death. Circulation. 1996;93:1354 1363. 17. Farb A, Tang AL, Burke AP, Sessums L, Liang Y, Virmani R. Sudden coronary death: frequency of active coronary lesions, inactive coronary lesions, and myocardial infarction. Circulation. 1995;92:1701 1709. 18. Burke AP, Farb A, Liang YH, Smialek J, Virmani R. Effect of hypertension and cardiac hypertrophy on coronary artery morphology in sudden cardiac death. Circulation. 1996;94:3138 3145. 19. Burke AP, Farb A, Malcom GT, Liang YH, Smialek J, Virmani R. Coronary risk factors and plaque morphology in men with coronary disease who died suddenly. N Engl J Med. 1997;336:1276 1282. 20. Burke AP, Farb A, Kolodgie FD, Malcom GT, Virmani R. Healed ruptured plaques are frequent in men with severe coronary disease and are associated with elevated total/high density lipoprotein (HDL) cholesterol. Circulation. 1997;96(suppl I):I-235. Abstract. 21. Burke AP, Farb A, Malcom GT, Liang Y, Smialek J, Virmani R. Effect of risk factors on the mechanism of acute thrombosis and sudden coronary death in women. Circulation. 1998;97:2110 2116. 22. Burke AP, Farb A, Malcom GT, Liang Y, Smialek JE, Virmani R. Plaque rupture and sudden death related to exertion in men with coronary artery disease. JAMA. 1999;281:921 926. 23. Fishbein MC, Siegel RJ. How big are coronary atherosclerotic plaques that rupture? Circulation. 1996;94:2662 2666. 24. de Feyter PJ, Ozaki Y, Baptista J, Escaned J, Di Mario C, de Jaegere PP, Serruys PW, Roelandt JR. Ischemia-related lesion characteristics in patients with stable or unstable angina: a study with intracoronary angioscopy and ultrasound. Circulation. 1995;92:1408 1413. 25. Velican D, Velican C. Atherosclerotic involvement of the coronary arteries of adolescents and young adults. Atherosclerosis. 1980;36: 449 460. 26. Wissler RW, Vesselinovitch D. Comparative pathogenetic patterns in atherosclerosis. Adv Lipid Res. 1968;6:181 206. 27. Armstrong ML, Heistad DD, Megan MB, Lopez JA, Harrison DG. Reversibility of atherosclerosis. Cardiovasc Clin. 1990;20:113 126. 28. Strong JP, Malcom GT, McMahan CA, Tracy RE, Newman WP III, Herderick EE, Cornhill JF. Prevalence and extent of atherosclerosis in adolescents and young adults: implications for prevention from the Pathobiological Determinants of Atherosclerosis in Youth Study. JAMA. 1999; 281:727 735. 29. Kim DN, Schmee J, Lee KT, Thomas WA. Atherosclerotic lesions in the coronary arteries of hyperlipidemic swine, part 1: cell increases, divisions, losses and cells of origin in first 90 days on diet. Atherosclerosis. 1987;64:231 42. 30. Schwartz SM, deblois D, O Brien ER. The intima: soil for atherosclerosis and restenosis. Circ Res. 1995;77:445 465. 31. McCaffrey TA, Du B, Consigli S, Szabo P, Bray PJ, Hartner L, Weksler BB, Sanborn TA, Bergman G, Bush HL Jr. Genomic instability in the type II TGF-beta1 receptor gene in atherosclerotic and restenotic vascular cells. J Clin Invest. 1997;100:2182 2188. 32. Tabas I, Li Y, Brocia RW, Xu SW, Swenson TL, Williams KJ. Lipoprotein lipase and sphingomyelinase synergistically enhance the association of atherogenic lipoproteins with smooth muscle cells and extracellular matrix: a possible mechanism for low density lipoprotein and lipoprotein(a) retention and macrophage foam cell formation. J Biol Chem. 1993;268:20419 21432. 33. Burke AP, Weber D, Farb A, Kolodgie FD, Malcom GT, Virmani R. Coronary calcification: insights from sudden coronary death victims. Z Kardiol. In press. 34. Davies MJ, Woolf N, Rowles PM, Pepper J. Morphology of the endothelium over atherosclerotic plaques in human coronary arteries. Br Heart J. 1988;60:459 464. 35. Taylor KE, Glagov S, Zarins CK. Preservation and structural adaptation of endothelium over experimental foam cell lesions: quantitative ultrastructural study. Arteriosclerosis. 1989;9:881 894. 