Definition and Clinical Evidence

A high-risk VP is a nonobstructive, silent coronary lesion that suddenly becomes obstructive and symptomatic (Fig. 59-1). The initial understanding of this process was described by Ambrose and Fuster in 1988 when they studied the baseline characteristics of nonobstructive lesions that evolved into acute myocardial infarction (AMI).⁷ The investigators found that at baseline, the lesions evolving to complete occlusion had a mean diameter stenosis of 48%. Multiple investigators reproduced this finding using both angiographic and intravascular imaging modalities and supporting the concept that the majority of lesions responsible for AMI originate from nonobstructive disease.⁸

Figure 59-1 Sequential coronary angiograms of the left anterior descending coronary artery performed with 12-month intervals. Rapid progression from being nonobstructive and asymptomatic to being severely obstructive and symptomatic fulfills the clinical definition of vulnerable plaques. A, Completely normal vessel by angiography. B, Non-obstructive, asymptomatic coronary disease in the proximal segment (black arrows). C, Obstructive, symptomatic coronary disease presenting with acute coronary syndrome (black arrows). D, Stenosis is 10% to 20% in the segments 12 months after local therapy with stent (36 months after baseline angiogram).

(From the cardiac cath lab at Mount Sinai Medical Center, New York, NY.)

Incidence

The number of nonobstructive, asymptomatic lesions that progress to obstructive, symptomatic lesions presenting as an acute event can give us the estimated incidence of this condition. Several studies have retrospectively addressed this issue. The first study included 1228 patients who underwent percutaneous coronary intervention (PCI) for symptomatic coronary artery disease (CAD). The incidence of nonobstructive, nonculprit lesions that required additional PCI was 12.4% in the first year and 5% to 7% per year from years 2 to 5 after the initial procedure (Fig. 59-2).⁹ The second study included 3747 post-PCI patients from the National Heart Lung and Blood Institute (NHLBI) registry.¹⁰ The incidence of nonobstructive, nonculprit lesions that required additional PCI was 6% in the first year, ranging from 4.4% to 12.8% according to the number of vessels involved (Fig. 59-3). A third study included a total of 1059 patients studied with computed tomographic angiography (CTA). VPs were classified according to the presence or absence of two variables: (1) positive remodeling, and (2) low attenuation (“soft”) morphology on CT.¹¹ The incidence of acute coronary events was 11% in plaques with one variable and up to 22% in plaques with both variables. Notably, the absence of these two variables represented a strong negative predictive value, with an incidence of coronary events as low as 0.5%. Another study using intravascular modalities recently reported a 12% incidence of TCFAs by optical coherence tomography (OCT) in sites other than the culprit lesion responsible for the acute coronary syndrome (ACS).¹² Additionally, they found that the majority of TCFAs occurred in the proximal segments of the coronary vasculature. Despite this incidence of 11% to 22%, the actual area occupied by these plaques in the coronary circulation may not be as large. Cheruvu and Virmani performed detailed histopathology in the entire vasculature of 50 human hearts and demonstrated that only 1.1% to 1.6% of the coronary vessel area was occupied by TCFAs. Out of all the lesions studied, 10.8% corresponded to TCFA.¹³ This is in agreement with a second study that identified a 11% incidence of TCFAs, with slightly higher incidence in patients with ACS, suggesting a more “unstable” profile of the coronary lesions in these subjects.^8,14 These retrospective and pathologic studies gave the foundation to design the first prospective natural-history study of VPs, also called the PROSPECT (Predictors of Response to CRT) Trial.¹⁵ A total of 697 patients with ACS underwent three-vessel coronary angiography and gray-scale and radiofrequency (RF) intravascular ultrasound (IVUS) after multi-vessel coronary stenting. The incidence of nonculprit lesions developing major adverse cardiovascular events (MACEs) was 11.6% at 3.4 years. Most of these events were rehospitalization for progressive angina. Cardiac death or MI occurred in only 4.9% of the population. Three independent variables predicted these events and include (1) plaque burden of 70% or greater (hazard ratio [HR] 5.03; P < 0.001); (2) minimal lumen area (MLA) of 4 mm² or less (HR 3.21; P = 0.001); and (3) RF classification of TCFA (HR 3.35; P < 0.001). Of great interest, plaques exhibiting all three features had a higher HR, up to 11.05 (P < 0.001). Nevertheless, the prevalence of these high-risk plaques was only 4.2%. As a result, we now know that high-risk plaques are uncommon and are composed by a combination of anatomic features, which includes plaque burden and MLA, in addition to TCFA.

Figure 59-2 Hazard rates per year for target-lesion (blue) and nontarget-lesion (red) events derived from life table survival analysis. Target-lesion events include any repeat revascularization or other event (i.e., death, myocardial infarction [MI], acute coronary syndrome [ACS], or congestive heart failure [CHF]) attributed to the target lesion. Non–target-lesion events include all repeat revascularizations involving the target vessel outside the target lesion, any non–target vessel revascularization, and any death, MI, ACS, or CHF that was clearly not attributable to the target lesion.

(Adapted with permission from Cutlip DE, Chhabra AG, Baim DS, et al: Beyond restenosis: Five-year clinical outcomes from second-generation coronary stent trials, Circulation 110:1226–1230, 2004.)

Figure 59-3 Kaplan–Meier analysis of nontarget lesion percutaneous coronary intervention (PCI) at 1 year, according to the initial degree of coronary artery disease.

(Adapted with permission from Glaser R, Selzer F, Faxon DP, et al: Clinical progression of incidental, asymptomatic lesions discovered during culprit vessel coronary intervention, Circulation 111:143–149, 2005.)

