US researchers found that using a score based on the amount of calcium present in coronary arteries as well as the traditional factors taken into account when assessing heart disease risk improved the prediction of risk and put more individuals in the most extreme risk category.

You can read about the study, conducted by first author Dr Tamar S. Polonsky, of the Northwestern University Feinberg School of Medicine, Chicago, and colleagues, in the 28 April online issue of JAMA, Journal of the American Medical Association.

Arterial calcified plaque results when fat and other substances build up under the inner layer of the artery wall. This material can calcify, a sign of atherosclerosis, a disease of the vessel wall, also called coronary artery disease, CAD.

People with CAD have an increased risk for heart attacks: as the plaque builds up, the arteries get narrower and narrower and can even stop blood flowing to the heart. The result is chest pain or angina, or a heart attack.

To determine a coronary artery calcium score (CACS) the radiographer takes a CT or CAT (computed tomography) scan and assesses how much calcium buildup is present in the plaque on the walls of the arteries of the heart.

The authors said in their background information that while large prospective studies, that is the kind that follows a population over a period of time, have suggested a link between the CACS score and risk of cardiovascular events, what remains uncertain is how such a measure affects the classification of coronary heart disease (CHD) risk over and above the traditional factors.

For this study, Polonsky and colleagues decided to investigate whether adding CACS to a prediction model based on traditional risk factors improved the classification of CHD risk.

They recruited 6,814 participants as part of the Multi-Ethnic Study of Atherosclerosis (MESA), a population-based cohort without known cardiovascular disease, who underwent CT scans and assessments for traditional CHD risk factors on enrollment, which started in July 2000. The participants were then followed up through May 2008.

The researchers then prepared two “hazard models” to assess CHD risk.

Model 1 (the traditional model) categorized five year CHD risk as 0 to less than 3 per cent (low risk), 3 to less than 10 per cent (intermediate risk), and 10 per cent and over (high risk), and accounted for the usual risk factors such as age, race, ethnicity, gender, tobacco use, blood pressure drug use, systolic blood pressure, and cholesterol levels (total and HDL).

Model 2 was just like Model 1 except that in addition to the usual risk factors, it also included CACS.

The researchers then compared the results from the two models and calculated a “net reclassification improvement”.

The results showed that:

  • 209 CHD events occurred in a final group of 5,878 participants over a median follow up of 5.8 years.
  • 122 of these events were major (eg heart attack, death from CHD, rescuscitated cardiac arrest).
  • Model 2 (including CACS) was significantly better at predicting risk than Model 1 (traditional, excluding CACS).
  • Model 1 (correctly) classified 69 per cent of the participants in the highest or lowest risk categories: Model 2 classified 77 per cent.
  • Adding CACS to the risk assessment meant an extra 23 per cent of participants who experienced CHD events were moved into the high risk category (ie the traditional model had not spotted them).
  • Adding CACS also changed the risk profile at the other end: an extra 13 per cent of participants who did not experience CHD events were moved to the lower risk category (ie the traditional model had assessed them as having a higher risk even though they did not eventually experience them).
  • Among participants of intermediate risk under Model 1, 16 per cent were moved into high risk and 39 per cent were moved into low risk under Model 2.

The authors wrote these findings show that:

“When CACS is added to traditional risk factors, it results in a significant improvement in the classification of risk for the prediction of CHD events in an asymptomatic population-based sample of men and women drawn from 4 US racial/ethnic groups.”

“Incorporation of an individual’s CACS leads to a more refined estimation of future risk of CHD events than traditional risk factors alone,” they added, explaining that the results are good enough to warrant moving onto the next stage, that is to assess the use of CACS on clinical outcomes.

In an accompanying editorial, Dr John P A Ioannidis, of the University of Ioannina School of Medicine in Greece, and Dr Ioanna Tzoulaki, of the Imperial College of Medicine in London, UK, agreed that CACS needs to be assessed via a randomized trial before we can say whether it should be routinely included in CHD risk assessments.

They explained that even with this exemplary study we don’t have enough evidence that the apparently more accurate risk profiles that result from adding CACS to the mix “can actually aid clinicians in better treating patients or improving their clinical outcomes”.

Also, costs and harms could be major hurdles, they wrote.

“Computed tomography costs $200 to $600 and routine implementation at the population level can be very expensive,” commented Ioannidis and Tzoulaki, adding that:

“The lifetime excess cancer risk due to radiation exposure from a single examination at age 40 years is 9 cancers per 100,000 men and 28 cancers per 100,000 women. This risk should be taken into account in formal risk-benefit analyses.”

They concluded that CACS looks promising, but there is still a lot of work to do before it can be recommended for routine use.

“Coronary Artery Calcium Score and Risk Classification for Coronary Heart Disease Prediction”
Tamar S. Polonsky; Robyn L. McClelland; Neal W. Jorgensen; Diane E. Bild; Gregory L. Burke; Alan D. Guerci; Philip Greenland.
JAMA, Vol. 303 No. 16, pages 1610-1616, published online April 28, 2010.

Sources: JAMA and Archives Journals,

Written by: Catharine Paddock, PhD