June 30, 2008

Seeking a Better Scorecard
By Beth W. Orenstein
Radiology Today
Vol. 9 No. 13 P. 12

Plaque distribution, not just the amount of plaque, may improve calcium scoring.

Each year, more than 700,000 Americans die of heart disease, according to the Centers for Disease Control and Prevention. Coronary artery disease (CAD), a buildup of calcific plaque in the coronary arteries leading to the heart, is the most common form of heart disease and the leading cause of death and disability for both men and women in the United States.

Early detection of CAD in patients, especially those who are asymptomatic, can lead to preventive treatments; thus helping reduce the likelihood of them having a heart attack or other cardiac event and dying.

Since the late 1990s and the introduction of fast, multidetector CT scans, doctors have been able to image the heart to determine the amount of calcium in a patient’s coronary arteries. The amount of calcium is presented as a coronary artery calcium score. Scores range from 0 to more than 400, with 0 being very low risk and more than 400 being high risk.

Originally, physicians debated the benefits of obtaining a calcium score. Some swore by the noninvasive imaging procedure, while others considered the scans, which cost $400 to $500, a waste of time and money and a risk to the credibility of the profession performing them. In July 2000, however, a writing group for the American Heart Association recommended obtaining calcium scores for people who, according to traditional measures, had an intermediate risk for CAD. The statement noted that a positive calcium score in such individuals could tip the scale strongly in favor of aggressively treating risk factors, possibly with aspirin and statin therapy.

Intermediate Risk
Since then, a consensus that calcium scanning is useful for evaluating those at intermediate risk (a 10% to 20% 10-year risk of CAD) has been growing, says Robert Detrano, MD, PhD, of the Harbor-UCLA Medical Center Research and Education Institute in Torrance, Calif. Detrano says the consensus has been coalescing even more since the publication of a paper in The Journal of the American Medical Association two years ago by Greenland et al and one in The New England Journal of Medicine in March by himself and colleagues.

“The primary issue with the coronary artery calcium [CAC] score is that it is an indirect, statistical assessment of risk of coronary artery disease,” says David A. Bluemke, MD, PhD, a professor of radiology and medicine and the clinical director of MRI at Johns Hopkins University School of Medicine in Baltimore. The current standard of coronary calcium measurement tells patients and their physicians the amount of calcific plaque present in the arteries, but not its spatial distribution.

In the June issue of Radiology, Detrano, Bluemke, and their colleagues reported on a method of calculating calcium scores that takes into account not only the amount of calcified plaque buildup in the coronary arteries but also its spatial distribution. Their study is titled, “Coronary Calcium Coverage Score: Determination, Correlates, and Predictive Accuracy in the Multi-Ethnic Study of Atherosclerosis.”

Location, Location, Location
Elizabeth Brown, ScD, a research assistant professor in the department of biostatistics at the University of Washington in Seattle and the study’s lead author, says that knowing the location of calcium in the arteries can be particularly important in estimating a patient’s potential risk.

“Currently, physicians only see the result in terms of an overall score designed to measure the amount of calcified plaque,” she says. “This new approach will provide physicians with a measure of the proportion of the arteries affected.”

 The additional information makes the coronary calcium coverage score (CCCS) a better predictor of future cardiac events, Brown says. The new score comes from the method of reading unenhanced cardiac CT scans and not from any difference in the way the scans are obtained or enhanced.

Adjusting Software
“The coronary calcium coverage score could become routine, although it would require manufacturers programming this method into their scanners,” Bluemke says. “It could be used with any CT scanner.”

“It should be more accurate with the modern multislice scanners,” adds Detrano, who also believes that commercial software manufacturers should be able to write the software so that the CCCS can be calculated as easily as the traditional calcium score.

Detrano came up with the idea for summarizing the spatial distribution of CAC after looking at thousands of scans. “I wondered if the distribution of calcium over the vessel would make a difference,” he says. “I wondered whether calcium which was diffusely spread over a vessel might suggest more diffuse atherosclerosis when compared to clumped calcium in one place.”

