ACC 2009: CAC Scoring Helps Reclassify Intermediate-Risk Patients

March 29, 2009

To earn CME related to this news article, click here.

March 29, 2009 (Orlando, Florida)Coronary artery calcium (CAC) scoring can help predict who is likely to have an MI or cardiac death among those who are at intermediate risk of coronary events according to traditional risk-factor assessment, a new five-year study shows. Dr Raimund A Erbel (University Clinic Essen, Germany) reported the findings of the Heinz Nixdorf Risk Factors Evaluation of Coronary Calcium and Lifestyle (Recall) study during a late-breaking clinical-trials session here today.

The findings appeared to be more pertinent for men, who had almost a 10-fold greater risk of cardiac death or MI if they were in the highest CAC quartile compared with the lowest, compared with twice the risk for women.

Among those who were deemed intermediate risk, addition of the CAC score into the assessment showed that 14% should be reclassified as high risk and just over 60% could be shifted to low risk, Erbel told attendees.

"Our results demonstrate that prediction of coronary events can be improved when calcium scoring is performed, especially in persons in the intermediate-risk category," said Erbel. As coronary-calcium levels are detectable long before other symptoms of coronary disease, those at intermediate risk who register a high coronary-calcium score should be encouraged to change their lifestyles and may require risk-modifying medication, he said. Conversely, those at intermediate risk with a low coronary-calcium score have a more favorable prognosis, he noted. "Calcium scoring has now been validated and reached a place in preventive cardiology," he added.

But Will Reclassifying Patients Change Their Treatment?

But some panel members questioned whether adding CAC scores really would change management. Dr Robert Califf (Duke University, Durham, NC) said: "We saw that CAC scoring could create tertiles of risk within that intermediate group. But just changing risk a bit one way or another doesn't necessarily move you across a threshold that would change your likelihood of treatment."

Even though these data seem to reallocate exact level of risk, CAC isn't cheap and there is some radiation exposure.

And Dr Deepak L Bhatt (Cleveland Clinic, OH) wondered whether measuring other things, such as high-sensitivity C-reactive protein (hs-CRP), would have served a similar purpose. "In the US particularly, we are concerned about cost-effectiveness, and even though these data seem to reallocate exact level of risk, CAC isn't cheap and there is some radiation exposure. Might hs-CRP be an easier way to reclassify patients where we already have a therapy that might be applied?"

Erbel replied that CAC scores "show exposure to risk over time, whereas lab tests [such as hs-CRP] show you the value in a moment that may be changed by treatment. We had more than 75% of intermediate-risk subjects reclassified [to either high or low risk]. This is a first glimpse, and our next step is to do cost-effectiveness analysis, and we will also present results with hs-CRP."

Combining Traditional Risk Factors and CAC Scores Gave Best Predictions

Erbel explained that total coronary-calcium burden is considered a measure of the extent of atherosclerosis, and a large amount of coronary calcium indicates a high likelihood of rupture-prone plaque somewhere in the coronary arteries. As up to 50% of first heart attacks are sudden deaths, and 60% of all cardiac deaths occur outside the hospital, prevention is key to reducing fatality from coronary artery disease, he noted.

In the Heinz Nixdorf Recall study, Erbel and colleagues recruited 4487 subjects without known coronary disease, ranging in age from 45 to 75 years, half of whom were women. The participants were placed into risk categories on the basis of standard cardiovascular risk factors, as defined by the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) guidelines. Electron-beam CT was used to measure the Agatston coronary-calcium score.

Of the 4137 participants with complete follow-up data, 93 suffered cardiac death or nonfatal MI--the primary end point--including 28 women. When coronary-calcium scores in the highest quartile were compared with those in the lowest, the relative risk of a cardiac event was 2.12 for women and 9.48 for men, when adjusted for NCEP ATP III category.

Combining CAC scores with information on NCEP ATP III risk category meant that prediction was better still. This analysis showed that those who are at low risk based on ATP III guidelines would not benefit from CAC screening, said Erbel, as a high CAC score (400 or greater) was found in only 3% of low-risk patients.

Among those at intermediate risk, the 14% who should be reclassified based on high CAC scores had an event rate of >8% over five years, while the 63% who should be reallocated to a low-risk group--because of CAC scores under 100--had an event rate of just 1% over five years, he noted.

The results also suggest that risk prediction could be improved in those at high risk according to the ATP III guidelines who score low on CAC screening, he said.

Erbel added that his team plans to follow up on patients for the next five years so that they can analyze the 10-year risk prediction capability of CAC scoring and other risk factors.

Appropriateness Testing Could Help Reduce Unnecessary SPECT-MPI Use

Meanwhile, a pilot study presented in the same session shows that assessment of the appropriateness of another imaging modality, single-photon-emission computed tomography myocardial perfusion imaging (SPECT-MPI), could help to reduce improper use of such technology.

Presenting the results, Dr Robert C Hendel (Midwest Heart Specialists, Fox River Grove, IL) said that the growth and cost of cardiovascular imaging has placed renewed attention on optimal test ordering. The true nature of under- or overuse is not known, he said, despite the development and publication of SPECT-MPI appropriate-use criteria (AUC) in 2005. Although these criteria are increasingly being adopted, evaluation in community-practice settings has been lacking, he noted.

In the study, he and his colleagues tracked real-world clinical use in six cardiac imaging practices in a total of just over 6000 patients, with staff members prospectively collecting and entering clinical data into an online database for all patients referred for SPECT-MPI. An algorithm automatically determined whether the study was suitable for each patient according to the AUC and tracked patterns of inappropriate imaging.

Overall, 66% of the SPECT-MPIs performed were deemed appropriate, 13.9% were uncertain, and 13.4% were found to be inappropriate. There was a large variability between sites, with a fivefold difference between the best- and worst-performing facilities. The most common instance of unnecessary testing was in asymptomatic, low-risk patients; these people accounted for almost half of all the inappropriate use, he noted.

And contrary to what is often assumed--that physicians order unnecessary tests at their own facilities--Hendel and colleagues found that most inappropriate test ordering came from outside a practice, primarily from noncardiologists in primary care. Of the inappropriate referrals, 19.5% came from noncardiologists, compared with 13.2% for cardiologists (p<0.0001).

"This tells us we need to focus our education initiatives on primary-care doctors," Hendel told a press conference. "If we focus on the top four inappropriate indications, we can reduce testing by 12% and inappropriate testing by 90%. This pilot demonstrates the feasibility of this approach: appropriateness can be measured, testing ordering can be evaluated, and physicians can be educated about how to improve their performance."

Erbel has received consulting fees and honoraria from AstraZeneca and Volcano and a research grant from GE Imatron. Hendel has received consulting fees and honoraria from PHx Health, Astellas Pharma, and GE Healthcare and research grants from GE Healthcare and Astellas Pharma.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.