INTRODUCTION

The development of intravascular ultrasound (IVUS) has been a major development in the invasive imaging of coronary arteries. Clinical studies in intravascular ultrasound began in 1989 with the development of catheters initially in the 5–6 French gauge (F) range with the most recent catheter size miniaturised to 2.6 F.¹ Intravascular ultrasound has permitted not only a greater understanding of plaque morphology and its response to interventional procedures but has provided accurate online quantitative information regarding lumen size and residual plaque load, an important predictor of restenosis. The presence of disease not only at the site of focal stenosis but also in reference segments believed by angiography to be free of disease has modified interventional practice significantly. With continued improvement in image quality through increasing ultrasound frequency from 30 MHz through 40 MHz currently available and ultimately 50 MHz, the morphology of plaque and the offline ability to characterise plaque will provide additional information in the management of atherosclerotic disease. It is likely that continued technical developments will enhance and define the role of intravascular ultrasound in coronary interventional practice.

BASIC INTERPRETATION

An appreciation of the coronary anatomy and its relationship to structures around it is important to the accurate interpretation of intravascular ultrasound images. These spatial relations are best appreciated at the time of slow catheter pull-back from distal to proximal vessel done using an automated pull-back device. Advances in image quality and improved tissue penetration has allowed the use of epivascular structures in addition to side branches as reference points for tomographic and axial orientation (Fig. 9.1).

Figure 9.1 The left panel is an image in the left coronary artery. Marginal veins cross around the artery in the horseshoe pattern, often associated with and opposite to the branch points of the right ventricular marginal arteries. Recurrent atrial branches generally emerge on the opposite side of the artery to the marginal branches.

The middle panel is an image of the proximal LAD artery. The anterior interventricular vein (AIV) lies to the left of the proximal artery in the majority of people (85%), with the first two diagonal branches on the same side. In one third of the people, the interventricular vein branches into two after second diagonal to lie either side of the LAD more distally. Pericardium may be seen as a bright reflection on the anterior side of the artery. Diagonal branches emerge on the same side as the AIV. Septal branches emerge from the LAD in a perpendicular fashion to myocardium, on the opposite side of the pericardium.

The right image is of the proximal left circumflex artery. Distally, the CFX is accompanied by the posterior left ventricular vein whereas in its proximal section, it is crossed and shadowed superiorly by the great cardiac vein. Recurrent atrial branches emerge from the circumflex artery towards the great cardiac vein in the opposite direction to obtuse marginal branches.

There are three concentric layers in the epicardial coronary arterial wall demonstrable at histology and seen by intravascular ultrasound imaging. The intima is the innermost layer and consists of endothelial cells and the subendothelial layer of smooth muscle cells and fibroblasts in a connective tissue matrix. The overall thickness of this layer can be just a few cells thick in childhood expanding to 150–200 *gmm in the adult. Beneath this, there is the internal elastic lamina which is intact in the normal state, consisting of fenestrated elastic fibres with a thickness less than 25 *gmm.

The media consists of multiple layers of smooth muscle cells arranged helically and circumferentially around the lumen of the artery, woven through a matrix of elastic fibers and collagen. The coronary arteries are less elastic than other similar sized arteries and thus resemble a transition towards more muscular peripheral arteries. The normal medial thickness ranges from 125–350 *gmm (mean 200 *gmm) although in the presence of plaque, the medial thickness may be considerably thinner, approximately 100 *gmm,² or completely involuted and replaced by plaque in severe disease. The external elastic lamina encircles the medial layer. It is composed of elastin but is thinner and more fenestrated than the internal lamina, and is not more than 20 *gmm in thickness.

The adventitia is essentially fibrous tissue, i.e. collagen (type III) and elastin, with the collagen orientated longitudinally in general, and to a lesser tissue density than media. It is a layer that is surrounded by the vaso vasorum, nerves and lymphatic vessels. The adventitia can extend from 300 to 500 *gmm in diameter beyond which it is considered perivascular stroma and epicardial fat.

The appearance of the three-layered appearance by intravascular ultrasound occurs due to the acoustic impedance between adjacent structures. For example, the lumen and intima are usually well-delineated due to the large acoustic impedance between fluid and tissue. The three-layered appearance of the vessel wall is dependent on the intima being of sufficient size to be identified with the resolution of the current generation of ultrasound transducers and in the presence of a sufficient acoustic interface between media and adventitia.³ At a frequency of 30 MHz, the threshold of intimal thickening required to resolve a definite intimal layer is approximately 160 *gmm. Previous work has shown that there is a progressive increase in the thickness of the intimal layer with increasing age.⁴ In an autopsy study done to evaluate the relationship between ultrasound images and tissue histology in 16 intact hearts from subjects ranging in age from 13 to 55 years with no history of coronary artery disease, segments with a three-layered appearance had a significantly greater intimal thickness (243 *b+ 105 *gmm) than non-layered segments (112 *b+ 55 *gmm) with a threshold between the two of 178 *gmm.⁴ As this threshold is crossed in males over the age of 30 years, it is apparent that histologically normal arteries will only have a two-layered appearance in the rare patients younger than this undergoing ultrasound examination.