36. Rosenfeld ME, Tsukada T, Chait A, Bierman EL, Gown AM, Ross R. Fatty streak expansion and maturation in Watanabe Heritable Hyperlipemic and comparably hypercholesterolemic fat-fed rabbits. Arteriosclerosis. 1987;7:24 34. 37. Kolodgie FD, Virmani R, Rice HE, Mergner WJ. Vascular reactivity during the progression of atherosclerotic plaque: a study in Watanabe heritable hyperlipidemic rabbits. Circ Res. 1990;66:1112 1126. 38. Kragel AH, Reddy SG, Wittes JT, Roberts WC. Morphometric analysis of the composition of atherosclerotic plaques in the four major epicardial coronary arteries in acute myocardial infarction and in sudden coronary death. Circulation. 1989;80:1747 1756. 39. Constantinides P. Plaque fissuring in human coronary thrombosis. J Atheroscler Res. 1966;6:1 17. 40. Paterson JC. Capillary rupture with intimal hemorrhage as a causative factor in coronary thrombosis. Arch Pathol. 1938;25:474 487. 41. Junqueira LCU, Bignolas G, Brentani RR. Picrosirius staining plus polarization microscopy: a specific method for collagen detection in tissue sections. Histochem J. 1979;11:447 455. 42. Davies MJ. Stability and instability: two faces of coronary atherosclerosis: the Paul Dudley White Lecture, 1995. Circulation. 1996:94: 2013 2020. 43. Glagov S, Weisenberg E, Zarins CK, Stankunavicius R, Kolettis GJ. Compensatory enlargement of human atherosclerotic coronary arteries. N Engl J Med. 1987;316:1371 1375. 44. Clarkson TB, Prichard RW, Morgan TM, Petrick GS, Klein KP. Remodeling of coronary arteries in human and nonhuman primates. JAMA. 1994;271:289 294. 45. Mintz GS, Popma JJ, Pichard AD, Kent KM, Satler LF, Wong C, Hong MK, Kovach JA, Leon MB. Arterial remodeling after coronary angioplasty: a serial intravascular ultrasound study. Circulation. 1996;94: 35 43. 46. Seo HS, Lombardi DM, Polinsky P, Powell-Braxton L, Bunting S, Schwartz SM, Rosenfeld ME. Peripheral vascular stenosis in apolipoprotein E-deficient mice: potential roles of lipid deposition, medial atrophy, and adventitial inflammation. Arterioscler Thromb Vasc Biol. 1997;17:3593 3601. 47. Thompson RW. Basic science of abdominal aortic aneurysms: emerging therapeutic strategies for an unresolved clinical problem. Curr Opin Cardiol. 1996;11:504 518. 48. Andersen HR, Maeng M, Thorwest M, Falk E. Remodeling rather than neointimal formation explains luminal narrowing after deep vessel wall injury: insights from a porcine coronary (re)stenosis model. Circulation. 1996;93:1716 1724. 49. Kakuta T, Usui M, Coats WD Jr, Currier JW, Numano F, Faxon DP. Arterial remodeling at the reference site after angioplasty in the atherosclerotic rabbit model. Arterioscler Thromb Vasc Biol. 1998;18:47 51. 50. Katsumata N, Shimokawa H, Seto M, Kozai T, Yamawaki T, Kuwata K, Egashira K, Ikegaki I, Asano T, Sasaki Y, Takeshita A. Enhanced myosin light chain phosphorylations as a central mechanism for coronary artery spasm in a swine model with interleukin-1. Circulation. 1997;96: 4357 4363. 51. Sangiorgi G, Taylor AT, Farb A, et al. Histopathology of postpercutaneous transluminal coronary angioplasty remodeling in human coronary arteries. Am Heart J. In press. 52. Courtman DW, Schwartz SM, Hart CE. Sequential injury of the rabbit abdominal aorta induces intramural coagulation and luminal narrowing independent of intimal mass: extrinsic pathway inhibition eliminates luminal narrowing. Circ Res. 1998;82:996 1006. 53. Kolpakov V, Polishchuk R, Bannykh S, Rekhter M, Solovjev P, Romanov Y, Tararak E, Antonov A, Mironov A. Atherosclerosis-prone branch regions in human aorta: microarchitecture and cell composition of intima. Atherosclerosis. 1996;122:173 189. 54. Bylock A, Bondjers G, Jansson I, Hansson HA. Surface ultrastructure of human arteries with special reference to the effects of smoking. Acta Pathol Microbiol Scand [A]. 1979;87A:201 209.

