WASHINGTON — Although the evaluation of noncalcified plaques with CT angiography currently is possible, there are still several obstacles to overcome before the technique is clinically useful, said Dr. Stephan Achenbach, a professor of medicine at the University of Erlangen in Germany, at the annual meeting of the Society of Cardiovascular Computed Tomography.
“As technology progresses, image quality gets better and better and our ability to visualize plaques gets better and better,” said Dr. Achenbach, who also is the past president of the Society of Cardiovascular Computed Tomography.
One of the criticisms of invasive angiography is that only the lumen can be seen, not the plaque itself. Contrast CT does allow for the visualization of noncalcified plaque in the coronary arteries.
“CT is able to show it—the slight lumen reduction and also noncalcified and partly calcified plaque,” said Dr. Achenbach. In fact, with high resolution, CT cross sections are similar to intravascular ultrasound (IVUS) for evaluating plaque composition. However, “this is indeed a tremendously difficult task to visualize these plaques by CT.”
One reason for this is that the plaques in the coronary arteries are extremely small. The spatial resolution of CT under optimal conditions is approximately 0.4 mm. “So we're trying to visualize something that is half a millimeter thick with a spatial resolution of 0.4 mm,” said Dr. Achenbach.
Another problem is contrast. Calcium is easy to see on CT because it has a very high contrast with the surrounding tissue. However, the contrast between noncalcified plaque and the surrounding tissue is much less. “So we have to deal with structures that give us very little contrast on CT,” said Dr. Achenbach.
Yet another challenge in using CT to visualize noncalcified plaque is the high level of image noise. Simply put, noise is the difference between real-world signals and an ideal signal and may be caused by a wide range of sources, such as variations in detector sensitivity and environmental variations.
The combination of low contrast between noncalcified plaque and surrounding structures and high image noise makes it very difficult to tell whether noncalcified plaque is present.
“Motion is another problem,” said Dr. Achenbach. Patient motion can produce blurring and dark areas that can be mistaken for noncalcified plaques.
The presence of calcium also can cause problems on occasion. Even under ideal conditions, calcium can appear to be surrounded by a dark rim. On inspection it can be unclear if this is really noncalcified plaque or not. “In the presence of calcium, we have tremendous difficulty ruling in or ruling out the presence of noncalcified plaque,” said Dr. Achenbach.
Despite these problems, “if image quality is really good, we continue to be amazed by how accurately and clearly CT angiography can visualize these noncalcified plaques,” he said.
Unfortunately, there are few data on the accuracy of CTA in identifying noncalcified plaque. In the studies that have been performed, researchers compared multidetector CT (usually 16-slice) with IVUS in patients without coronary stenoses. The accuracy of multidetector CT (MDCT) in the detection of nonstenotic plaque ranged from 80% to 90%. However, many of the plaques identified were at least partly calcified. The accuracy of MDCT detection of purely noncalcified plaque was closer to 50%.
Beyond plaque characterization, can CT quantify? “Theoretically, you can measure the size of the plaque and you can measure the size of the lumen,” said Dr. Achenbach.
In general, the correlation between MDCT and IVUS with regard to measuring plaque area and volume is good. “But you're not really able to very accurately measure a single coronary plaque,” said Dr. Achenbach.
In addition, interobserver variability is a problem when it comes to quantifying plaque using MDCT. In a study performed at his own institution, interobserver variability ranged from 19% to 32%, depending on the vessel.
The ultimate goal, though, is to be able to identify vulnerable plaques—those at the greatest risk of rupture.
Histologically, the markers of plaque vulnerability include a thin fibrous cap, a necrotic core, and macrophage infiltration. These markers are very hard to see on CT.
“However, there are some other measures that are also tied to plaque vulnerability and the [risk of it causing] an event in the future that might be amenable to CT,” said Dr. Achenbach.
The remodeling index—defined as the lesion external elastic membrane (EEM) area divided by the EEM area for a reference vessel—is a potential measure. “It has been shown that the remodeling index in CT correlates quite well with the remodeling index in IVUS,” said Dr. Achenbach. Strong positive remodeling has been associated with greater risk of plaque rupture.