Applied Radiology

Dr. Bradley is the Chairman of and a Professor in the Department of Radiology, University of California San Diego Healthcare, San Diego, CA.

Atherosclerosis is a topic very close to my heart. It is fitting that this issue of Applied Radiology addresses cardiac imaging, given the tremendous strides made over the past decade, particularly with magnetic resonance imaging (MRI) and computed tomography (CT).

Much of heart disease is acquired as a result of atherosclerosis. For the past 10 years, there has been a fundamental shift in the way we image atherosclerotic disease, in the coronary arteries as well as in the carotids and any other arteries where flow may be limited by an occlusive plaque.

For the greater part of the last century, imaging was focused on quantitating the degree of stenosis. A 70% reduction in vascular diameter, ie, 50% reduction in cross-sectional area, was considered "flow limiting." This 70% figure was given additional weight by the 1991 NASCET trial, which showed that at least a 70% stenosis of the carotid artery was necessary to justify carotid endarterectomy in symptomatic patients. ¹ The NASCET trial used two numbers from a two-view cut-film carotid angiogram: the carotid diameter on the most narrowed view of the residual lumen divided by the normal diameter of the internal carotid artery just distal to the carotid bulb. Although the NASCET trial was based on cut-film angiography, the same number has come to be used with more modern digital imaging modalities, such as ultrasound, CT angiography, and MR angiography. Given the oft-noted point that atherosclerotic plaque can be eccentric and that two views from a routine angiography study may not estimate degree of stenosis accurately, many now favor the use of a cross-sectional area now that the newer digital modalities are being acquired in the axial plane. Thus, instead of a 70% ratio of diameters, stenosis would be measured in square millimeters. This should improve quantitation of stenosis; however, degree of stenosis may not be the most important factor in predicting stroke (as in NASCET) or myocardial infarction.

Over the last decade, attention has been increasingly focused on the vulnerability of the plaque itself, rather than on the degree of stenosis. While chronic narrowing of the coronary arteries may lead to symptoms (ie, angina with exercise) or findings (eg, ST depression/elevation on a stress EKG), such findings, at most, may lead to additional confirmatory testing up to an angiogram on an elective basis.

Of much greater concern than chronic narrowing is acute rupture of the atherosclerotic plaque. This forms a thrombogenic surface, which leads to acute occlusion of the vessel locally by clot or to an embolus being sent downstream. The former is more common in the coronary arteries as a cause of myocardial infarction. The latter is more common in the carotid arteries as a cause of cerebral infarction. Thus, current research in atherosclerosis is less concerned with degree of stenosis than with characterization of plaque stability. Identification of vulnerable plaque, ie, that which is more likely to rupture acutely, should be the goal of imaging, as such lesions are more often the cause of fatal cerebral or myocardial infarction.

The mature atherosclerotic plaque can be characterized simply as a fibrous cap covering a lipid core. The thicker the fibrous cap, the more stable the plaque and the less likely it is to rupture. Vulnerable plaques have thin fibrous caps. In this context, "thin" is on the order of 50 to 100 microns (0.05 to 0.10 mm), which is a much higher spatial resolution than is currently available with either CT or MRI. Stable plaques tend to be calcified on CT. This poses an interesting conundrum: high coronary calcium scores may actually represent more stable disease than low scores do. As expected, large lipid cores are low density on CT. On MRI, a lipid core is bright on T2-weighted images and the fibrous cap enhances with gadolinium. It is hoped that using local coil technology and k-space re-registration techniques, such as periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) MRI, ² to get rid of motion artifact, the spatial resolution will improve to the point at which the thickness of the fibrous cap can be evaluated and disruptions can be visualized. While plaque characterization will most likely be performed initially in the carotid arteries (which could serve as a surrogate for coronary disease), we should expect the technology to transfer fairly quickly to the heart.