The media appears as a thin middle layer by intravascular ultrasound and is often referred to as the sonolucent zone as it is less echodense than the intima or adventitia due to a lesser collagen content. Intravascular ultrasound imaging was performed in vitro on 6 histologically normal and 104 minimally diseased arteries in patients aged 13 to 83 years to test the hypothesis that normal coronary arteries produce a three-layer image that corresponds to the histologic layers of intima, media and adventitia.⁵ The results showed a very strong correlation between area of the echolucent ultrasound layer with the media and the inner echogenic layer with intimal area. In addition, a three-layered appearance was consistently seen when the internal elastic membrane was present with or without intimal hyperplasia. If the internal elastic membrane was absent, a three-layer appearance was still seen if the collagen content of the media was low. However, a two-layer appearance was observed when there was absence of the internal elastic membrane as well as a high collagen content of the media.⁵ In addition, the relative composition of the intimal layer also determined the ability to discriminate the three vessel wall layers. Thus, over a given coronary artery segment, the three-layered appearance may alternate with a two-layered appearance due to the relative content of elastin and collagen. However, for the purposes of quantitation, the acoustic impedance between the combination of adventitia and external elastic lamina with the intima permits accurate measurement of plaque and vessel area. In the left main stem and at the proximal part of the right coronary artery, the three-layered appearance may be lost due to the increase in elastin content in transition from the highly elastic aortic root.

ATHEROSCLEROTIC DISEASE

Intravascular ultrasound is the current imaging technology of choice for studying the morphology of atherosclerotic plaque in vivo and continues to have an important diagnostic role (Table 9.1). Early studies used a large 8 F catheter to correlate ultrasound appearances with histological findings in arteries collected at the time of autopsy.⁶ The arteries studied were a combination of elastic, transitional (musculo-elastic) and muscular types with respect to media/adventitia appearance. The results confirmed that a highly accurate measurement of luminal area was achieved comparing ultrasound to direct measurement of perfused isolated arteries. A distinct interface between media and adventitia was obtained only where there was a significant difference in the acoustic qualities of the two layers (namely, loose collagen in the adventitia of elastic arteries or where there was a minimal smooth muscle cell component in the adventitia of transitional/muscular arteries). The interface between plaque and media was only apparent where there was a dense internal elastic lamina or a significant amount of necrotic material in the plaque.

TABLE 9.1 DIAGNOSTIC ROLE OF IVUS

The earliest accumulations of atherosclerotic plaque consist of crescentic intimal thickening of an intermediate echointensity. A common site for initial and increased plaque accumulation occurs at branch points and bifurcations due to the shear stress effect of blood flow. Transplant vasculopathy is a good model for the early development of coronary artery disease as these patients undergo ultrasound studies at angiographic follow-up early after transplantation. In one study, intravascular ultrasound was used to study epicardial arteries in 25 recently transplanted hearts from young donors (mean age 28 years).⁷ In this unique study group, all donors aged under 25 years had a homogeneous non-layered vessel wall. Another group of donors of mean age 32 years manifested a three-layered appearance. In five hearts, significant eccentric intimal thickening >500 *gmm was shown in donors with risk factors for coronary disease, implying early coronary disease in the presence of angiographically normal arteries. Subsequent work by the same group in a larger group of transplant recipients over a period of long-term follow-up has shown that all 60 hearts had variable degree of concentric intimal thickening after one year, 42 of whom had normal coronary arteries at angiography.⁸

Occasionally, image interpretation can be obscured in abnormal vessels due to incorrect assumptions of the borders of the vessel layers traced manually. This can occur because the internal elastic lamina may not be a separate layer in the presence of plaque. Media may also appear unusually thick due to attenuation of the ultrasound beam passing through intimal plaque. By contrast, the media layer may also be thinner than expected due to the spread of the signal from an area of high reflectivity (plaque) to a low one (media). For these reasons, the outer border of the plaque is usually defined as at the media/adventitia border (the external elastic lamina) which is believed to be a fair assumption given the relative contribution of media to the plaque area. It is implicit in image interpretation that frames are selected with the best image quality and done so by an experienced operator.

With a greater accumulation of plaque, there is a greater complexity to the plaque which may be differentiated by broad ultrasound criteria. A fibrous plaque has an echodensity intermediate between less echodense media or lipid and more echodense calcification. Thus by comparing the brightness of the tissue in question to that of the adventitia, a relative grading of the plaque may be obtained. Such fibrous plaques with similar brightness to adventitia may then be described as hard or soft (with respect to the grey scale) depending on the presence or absence of shadowing behind the plaque (Fig. 9.2). Fatty plaques are significantly more echolucent and when large may be appreciated as lipid pools. However, because shadowing in relation to a fibrous plaque may be misinterpreted as a lipid collection, there is a tendency to broadly classify plaques containing lipid as fibrofatty in nature.

Figure 9.2 These three panels demonstrate different plaque morphologies. The left panel shows a concentric mild fatty plaque which is less echodense then adventita. The middle panel shows an extensive fibrofatty plaque with echodensity similar to adventitia. The right panel is a more echodense eccentric plaque with an area of acoustic shadowing between 5 and 8 o’clock due to the presence of dense fibrous plaque.

Calcification is commonly seen by intravascular ultrasound as a bright echo with shadowing behind often associated with reverberation artifact in the area of the shadow due to oscillation of the ultrasound beam between calcium and transducer (Fig. 9.3). Calcium may be seen in relatively small plaque accumulations indicating the age of the plaque or the site of previous plaque rupture and repair. In one series of patients undergoing balloon angioplasty, 82% of arterial segments exhibited small areas of calcium which were visible in only 8% of angiograms at the lesion site or in 155 of more proximal segments by fluoroscopy.⁹ In the GUIDE (Guidance by Ultrasound Imaging for Decision Endpoints) trial (Phase I), 70% of target lesions had areas of calcium by ultrasound, compared to 40% of angiograms.¹⁰ Calcification may be graded from absent (0) to severe (3+) by the extent of the arc subtended by a fibrocalcific matrix.¹¹