Risk Factors

Multiple regression analysis identified multi-vessel CAD at baseline (three- and two-vessel CAD), previous PCI, and age less than 65 years as independent predictors for VP. Of note, treatment with statins failed to protect patients within the first year.¹⁰ IVUS studies also identified age, hypertension, diabetes mellitus (DM), heart failure, and dyslipidemia as predictors for IVUS-derived TCFA (ID-TCFA).¹⁶ Diabetes is strongly associated with higher rates of ID-TCFA (21.6% vs. 13.1%, P = 0.01), and up to 54.4% in patients with a diagnosis of DM greater than 10 years.¹⁷ The incidence of TCFA was also seven times higher in males compared with females.¹³ In terms of biomarkers and VP, serum myeloperoxidase is a new potential marker of plaque vulnerability.^18,19 Similarly, Lipoprotein-associated phospholipase (Lp-PLA2) may play a key role in necrotic core expansion as demonstrated by the Integrated Biomarkers and Imaging Study-2 (IBIS-2).²⁰ In isolated studies, highly sensitive cross-reactive protein (hs-CRP), white blood cell (WBC), interleukin (IL) 18, and tumor necrosis factor alpha (TNFα) inversely correlated with fibrous cap thickness. However, hs-CRP appeared to be the only independent predictor by regression analysis.²¹ Furthermore, hs-CRP has been postulated as a modulator of neovascularization.²² Nevertheless, the clinical evidence for primary prevention of hs-CRP is rather limited and still subject to controversy, and further research is needed to completely elucidate the role of biomarkers to predict plaque rupture and coronary events.^23–25

Anatomic Distribution

The anatomic distribution of coronary lesions responsible for AMI is dominated by the proximal segment, which is responsible for 80% of MIs in all three major vessels.²⁶ These data were recently reproduced by OCT studies in vivo (Fig. 59-4).^12,27,28 When located in bifurcations, TCFAs are predominately located in the proximal rim of the bifurcation, as evaluated by OCT and IVUS.²⁹ After reviewing clinical characteristics, the next section provides the basis for understanding the pathophysiology of the disease. Most importantly, this section offers the foundation to critically evaluate novel imaging techniques that claim effectiveness in the diagnosis of high-risk VPs. For the interventionalist, the incidence of high-risk VPs evolving into clinical events is close to 13% in 3 years, in patients with ACS and multi-vessel disease. These lesions are positively remodeled, have large plaque burden, and are usually located in the proximal segment of the coronary arteries.

Figure 59-4 Distribution of thin cap fibroatheromas (TCFAs) in the proximal, middle, and distal segments of the three coronary arteries.

(Reproduced with permission from Tanaka A, Imanishi T, Kitabata H, et al: Distribution and frequency of thin-capped fibroatheromas and ruptured plaques in the entire culprit coronary artery in patients with acute coronary syndrome as determined by optical coherence tomography, Am J Cardiol 102:975–979, 2008.)

Thin Fibrous Cap with Increased Stress–Strain Relationship

Autopsy studies have shown that ruptured plaques are characterized by a very thin fibrous cap, measuring 23 ± 19 microns (µm) in thickness. Most importantly, 95% of ruptured caps measured 64 µm or less in the coronary and 60 µm or less in the aorta.³² As a result, the first and probably most important histologic feature of TCFA is a fibrous cap 65 µm or more in thickness. These thin caps are unable to withstand the circumferential tensile stress applied by the oscillations of arterial blood pressure. The ratio of the circumferential tensile stress to the radial strain of the fibrous cap equals the stiffness of the tissue.³³ Hence, soft (fatty) tissue will be more strained than stiff (fibrous) tissue when equally stressed. Furthermore, as caps become thinner, the stress increases in an exponential pattern (Fig. 59-8). In addition, as lipid pools become larger, stress also increases. Therefore, the strength of a cap may be as important as the actual thickness of a fibrous cap. This stress–strain relationship in the fibrous cap is therefore considered a feature for plaque vulnerability.³⁴ Technology aiming to detect TCFA should have a radial resolution less than 65 µm and the ability to quantify the stress–strain relationship in the fibrous cap.

Figure 59-8 Relationship between circumferential stress (vertical axis) and fibrous cap thickness (horizontal axis). Note the exponential increase in circumferential stress when cap thickness is reduced to fewer than 200 microns.

(Adapted with permission from Loree HM, Kamm RD, Stringfellow RG, Lee RT: Effects of fibrous cap thickness on peak circumferential stress in model atherosclerotic vessels, Circ Res 71:850–858, 1992.)

Large Necrotic Core with Increased Free Cholesterol to Esterified Cholesterol Ratio

Modified oxidized low-density lipoprotein (LDL) is avidly taken up by macrophages via scavenger receptors, leading to cytoplasm overload with lipid droplets. Continuous inflow of oxidized isoforms of LDL (oxLDL) leads to cell death with extracellular lipid accumulation within the matrix of the plaque. Preclinical experimentation has demonstrated how the oxLDL are associated with plaque vulnerability depending on the type of predominant molecule, mostly regulating apoptotic cell death.^35,36 This is the basic mechanism of the necrotic core, which is formed after death by necrosis, or apoptosis, of lipid-laden macrophages, foam cells, and erythrocytes. Active collagen dissolution by metalloproteinases contributes to core expansion, which plays a major role in plaque vulnerability. In the aorta, TCFAs exhibit necrotic core areas of 40%, and ruptured plaques up to 50% of total plaque area.³⁷ However, other studies in coronary arteries show lower necrotic areas, down to 24% and 34% in TCFAs and ruptured plaques, respectively.³⁰ Core composition may influence the propensity for plaque rupture and thrombosis. An increased free cholesterol–esterified cholesterol ratio, with oxidized cholesterol, increases the likelihood of thrombosis by interacting with the oxLDL receptor-1 (LOX-1), and enhancing the expression of tissue factor.^38–40