Brown came up with the CCCS as the summary measure and developed the algorithm to calculate it. To develop the score, the researchers conducted a study using data from the Multi-Ethnic Study of Atherosclerosis (MESA). MESA was initiated in July 2000 to investigate the prevalence and progression of subclinical cardiovascular disease in individuals without known cardiovascular disease. MESA included 6,814 men and women between the ages of 45 and 84 from six locations in the United States, including Chicago, Los Angeles, and New York.

Study Details
The researchers compared CT image data for 3,252 participants with calcific plaque with data collected from 3,416 patients without calcific plaque. Because of insufficient CT image data, 146 MESA participants had to be excluded from the analysis. The scans were read centrally at the Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center to identify and quantify coronary artery calcification.

The cardiologist or radiologist reading the scans identified the coronary arteries—left main, left anterior descending, left circumflex, as well as the right coronary. Once trained, it took the readers about two minutes per case to identify these waypoints. The waypoints were then used to determine the three dimensional course of the arteries.
“We determined the presence of calcific plaque in short intervals along the arteries, which we called subdivision,” Brown says.

An absolute subdivision was defined as a 5-millimeter linear segment of the arterial trajectory and a relative subdivision as 5% of the total length of the artery. To calculate the patient’s calcium coverage score, the researchers divided the number of absolute subdivisions in which cardiac plaque was present by the total number of absolute subdivisions in the coronary arteries and then multiplied the quotient by 100. The traditional Agatston method and mass score method of calculating calcium scores also were calculated for each subdivision. (Details of how the lengths of arteries were calculated and the subdivisions identified can be found at http://radiology.rsnajnls.org/cgi/content/full/2473071469/DC1.)

 The patients were then followed for a median period of 41 months to determine if there was a relationship between the distribution of the calcium shown in their CT images and the likelihood of heart attack or other cardiac event. The follow-up was conducted by telephone, when each patient or his or her family was contacted and asked whether the patient had been admitted to the hospital, undergone outpatient treatments, or  died. Hospital and outpatient records and death certificates were used to verify the patient’s or family’s reports, Brown says.

Better Scoring
The study found that the CCCS was a better predictor of future cardiac events than the currently used Agatston and mass scores, which gauge only the amount of calcium present. A twofold increase in the CCCS indicated a 34% increase in the risk of heart attack or other serious cardiac event and a 52% increase in the risk of any cardiac event.

 The study also found that diabetes, hypertension, and dyslipidemia (abnormal concentrations of lipids [fats] or lipoproteins in the blood) were highly associated with the CCCS. On average, compared with patients without diabetes, patients with diabetes had 44% more of their coronary arteries affected by plaque.

In their study, the researchers point out that the calcium coverage score has some limitations. For example, the CCCS depends on an accurate tracing of the arteries along their entire length, which may prolong the reading time. “Also, since arterial tracing adds an additional element of variability to the calcium scoring process, in our study, the CCCS was slightly less reproducible compared with the Agatston score or the calcium score,” Brown says.

Brown also says that while the researchers demonstrated that overall cardiovascular events were better predicted with the CCCS than with the Agatston or mass score, they observed no differences in the prediction of hard events (myocardial infarction or death). “This may have been due to a true lack of association or to the smaller number of hard events,” Brown says. “We were limited by the length of the follow-up and the number of adjudicated events we had in the MESA for participants at the time of the analysis.”
The researchers plan to explore their findings further with longer follow-ups, Brown says. They also hope to explore the possibility of providing more prognostic information by combining Agatston and mass scores with the CCCS.

“Our current study is following more than 5,000 individuals for changes in coronary artery calcium,” Bluemke says.

Another issue, Brown says, is people who have zero calcium scores or no calcium indicated by a CT scan. “We are thinking about whether or not there is information we are missing in the people who have zero Agatston scores. That raises the question of whether we need to change the CT scan reading protocol that we use,” she says.

The researchers believe that the CCCS will prove useful in the clinical setting by helping physicians more accurately classify patients according to risk than conventional calcium scoring methods. That ability should allow physicians to improve and individualize treatment strategies. “This could definitely guide their decisions for treatment or for monitoring,” Brown says.

Detrano believes that if the CCCS proves to be a more accurate predictor of coronary events, as their paper shows, “then it should supplant the standard Agatston coronary score in clinical calcium scoring.”

— Beth W. Orenstein is a freelance writer from Northampton, Pa., and a regular contributor to Radiology Today.