Virmani et al Lesion Classification and Sudden Coronary Death 1275 55. Pittilo RM, Bull HA, Gulati S, Rowles PM, Blow CM, Machin SJ, Woolf N. Nicotine and cigarette smoking: effects on the ultrastructure of aortic endothelium. Int J Exp Pathol. 1990;71:573 586. 56. Bennett MR, Evan GI, Schwartz SM. Apoptosis of human vascular smooth muscle cells derived from normal vessels and coronary atherosclerotic plaques. J Clin Invest. 1995;95:2266 2274. 57. Geng YJ, Henderson LE, Levesque EB, Muszynski M, Libby P. Fas is expressed in human atherosclerotic intima and promotes apoptosis of cytokine-primed human vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 1997;17:2200 2208. 58. Taylor AJ, Farb AA, Angello DA, Burwell LR, Virmani R. Proliferative activity in coronary atherectomy tissue: clinical, histopathologic, and immunohistochemical correlates. Chest. 1995;108:815 820. 59. O Brien ER, Alpers CE, Stewart DK, Ferguson M, Tran N, Gordon D, Benditt EP, Hinohara T, Simpson JB, Schwartz SM. Proliferation in primary and restenotic coronary atherectomy tissue: implications for antiproliferative therapy. Circ Res. 1993;73:223 231. 60. Nobuyoshi M, Kimura T, Ohishi H, Horiuchi H, Nosaka H, Hamasaki N, Yokoi H, Kim K. Restenosis after percutaneous transluminal coronary angioplasty: pathologic observations in 20 patients. J Am Coll Cardiol. 1991;17:433 439. 61. Pollman MJ, Hall JL, Mann MJ, Zhang L, Gibbons GH. Inhibition of neointimal cell bcl-x expression induces apoptosis and regression of vascular disease. Nat Med. 1998;4:222 227. 62. Imanishi T, McBride J, Ho Q, O Brien KD, Schwartz SM, Han DKM. Expression of cellular FLICE-inhibitory protein in human coronary arteries and in a rat vascular injury model. Am J Pathol. 2000;156: 125 137. 63. Henney AM, Wakeley PR, Davies MJ, Foster K, Hembry R, Murphy G, Humphries S. Localization of stromelysin gene expression in atherosclerotic plaques by in situ hybridization. Proc Natl Acad Sci U S A. 1991;88:8154 8158. 64. Libby P, Hansson GK. Involvement of the immune system in human atherogenesis: current knowledge and unanswered questions. Lab Invest. 1991;64:5 15. 65. Nicoletti A, Kaveri S, Caligiuri G, Bariety J, Hansson GK. Immunoglobulin treatment reduces atherosclerosis in apo E knockout mice. J Clin Invest. 1998;102:910 918. 66. Zhou X, Stemme S, Hansson GK. Evidence for a local immune response in atherosclerosis: CD4 T cells infiltrate lesions of apolipoprotein-edeficient mice. Am J Pathol. 1996;149:359 366. 67. Zhou X, Paulsson G, Stemme S, Hansson GK. Hypercholesterolemia is associated with a T helper (Th) 1/Th2 switch of the autoimmune response in atherosclerotic apo E-knockout mice. J Clin Invest. 1998;101: 1717 1725. 68. Galis ZS, Sukhova GK, Libby P. Microscopic localization of active proteases by in situ zymography: detection of matrix metalloproteinase activity in vascular tissue. FASEB J. 1995;9:974 980. 69. Sukhova GK, Shi GP, Simon DI, Chapman HA, Libby P. Expression of the elastolytic cathepsins S and K in human atheroma and regulation of their production in smooth muscle cells. J Clin Invest. 1998;102: 576 583. 70. Sukhova GK, Schonbeck U, Rabkin E, Schoen FJ, Poole AR, Billinghurst RC, Libby P. Evidence for increased collagenolysis by interstitial collagenases-1 and -3 in vulnerable human atheromatous plaques. Circulation. 1999;99:2503 2509. 71. Schonbeck U, Mach F, Sukhova GK, Atkinson E, Levesque E, Herman M, Graber P, Basset P, Libby P. Expression of stromelysin-3 in atherosclerotic lesions: regulation via CD40-CD40 ligand signaling in vitro and in vivo. J Exp Med. 1999;189:843 853. 72. Desrochers PE, Mookhtiar K, Van Wart HE, Hasty KA, Weiss SJ. Proteolytic inactivation of 1-proteinase inhibitor and 1-antichymotrypsin by oxidatively activated human neutrophil metalloproteinases. J Biol Chem. 1992;267:5005 5012. 73. Hazen SL, Hsu FF, Gaut JP, Crowley JR, Heinecke JW. Modification of proteins and lipids by myeloperoxidase. Methods Enzymol. 1999;300: 88 105. 74. Nemerson Y, Esnouf MP. Activation of a proteolytic system by a membrane lipoprotein: mechanism of action of tissue factor. Proc Natl Acad Sci U S A. 1973;70:310 314. 75. Wilcox JN, Smith KM, Schwartz SM, Gordon D. Localization of tissue factor in the normal vessel wall and in the atherosclerotic plaque. Proc Natl Acad Sci U S A. 1989;86:2839 2843. KEY WORDS: atherosclerosis/classification coronary vessels/pathology erosion rupture sudden coronary death

Lessons From Sudden Coronary Death: A Comprehensive Morphological Classification Scheme for Atherosclerotic Lesions Renu Virmani, Frank D. Kolodgie, Allen P. Burke, Andrew Farb and Stephen M. Schwartz Arterioscler Thromb Vasc Biol. 2000;20:1262-1275 doi: 10.1161/01.ATV.20.5.1262 Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright 2000 American Heart Association, Inc. All rights reserved. Print ISSN: 1079-5642. Online ISSN: 1524-4636 The online version of this article, along with updated information and services, is located on the World Wide Web at: http://atvb.ahajournals.org/content/20/5/1262 Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online at: http://atvb.ahajournals.org//subscriptions/