Increased Plaque Inflammation

Macrophages and T cells are capable of degrading the extracellular matrix by secretion of proteolytic enzymes such as plasminogen activators and matrix metalloproteinases (MMPs), including collagenases, elastases, gelatinases, and stromelysins, weakening the already thin fibrous cap and predisposing it to rupture.⁴¹ This has been recently confirmed by Suzuki et al, who found increased levels of MMP-1, MMP-13, and IL-6 in the coronary vessels of patients who had experienced MI, supporting the evidence behind the role of MMP in plaque rupture that leads to clinical events.⁴² Among the multiple MMPs identified to have a role in atherosclerosis, two particular subtypes, MMP-7 and MMP-12, seem more selectively localized in specific subsets of macrophages; especially, those located in the necrotic core have low arginase-I expression and represent a marker of activation and enhanced inflammation.^43–45 The continuous entry, survival, and replication of monocytes or macrophages within plaques are aimed to remove oxLDL and reduce oxidation and reactive oxygen generation (ROS) products. In situations where the macrophage scavenger capacity is overloaded, cell death is activated by apoptosis, releasing MMPs and tissue factor.⁴⁶ This link among inflammation, apoptosis, and thrombosis was elegantly documented by Hutter et al in human and murine atherosclerotic lesions.⁴⁷ Therefore, plaque inflammation is a pivotal feature of plaque vulnerability.^4,48 Recent studies have reported two different subclasses of monocytes or macrophages inside the atheroma.⁴⁹ The first class of macrophages, classically known as activated macrophages, or M1, promote inflammation. The second subtype, identified as M2, appears to be anti-inflammatory. Therefore, the ratio between M1 and M2 macrophages has an impact on atherosclerosis progression, plaque regression, or both. It has been described that different factors shift this ratio toward an increased content of M2 cells, including increased T helper–secreted molecules, IL-4, IL-13, peroxisome proliferator-activated receptor delta (PPAR-δ) and the lipid sphingosin 1-phosphate (S1P).^50–52 These M2 macrophages are also responsible for efferocytosis, or the removal of short-lived apoptotic bodies from the atheroma. Defective efferocytosis leads to secondary necrosis, a process linked to advanced disease. The failure of macrophage efflux, prothrombotic factors, inflammatory chemokines, and collagenases increase the necrotic core, leading to plaque rupture and thrombosis.^53,54

Technology aiming to detect TCFAs should have sufficient resolution to identify and quantify macrophages in the fibrous cap and shoulders of the atherosclerotic plaque.

Degree of Vascular Remodeling

As a defense mechanism, the vessel wall can expand significantly to harbor large atheromas without obstructing the lumen (Fig. 59-9). Also known as remodeling, this process is linked to high-risk atherosclerosis. The mechanisms responsible for remodeling involve an inflammatory process at the base of the plaque, which leads to the digestion of the internal elastic lamina (IEL) and involves the deeper layers of the vessel wall, including the tunica media and the adventitia. Several studies have shown increased expression of MMPs within the intimo-medial interface of remodeled plaques.^55,56 Furthermore, disruption of the IEL is associated with medial and adventitial inflammation, and evidence suggests that MMP-10 plays a regulatory role in atherosclerosis progression by influencing plaque inflammation and vascular remodeling.^32,57 Concordantly, Burke et al demonstrated that marked expansion of the IEL occurred also in plaque hemorrhage, with or without rupture.⁵⁸ The clinical relevance of remodeling was pioneered by Schoenhagen et al, who studied 85 patients with unstable coronary syndromes and 46 patients with stable coronary syndromes by using IVUS before PCI.⁵⁹ Remodeling ratio (RR) was defined as the area of the external elastic membrane at the lesion divided by the same area at the proximal reference site. Positive remodeling was defined as an RR greater than 1.05 and negative remodeling as an RR less than 0.95 (Fig. 59-10). The RR was higher at target lesions in patients with ACS than in patients with stable angina. As a result, positive remodeling was more frequent in ACS (51.8% vs. 19.6%), whereas negative remodeling was more frequent in stable angina (56.5% vs. 31.8%) (P = 0.001), confirming the histopathologic associations between plaque remodeling and vulnerability.⁵⁹ New technology aiming to detect TCFAs should provide exact measurements to quantify the degree of vascular remodeling. Of clinical relevance, Corti et al were the first to document the same eccentric pattern for plaque regression after aggressive lipid therapy.⁶⁰ More recently, multiple studies have confirmed this observation (Figure 59-11).^61–63 Considering that lipid is the main plaque component that can be reversed with therapy, this eccentric pattern of plaque regression suggests an effective reverse lipid transport system through the deeper layers of the vessel wall, probably mediated by vasa-vasorum neovascularization.^4,64

Figure 59-9 Large human thrombotic plaque in the left main coronary artery, with extensive remodeling containing a large necrotic core.

(Courtesy of Dr. K-Raman Purushothaman, Mount Sinai Hospital, New York, NY.)

Figure 59-10 Cartoon explaining the direction of positive and negative remodeling. (See text for details.)

(Adapted with permission from Schoenhagen P, Ziada KM, Kapadia SR, Crowe TD, Nissen SE, Tuzcu EM: Extent and direction of arterial remodeling in stable versus unstable coronary syndromes: An intravascular ultrasound study, Circulation 101:598–603, 2000.)

