Picking Plaques That Pop …

Characteristics of vulnerable plaque

Histopathologic characteristics of ruptured plaques are well defined (4) (Table 1)and, except for a fibrous cap which is intact, it is reasonable to assume that vulnerable plaques exhibit similar features. Knowledge of such histopathologic characteristics should allow development of appropriate imaging techniques (5) to detect these lesions. The disrupted plaques have significantly attenuated fibrous caps, which almost always demonstrate intense macrophage infiltration. The thickness of the cap near the plaque rupture site has been reported to be 23 ±19 μm and plaques with caps 2 standard deviations above the mean (64 μ) are presumed to be vulnerable to rupture (6). Accordingly, vulnerable plaques are termed “thin cap fibroatheroma” (TCFA). The attenuation of the fibrous cap is directly related to the extent of macrophage infiltration (4). The site of minimum thickness along the longitudinal or circumferential course of the plaque is believed to be prone to rupture, and this site may not necessarily correspond to the shoulders or summit of the plaque.

Although it is a common knowledge that the plaques undergoing rupture might not be critically occlusive, the pathologic characterization of vulnerable plaques and ruptured plaques (disrupted and healed ruptures) reveals that they need to mature to sizable extent before disruption (4,7). Two-thirds of TCFA are >50% and one-quarter are >75% cross-sectionally stenotic, whereas three-quarter ruptures are >50% and one-half are >75% stenotic; <50% cross sectional narrowing is observed in only 40% TCFA and 20% in ruptures (7). It is therefore reasonable to presume that detection of sufficiently large and thin-capped plaques would be possible by morphology-based imaging techniques, such as intravascular ultrasound (IVUS).

The necrotic core in a given plaque is independent of luminal narrowing and needs to exceed a critical threshold to convey vulnerability. It occupies more than 25% of the plaque area in at least two-thirds of disrupted lesions (2,6). The necrotic core in ruptured plaques and TCFA varies from 2 to 22 mm in length and demonstrates >60° circumferential involvement of the vessel. Such dimensions may offer an adequate target to the imaging strategies. Macrophage infiltration is intimately linked to the evolution of necrotic core. During the process of subintimal migration, uninhibited ingestion of lipids converts macrophages into foam cells. Cholesterol esters and free cholesterol pack the foam cells and eventually lead to their annihilation; dying foam cells contribute to the enlargement of necrotic core. Progression from non-disrupted to disrupted lesion is associated with an increase in ratio of free-to-esterified cholesterol (8) which may result from intraplaque hemorrhage and red blood cell (RBC) deposition in the core area. The cholesterol content of the erythrocyte membrane constitutes 40% of its weight; it is elevated in hypercholesterolemia and decreases with statin treatment (9). The extent of iron deposits in the plaque and the reactivity to glycophorin (a protein specific to the RBC membrane) correspond to the size of the necrotic core. Changes in these variables parallel an increase in macrophage density, suggesting that hemorrhage may serve as an inflammatory stimulus (8). Intraplaque hemorrhage is believed to occur from the disruption of thin-walled microvessels that are lined by a discontinuous endothelium without supporting smooth muscle cell (SMC) (10). These microvessels represent a part of neoangiogenesis within the plaque confines and are associated with extensive vascular proliferation around the adventitial vascular layer.

Plaque burden usually does not compromise lumen size unless 40% or greater cross-sectional luminal narrowing has occurred, causing compensatory enlargement or remodeling of the entire vessel (11,12). In particular, lesions with hemorrhage, large necrotic cores, and macrophage inflammation are more likely to demonstrate vascular remodeling. Conversely, with nonrupture-related acute coronary events, such as those resulting from plaque erosion, plaques demonstrate arterial shrinkage. The extent of arterial remodeling is closely associated with the expression of activated metalloproteinases (13). The expansile growth of vessel ensures that vulnerable plaques are sizable and enhances their ability to be imaged.

More than one-half of TCFA show either an absence or only speckled calcification, and the other half demonstrate a large variation in the degree of calcification (5), which is dependent on patient age. From the pathologic standpoint, coronary calcification correlates with the magnitude of plaque burden, but does not offer further association with plaque vulnerability. Based on the available information, it is unlikely that a measure of calcification alone would allow assessment of plaque vulnerability.

Incidence and distribution of TCFA

In patients dying of an AMI, the mean number of TCFA could be as high as 2.5 in men and 1.5 in women (3,7). In contrast, the incidence of TCFA in patients dying suddenly is a little over one per heart and similar in both genders. The majority of TCFA occur predominantly in the proximal portion of the three major coronary arteries, and about one-half arise in the mid-portion of these arteries; they are only infrequently seen in the distal coronary circulation. Because the vulnerable plaques are not abundant and are often located proximally in major arteries, an effort to detect vulnerable plaques is justified.