Figure 59-11 A, Volume-rendered high-resolution, three-dimensional micro-CT image of the descending aorta vasa vasorum. B and C, Corresponding histologic cross-sections demonstrate atherosclerotic lesions in the inferior vena cava (black arrow). D, Highlighted differentiated arterial (red) and venous (blue) vasa-vasorum. (Masson trichrome stain, bar 500 microns).

(Reproduced with permission from Langheinrich AC, Michniewicz A, Sedding DG, et al: Correlation of vasa vasorum neovascularization and plaque progression in aortas of apolipoprotein E(-/-)/low-density lipoprotein(-/-) double knockout mice, Arterioscler Thromb Vasc Biol 26:347–352, 2006.)

Vasa-Vasorum Neovascularization

Atherosclerotic neovascularization evolves in early atherogenesis as a defense mechanism against hypoxia and oxLDL deposition within the tunica intima.^65,66 In advanced disease, neovessels may play a defensive role allowing for lipid removal from the plaque through the adventitia leading to plaque regression, as described above. The adventitial vasa-vasorum is the main source of neovascularization in atherosclerotic lesions (Fig. 59-12). Neovascularization, elegantly delineated by Barger et al, who used cinematography (Fig. 59-13), is distributed from the epicardial fat to the plaque throughout vessel wall, although some can originate directly from the vessel lumen.⁶⁷ Nevertheless, neovessels from the adventitial vasa were 28 times more numerous (96.5%) compared with those from the luminal side (3.5%).⁶⁸ Neovessels from the adventitial vasa characterized severely stenotic lesions and correlated with the extent of inflammatory cell infiltration and lipid core size. Conversely, neovessels from a lumen origin were found in plaques with 40% to 50% stenosis and were associated more often with intra-plaque hemorrhage or hemosiderin deposits.⁶⁸ Neovessels may also serve as a pathway for leukocyte recruitment to high-risk areas of the plaque, including the cap and the shoulder. Expression of vascular cell adhesion protein 1 (VCAM-1), intercellular adhesion molecule 1 (ICAM-1), and E-selectin is twofold to threefold higher on neovessels compared with the arterial luminal endothelium, which confirms the pivotal role of neovessels as a pathway for leukocyte recruitment in human coronary plaques.⁶⁹ Angiogenesis is stimulated by inflammatory cells. For instance, apoptotic microvesicles at the submicron level, found in atherosclerosis plaques, are highly proangiogenic by regulating CD40L and are produced mostly from macrophages.^70,71 Histologic evidence for atherosclerotic neovascularization as a pathway for macrophage infiltration in advanced, lipid-rich plaques is also documented (Fig. 59-14).⁷² Neovessel content was significantly increased in plaques with severe inflammation, associated with both increased macrophage and T lymphocyte infiltration.^73,74 Moreover, ruptured plaques exhibited the highest degree of neovascularization.⁷⁵ Further analysis of plaque angiogenesis in diabetes documented a complex morphology, including sprouting, red blood cell (RBC) extravasation, and perivascular inflammation.⁷⁶ Lastly, neovascularization may increase calcification, as demonstrated by the close link between immature endothelial cells and osteoblasts in atherosclerotic plaques and in cellular tissue, expressing vascular calcification–associated factor (VCAF).⁷⁷ Using micro-CT imaging, recent studies have shown increased vasa-vasorum density, along with iron and glycophorin-A content in nonstenotic, noncalcified plaques. Moreover, calcium content was inversely proportional to neovessel content.⁷⁸ Technology detecting TCFAs should quantify vasa-vasorum neovascularization in the adventitia, in the tunica media, and within the atherosclerotic plaque.

Figure 59-12 Top panels, The baseline and follow-up intravascular ultrasound (IVUS) images of a single coronary cross-section after 24 months of rosuvastatin treatment. Bottom two panels, Measurements superimposed on the same cross-sections, demonstrating the reduction in atheroma area. EEM, external elastic membrane.

(Reproduced with permission from Nissen SE, Nicholls SJ, Sipahi I, et al: Effect of very high-intensity statin therapy on regression of coronary atherosclerosis: The ASTEROID trial, JAMA 295:1556–1565, 2006.)

Figure 59-13 Color prints taken during the injection of silicone polymer into the coronary arteries of cleared hearts, demonstrating regions of vessels with abundant neovascularization. The positive regions are composed of networks of small-caliber vessels coming from the adventitia and penetrating the tunica media into the atherosclerotic plaque.

(Reproduced with permission from Barger AC, Beeuwkes R, 3rd, Lainey LL, Silverman KJ: Hypothesis: Vasa vasorum and neovascularization of human coronary arteries. A possible role in the pathophysiology of atherosclerosis, N Engl J Med 310:175–177, 1984.)

Figure 59-14 Histologic evidence of atherosclerotic neovascularization as a pathway for macrophage infiltration in human aortic plaques obtained at autopsy. Bicolor, contrasting immunohistochemical technique showing microvessels in cross-sections (red arrows) identified with the monoclonal endothelial cell marker CD34 linked to a blue chromogen and inflammatory cells identified with a combined macrophage–T cell marker CD68-CD3 linked to a red chromogen.

(Reproduced with permission from Fuster VM, Fayad PR, Corti ZA, Badimon RR: Atherothrombosis and high-risk plaque part I: Evolving concepts, J Am Coll Cardiol 46:937–954, 2005.)

In summary, plaque neovascularization, initially a defense mechanism to provide oxygen and to remove lipid from the plaque, may eventually fail leading to extravasation of RBCs, perivascular inflammation, and intra-plaque hemorrhage.

Intra-plaque Hemorrhage

Neovessel leakage leads to extravasation of plasma content and RBCs into the plaque, which is known as intra-plaque hemorrhage (IPH). The mechanisms by which IPH occurs involves the leakage of RBCs from immature neovessels, which is mediated by various growth factors and chemokines that are expressed more in VPs.⁷⁹ Compared with nonhemorrhagic atheromas, evidence has demonstrated lower levels of vascular endothelial growth factor (EGF), placental growth factor (PGF), and angiopoietin-1, in combination with an increased vascular endothelial growth factor (VEGF) expression, which are characteristic of hemorrhagic plaques.^80,81 Lysis of RBCs contributes to lipid deposition. In addition, extracorpuscular hemoglobin (Hb) can induce oxidative tissue damage by virtue of its heme iron, with subsequent production of reactive oxygen species.⁷⁹ The defense mechanism against free Hb is haptoglobin (Hp), which irreversibly binds to free Hb forming an Hp–Hb complex. Two classes of alleles (Hp-1 and Hp-2) characterize the human Hp locus at chromosome 16q22. The protein products of the two Hp alleles are structurally different, and the cardiovascular effects of this Hp polymorphism play a major role in patients with DM.^82–84 Multiple independent epidemiologic studies examining incident cardiovascular disease have demonstrated that individuals with DM and the Hp 2-2 (homozygous for the Hp 2 allele) genotype have the risk of cardiovascular events four to five times higher compared with individuals with the Hp 1-1 (homozygous for the Hp 1 allele) genotype. The mechanism by which this Hp 2-2 phenotype regulates inflammation and enhances plaque vulnerability is related to decreased Hp clearance in the Hp 2-2 phenotype. This is explained, in part, by a defective CD163 receptor in macrophages, which is the receptor in charge of removing the Hp–Hb complex from the atheroma.⁸⁵ Technology detecting TCFAs should identify intra-plaque hemorrhage, iron deposition, RBC membranes, and hemosiderin deposits in macrophages. In patients with DM, Ha genotyping may offer additional prognostic value.

Plaque Composition: Summary

Atherosclerotic plaques at high risk for rupture are composed of several features, including a large lipid core, a thin fibrous cap with increased shear stress, macrophage infiltration, positive remodeling, vasa-vasorum neovascularization, and increased intra-plaque hemorrhage. These concepts apply only to lesions at risk for plaque rupture and thrombosis (TCFAs). Of note, plaques at risk for erosion and thrombosis do not exhibit any specific morphologic feature that can be detected by any imaging technique at this point. This is a significant limitation of the individual, lesion-oriented approach to the high-risk VP hypothesis. Nevertheless, even by merely identifying and treating TCFAs, a significant reduction of clinical events can be achieved.

Invasive Techniques

Coronary angiography has the ability to delineate the coronary lumen, but it does not provide any information about the vessel wall. Therefore, this technique is not appropriate for the detection of VPs. Thus multiple intracoronary imaging techniques have been proposed to identify TCFAs. Of pivotal importance, every imaging technique should have appropriate validation, with histology as the gold standard to identify key histo-morphologic components. Considering that the majority of atherosclerotic plaques (TCFAs and non-TCFAs) will have a certain degree of fibrous cap thickness, shear stress patterns, necrotic core area, macrophage area, positive remodeling, and vasa-vasorum neovascularization, the presence or absence of these features alone (sensitivity and specificity) is not enough. Proper histologic validation must include accurate assessment of the degree of these components, which involves linear regression analysis. This validation process should be confirmed in animal models of TCFA before being applied in human coronary arteries.⁸⁶ Then, the ultimate test should be a natural history study of all these techniques to determine if specific plaque components have any prognostic implications. Novel intracoronary techniques to detect TCFAs include (1) IVUS, (2) virtual histology (VH), (3) palpography, OCT, (4) IVUS elastography, or (5) intravascular magnetic resonance imaging (MRI), (6) angioscopy, (7) spectroscopy, and (8) thermography. A summary of these techniques, the component detected, and the resolution or accuracy is presented in Table 59-1. The first seven are already available for use in clinical practice or are under active evaluation in humans. With regard to thermography and the noninvasive modalities, an evidence-based approach for the understanding of their clinical usefulness will be presented. The interventionalist must develop a critical approach to evaluating these novel techniques, understanding their potential, and, most importantly, discerning their multiple limitations before considering them for clinical use.

TABLE 59-1 Summary of Current Invasive Detection Technologies

Intravascular Ultrasound

Unlike angiography, IVUS allows proper visualization of the disease in the vessel wall and provides cross-sectional and longitudinal images of atherosclerotic plaques in vivo.⁸⁷ IVUS is based on transmitting and receiving high-frequency sound waves from tissue through a low-profile catheter (approximately 1 mm); reaching a radial resolution between 100 to 250 µm. IVUS is safe, quick, and easy. Most importantly, IVUS allows identification of hemodynamically significant lesions that may be underestimated by angiography, particularly in nonocclusive plaques with positive remodeling. In addition, IVUS delineates the degree of calcification, plaque burden, and the degree of arterial remodeling. It uses the amplitude of the backscattered ultrasound signal to differentiate highly echogenic components such as calcium and dense fibrous tissue from echolucent tissue, including lipid and necrotic core. However, it cannot clearly differentiate between fibrous and fatty plaques.⁸⁸ As a consequence, it is accepted that grayscale IVUS, as an isolated technology, is not capable of distinguishing plaque types. Thus, the application of other imaging protocols and algorithms to IVUS, including integrated backscatter IVUS analysis and Virtual Histology (IVUS-VH), which is discussed below, may contribute to better identification of the different plaque components. Researchers have studied backscatter analysis that extracts frequency components of a signal buried in the original IVUS signal. The imaging signal from a small volume of tissue creates an integrated backscatter (IB) pattern.⁸⁹ Several studies have reported on the IVUS characteristics of culprit lesions and the presence of multiple ruptured plaques in patients with acute coronary events.^87,90 A recent study evaluated the long-term outcomes of VPs arbitrarily defined as plaques with rupture, lipid core, dissection, or thrombus by conventional IVUS during ACS both in culprit and nonculprit locations. Multiplicity of VPs in the nontarget vessels (HR 2.2; 95% confidence interval [CI] 1.4–3.4, P = 0.001) was the only independent predictor of long-term critical events. Finally, DM and ACS were significantly associated with the multiplicity of VP.⁹¹