IVUS for the detection of vulnerable plaques

Cross sectional IVUS images of normal coronary arteries demonstrate a sharp endothelial border, thin, clear echolucent media, and dense adventitial layer (14). When intimal thickening occurs during development of atherosclerotic plaques (11), signal free zones represent lipid accumulations, soft echoes correspond to SMCs, fibrosis or collagen deposition, and bright intensity indicates calcification (15,16). Further, rupture or ulceration of lesions can often be detected in culprit vessels (17). An IVUS interrogation of all three epicardial vessels in more than 200 patients (18) closely concurred with the pathologic literature that plaque rupture of the infarct-related vessel occurred in two-thirds of patients presenting with AMI; of these, one-fifth of the patients had evidence of multiple ruptures. The plaques underlying culprit lesions were hypoechoic, eccentric, and positively remodeled. It is therefore expected that vulnerable plaques would demonstrate similar findings. Such presumption has been vindicated by serial observations in a two-year follow-up of 100-plus patients (19). The lesions that developed into an acute coronary event were characteristically large and contained relatively shallow but prominent echolucent zones. These features support the concept that vulnerable plaques are generally not trivial, have a large necrotic core and thin fibrous covers, and demonstrate outward remodeling at the site of culprit lesions.

Supplementing ultrasonic strategies for the detection of vulnerable plaques

Several modifications of ultrasonic techniques have allowed improved characterization of vulnerable plaques. For instance, palpography (or elastography) assesses local deformation of the fibrous cap tissue caused by the application of intraluminal pressure on the plaque surface (20,21). This technique can demonstrate significant differences between fibrous and fatty tissue in autopsied coronary and femoral arteries. Clinical studies have revealed high strain (1% to 2%) in non-calcified and low strain (0% to 0.2%) values in calcified areas of plaque. It is expected that high strain or vulnerable regions will be better characterized with the development of three-dimensional elastography for the full length interrogation of a coronary artery.

Similarly, contrast echocardiography has been used to evaluate plaque vulnerability by IVUS (22). Since vulnerable plaques have extensive adventitial network of vasa vasorum and intraplaque neoangiogenesis, it has been presumed that increased contrast intensity should indicate plaque vulnerability. Further, negatively-charged microbubbles may transiently attach to positively charged endothelium (of the neovessels) to enhance their residence time and detectability (23). The increased porousness of the neoangiogenic vessels, which allows constant RBC leakage, may (although untested) allow similar exit for microbubbles, which might then be phagocytosed by core macrophages to render them echogenic (24). Such detection can be further enhanced by imparting targeting specificity to microbubbles directed at the novel receptors expressed on macrophages. In one such exploitation of shell properties in an experimental model of atherosclerosis, phosphatidylserine was incorporated in the liposome contrast agent to target the ubiquitously expressed phosphatidylserine receptors of macrophages (25).

Two studies published in this issue of the Journal(26,27), offered refined analysis of IVUS data s for quantitative tissue characterization of plaque composition. Whereas Kawasaki et al. (26) applied three-dimensional IVUS with integrated color-coded backscatter maps, Murashighe et al. (27) used a mathematical model for wavelet analysis of radiofrequency IVUS signals. These studies suggest that identification of lipid-rich plaques from fibrous lesions may become clinically feasible. The former study evaluated the interval change in atherosclerotic lesion composition in response to treatment with statins or placebo, and demonstrated the feasibility of detection of reduction in lipid volume and concomitant increase in fibrous (and mixed) lesion volume in response to statins even before regression in geometric parameters became apparent (26). The latter study (27) accurately differentiated lipid-rich from fibrotic plaques that were histologically verified in tissue specimens from necropsy or directional atherectomy.

Quo vadis vulnerable plaque imaging

Although detection of various morphologic features of vulnerable plaque should become possible with IVUS techniques, it will be reasonable to propose that simultaneous efforts are invested in development of strategies that define macrophage infiltration of fibrous caps. For IVUS to become capable of such identification, it will be necessary to refine targeted imaging with microbubbles (28). It may also be possible to superimpose functional imaging strategies onto morphology-based techniques. The functional imaging can be performed by radionuclide agents, such as F-18 FDG (29) or Tc-99m-labeled Annexin-A5 which target macrophage infiltration (30,31). Intravascular beta detectors are now being investigated to provide estimates of macrophage infiltration within the atherosclerotic lesions (32).

Another novel attempt employing intravascular magnetic resonance imaging (MRI) is presented in this issue of the Journal(33). Schniderman et al. (33) used a self-contained MRI probe integrated within the tip of a coronary catheter that provided cross-sectional images with a field of view of 250 μm in depth and lateral resolution of 60°. Intravascular MRI was able to correctly identify shallow lipid cores, predominantly fibrous lesions and interestingly foam cell rich fibrous caps. Although the need to identify vulnerable plaque (beyond detecting the vulnerable patient) has not yet found universal acceptance, we believe that it will constitute one of the most important developments in our fight against acute coronary events.