The ability of IVUS to identify TCFAs can be summarized as follows:

1. Fibrous cap thickness: Considering that the resolution of IVUS is lower than that needed to detect TCFAs, efforts at quantifying cap thickness with IVUS will always overestimate the expected values provided by histology. With IVUS, ruptured plaques show thinner caps compared with nonruptured plaques (Fig. 59-15).⁹² However, when ruptured plaques are evaluated with histology, the mean cap thickness is about 8 to 10 times lower than the resolution of IVUS, 23 ± 19 µm in the coronary, and 34 ± 16 µm in the aorta.³⁷ This significant overestimation of cap thickness by IVUS is related to its poor axial resolution, which makes it very unlikely for any IVUS-related technology to be able to detect TCFAs, the most common form of VP.⁹³

2. Necrotic core area: The sensitivity of IVUS to detect the necrotic core has been reported to be 46%, with a specificity of 97%. Several studies have been performed trying to improve these results, using an integrated back scatter approach.^94,95 A prospective study demonstrated that 93% of the clinical events occurred in plaques with large echolucent areas, a surrogate of the necrotic core, suggesting a prognostic value for IVUS.⁹⁶ Similarly, recent studies propose the use of attenuation of the echo signal inside the plaque as being indicative of large necrotic cores, but no histologic validation has corroborated this finding.⁹⁷ We conclude that although IVUS provides useful information about plaque echogenicity, the exact sensitivity and accuracy to identify necrotic cores is unclear, and the resolution may not be sufficient to properly quantify this important feature of plaque vulnerability.⁸⁸

3. Plaque inflammation: Detection of macrophages within the fibrous cap requires a resolution within 10 to 20 µm. Considering that the IVUS resolution is 10 to 20 times lower, it is impossible for IVUS to detect macrophages in atherosclerotic plaques.

4. Degree of positive remodeling: IVUS is an excellent tool for detecting remodeling, a major feature of plaque vulnerability, in contrast to detection of other features of plaque composition.⁹⁸ IVUS-derived arterial remodeling helped understand the actual paradox of lumen and plaque size in VP, as large plaques can appear as nonobstructive on angiography, leading to the realization that ruptured plaques are larger compared with nonruptured plaques.^37,90 No other imaging modality can show remodeling better than IVUS, so IVUS is considered the gold standard for this parameter in vivo.⁹⁹

5. Plaque neovascularization: IVUS has the potential to detect flow within the plaque and thus identify functional neovessels. While real-time IVUS is limited to evaluating plaque perfusion, recent developments with contrast agents have dramatically improved the quality of Doppler ultrasound. Intravascular injection of microbubbles (i.e., small encapsulated air or gas bubbles or albumin microspheres) can boost the Doppler signal from blood vessels. Microbubbles can help in visualizing flow in smaller vessels, even at the capillary level, as has been shown by contrast-enhanced echocardiography (CEE).¹⁰⁰ Direct visualization of atherosclerotic plaque microvessels using CEE was successfully done by Feinstein in carotid plaques; this has been also validated with histology in animal models and in humans.^101–103 In coronary arteries, IVUS-CEE has successfully identified plaque neovessels with spatio-temporal changes and enhancement–detection techniques (Fig. 59-16).¹⁰⁴ To improve resolution, an IVUS prototype using “harmonic” imaging, with transmission of ultrasound at 20 MHz (fundamental) and detection of contrast signals at 40 MHz (second harmonic), was developed with the aim to identify adventitial neovessels in rabbits models.^105,106 However, to date, vasa-vasorum detection of coronary atherosclerosis with IVUS imaging is still evolving for clinical use.

Figure 59-15 Intravascular ultrasound (IVUS) images of ruptured plaques, highlighting the fibrous cap and a large echolucent area under the cap suggestive of large necrotic cores (upper panel). Lipid area, cap thickness, and lipid percent area in ruptured (red) versus nonruptured (black) plaques (lower panel).

(Adapted with permission from Ge J, Chirillo F, Schwedtmann J, et al: Screening of ruptured plaques in patients with coronary artery disease by intravascular ultrasound, Heart 81:621–627, 1999.)

Figure 59-16 Differential intravascular ultrasound (IVUS) images to identify the vasa-vasorum, showing the subtracted post-injection signals from baseline signals. A, Black and white (signal intensity of Figure 59-1 A–C). B, Color-coded, panel A. C, Thresholded to show most significant areas of enhancement.

(Adapted with permission from Vavuranakis M, Kakadiaris IA, O’Malley SM, et al: Images in cardiovascular medicine. Detection of luminal-intimal border and coronary wall enhancement in intravascular ultrasound imaging after injection of microbubbles and simultaneous sonication with transthoracic echocardiography, Circulation 112:e1–e2, 2005.)

Virtual Histology

Considering the significant limitations of IVUS, Nair and Vince at Cleveland Clinic studied the ultrasound scattered reflection wave as a possible alternative to improve tissue characterization.¹⁰⁷ This backscattered reflection wave is received by the transducer, where it is converted into voltage. This voltage is known as backscattered radiofrequency (RF) data. Using a combination of previously identified spectral parameters of the backscattered ultrasound signal, a classification scheme was developed to construct an algorithm to test plaque composition ex vivo. Four major plaque components were tested, including fibrotic tissue (dark green), fibro-fatty tissue (yellow-green), calcific-necrotic core (red), and dense calcium (white). A color was assigned for each of these components and is displayed on the IVUS image (Fig. 59-17). The Movat-stained histologic images identified homogeneous regions representing each of the four plaque components (Fig. 59-18). The unit of analysis (also called box) was initially composed of 64 backscattered RF data samples in length (480 µm).¹⁰⁷ The algorithm developed was then validated ex vivo, with sensitivities and specificities between 79% and 93% for all four-plaque components.¹⁰⁷ The initial studies were performed in ex vivo human coronary specimens with a 30-MGz, 2.9F, mechanically rotating IVUS catheter (Boston Scientific Corp); the initial catheter approved by the U.S. Food and Drug Administration (FDA) was a 20-MHz (Eagle-Eye® Gold) device. Recently, the catheter was upgraded with a 45-MHz transducer that is currently available for clinical practice. In vivo validation studies have shown positive results.^108,109 Virtual histology gained significant attention with the PROSPECT Trial.¹⁵ As previously discussed in this review, VH-derived TCFA was associated with increased events (HR 3.35, P < 0.001). Most importantly, the highest-risk lesions (HR 11.05, P < 0.001) were a conglomerate of several features, including greater than 70% plaque burden, low MLA, and TCFA morphology. Therefore, isolating TCFA by using VH may not be enough to categorize a lesion as VP. The possibility of a PROSPECT II study randomizing patients with these lesions to aggressive medical therapy with or without stenting may be considered. However, the large number of patients needed and the increased costs may limit the ability to test this hypothesis.

Figure 59-17 Color-coded reproduction of intravascular ultrasound Virtual Histology (IVUS-VH) plaque composition displayed in vivo in the cath lab.

Figure 59-18 Definitions of the different plaque components obtained with intravascular ultrasound Virtual Histology (IVUS-VH).

(Reproduced with permission from Vince, DG, personal communication).

The individual ability of VH to identify TCFAs can be described as follows:

1. Fibrous cap thickness: IVUS-VH is limited in cap thickness evaluation. This was elegantly addressed in the initial publication by Nair and Vince, in which they comment on the limitations: “The window size currently applied for selection of regions of interest and eventual tissue map reconstructions is 480 microns in the radial direction. Therefore, detection of thin fibrous caps (≤65 microns below the resolution of IVUS) would be compromised, restricting the detection of vulnerable atheromas”.¹⁰⁷ As a result, lesions with fibrous cap thickness greater than 65 µm will be incorrectly classified as TCFAs and perhaps overestimated. Therefore, with its axial resolution of 250 µm, it is insufficient to determine fibrous cap thickness, and investigators have proposed a classification of “IVUS-VH–derived TCFA”.¹¹⁰ This has been defined as a plaque with a rich necrotic core (>10%), without evident overlying fibrous tissue, and with a percent plaque volume of 40% seen on at least three consecutive images. Such features are reflective of histologic TCFAs that are more prone to rupture. As also defined by histology, IVUS-VH-derived TCFAs cluster around the proximal segments of the arteries, are more often associated with positively remodeling, and are more frequently found in patients with ACS compared with those with stable angina.^111,112 Sawada et al evaluated the ability of both IVUS-VH and OCT to detect TCFAs in the same coronary lesions. IVUS-VH was very effective in detecting the absence of TCFA. However, IVUS-VH only diagnosed half of the TCFAs compared with OCT.^99,113 One of the initial concerns about the IVUS-VH–derived TCFA was its accuracy to serve as a surrogate for VPs and the actual prognostic value that this specific finding could provide. This question was mostly answered by the PROSPECT trial (described above), which highlighted the value of IVUS-VH–derived TCFA. In a similar way, this concept was reinforced by a recent publication by Kubo et al, which addressed the natural history of IVUS-VH-TCFA in 99 patients undergoing PCI, who were followed up for 12 months with serial evaluations of coronary vasculature with the use of IVUS-VH. They found that 75% of the 20 VH-TCFAs healed by either becoming TCFAs (65% of total) or by becoming fibrotic (10% of total), whereas only 25% of those remained unchanged. They also reported the occurrence of 12 new VH-TCFAs that developed during the follow-up period: 6 of them from pathologic intimal thickening and 6 from TCFAs identified at baseline. Notably, no acute coronary events occurred during the follow-up period from any of the initially identified or the newly formed VH-TCFAs.¹¹⁴

2. Necrotic core area: IVUS-VH was initially developed to identify calcific necrotic cores. However, the incidence and degree of calcification in necrotic cores is variable, and therefore necrotic cores without calcification may not be properly identified.¹¹⁵ Most importantly, the majority of advanced atherosclerotic lesions will display a certain degree of necrotic core. As a result, when validating necrotic core using IVUS-VH, not only the presence or absence of the necrotic core is important (sensitivity/specificity) but also the area.¹¹⁶ IVUS-VH routinely reports necrotic core area (mm²) and percent of total plaque area. A recent substudy from the PROSPECT cohort reported that plaques with large areas of attenuation by grayscale analysis are associated with large amount of VH-IVUS necrotic core and are markers of the presence of fibroatheromas (VH-TCFA or thick-capped fibroatheromas [VH-ThFA]).¹¹⁷ However, despite the cumulative clinical evidence and proposed clinical applications of IVUS-VH, proper validation of these areas with histology using linear regression analysis have yet to be performed in humans and probably never will be. Limited studies have attempted to identify the correlations between necrotic core areas by using IVUS-VH with histology; these were done in a porcine model, demonstrating that IVUS-VH is unreliable in terms of necrotic core assessment.^118,119 As discussed earlier in this review, multiple pathologic studies have established the concept that necrotic core areas from patients with ACS are larger compared with necrotic core areas from patients with chronic stable angina.^4,30 Conversely, fibrous plaque areas (collagen) have been found to be significantly smaller in patients with ACS.³⁸ Surmely et al concluded that data on plaque composition obtained by IVUS-VH contradict previously published histopathologic data.¹²⁰ However, Rodriguez-Granillo et al have identified larger necrotic areas in ruptured plaques and in nonculprit lesions from patients with ACS.^121,122 Validation with histology in carotid atherosclerosis shows diagnostic accuracy of 99.4% for TCFA, 96.1% for calcified TCFA, 85.9% for fibroatheroma, 85.5% for fibrocalcific atheroma, 83.4% for pathologic intimal thickening, and 72.4% for calcified fibroatheroma.¹²³ In a small prospective cohort, IVUS-VH analysis was correlated with coronary atherectomy specimens. The correlation coefficients ranged from 0.90 to 0.97 for plaque components.¹²⁴

3. Plaque inflammation: Detection of macrophages within the fibrous cap requires a resolution within 10 to 20 µm. Considering that IVUS-VH resolution is at least 10 to 20 times lower, it is impossible for VH to detect macrophages in vivo.

4. Degree of positive remodeling: IVUS is an excellent tool to detect remodeling, and IVUS-VH should preserve this advantage. As reviewed earlier, positive remodeling is related to large necrotic core areas, more frequently seen in patients with ACS. Conversely, plaques with negative or constrictive remodeling are associated with smaller necrotic core areas, usually seen in patients with chronic stable angina. Recent studies evaluated IVUS-VH–derived necrotic core areas in plaques with positive and negative remodeling and found smaller necrotic core areas in positively remodeled plaques.⁹⁰ Other investigators have confirmed these data.¹²⁵ However, some authors have demonstrated contradictory data with strong correlations between large IVUS-VH–derived necrotic core areas and positive remodeling.^117,126 Finally, vasa-vasorum neovascularization and IPH require highly sophisticated technology and cannot be identified with IVUS-VH. On the basis of the data presented above, it is difficult to reconcile these contradictory findings and, therefore, impossible to elucidate the real clinical value of IVUS-VH–derived plaque composition in clinical practice.

Elastography and Palpography

The mechanical stress induced by changes in blood pressure induces deformation on the fibrous cap (strain).¹²⁷ This stress–strain relationship in coronary lesions can be identified by using another IVUS-derived technique called elastography or palpography, which displays data in a color-coded scale (see Figure 59-18).¹²⁸ Purple indicates a low strain (hard, stiff), whereas yellow indicates a region of high strain (soft, deformable).¹²⁹ Palpography has high sensitivity and specificity to detect VP, with a deformation of more than 2% reflecting increased macrophage infiltration, reduced smooth muscle cell, and low collagen content (Fig. 59-19).¹³⁰ The Rotterdam Classification (ROC) divides strain into four subclasses; the worst is ROC IV, with a deformation greater than 1.2%.¹²⁹ Palpography shows correlations with CRP, and ST elevation myocardial infarction (STEMI). Aggressive treatment using statins can reduce, over a period of 6 months, the intensity and frequency of these high-strain spots. Recent improvements using reconstructive compounding by motion artifact correction are promising.^131,132 Nonetheless, palpography cannot detect fibrous cap thickness, necrotic core, degree of remodeling, neovascularization, or IPH.

Figure 59-19 Vulnerable plaque marked in intravascular ultrasound (IVUS) (A), elastogram (B), macrophage staining (C), and collagen staining (D). In the elastogram, a vulnerable plaque is indicated by a high strain on the surface. In the corresponding histology, a high amount of macrophages (C) is visible with a thin cap (D) and a lipid pool (LP). See also box plots correlating the macrophage (upper right) and smooth muscle cell content (lower right) with the level of strain. High levels of strain are associated with significant increases in macrophage content and significant decreases in smooth muscle cell content. SMCs, smooth muscle cells.

(Adapted with permission from Schaar JA, De Korte CL, Mastik F, et al: Characterizing vulnerable plaque features with intravascular elastography, Circulation 108:2636–2641, 2003.)

Optical Coherence Tomography

OCT is a novel high-resolution intravascular imaging technique that is currently approved for clinical use. Of all of the invasive modalities, OCT provides the highest resolution (5 to 20 µm).¹³³ Excellent sensitivity and specificity, between 92% and 100%, have been documented for all components of TCFA.¹³³ Superb resolution allows for improved images (Fig. 59-20). Recent advances using optical frequency domain imaging (OFDI) allow for high-speed comprehensive imaging, scanning up to 5 cm with one single saline flush (Fig. 59-21).¹³⁴ OCT is currently being used in clinical practice, providing significant information for the identification of plaque rupture, fibrous cap erosion, intracoronary thrombus, and TCFA location.^135,136 Similarly, OCT can predict no-reflow phenomenon in patients with large lipid cores who undergo PCI during ACS.¹³⁷ A direct comparison of OCT with IVUS-VH to detect TCFAs is shown in Figure 59-22. With all these promising facts on OCT, the interventionalist needs objective information regarding the ability of OCT to detect TCFA, which can be summarized as follows:

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Volume and Outcome
2. Regional Centers of Excellence for the Care of Patients with Acute Ischemic Heart Disease
3. Role of Adjunct Devices: Cutting Balloon, Laser, Ultrasound, and Atherectomy
4. Intervention for Non-ST-Segment Elevation Acute Coronary Syndromes

Stay updated, free articles. Join our Telegram channel

Join