In the JournalsPerspective

CAC score better predictor than standard risk factors of major coronary stenosis

In symptomatic patients, coronary artery calcium score predicted coronary stenosis of more than 50% more accurately than conventional risk factors, according to results of the Euro-CCAD study.

For the retrospective study, researchers analyzed 5,515 symptomatic patients (mean age, 60 years; 60.6% men) from seven centers in six countries. All were assessed for conventional risk factors and had a coronary artery calcium (CAC) score taken using local CT protocol. All were assessed for presence and amount of coronary artery stenosis by conventional or CT coronary angiography.

Rachel Nicoll, BSc, a PhD candidate from the department of public health and clinical medicine, Umeå University and Heart Centre, Umeå, Sweden, and colleagues determined that 28% of patients had significant coronary stenosis, defined as > 50%; 64% had CAC score > 0; median CAC score was 21 and mean number of risk factors was 2.6 per patient.

Predictors of major coronary stenosis

In a multivariate analysis of the conventional risk factors, the following variables predicted coronary stenosis > 50%: male sex (B = 1.07; OR = 2.92; 95% CI, 2.52-3.39), diabetes (B = 0.7; OR = 2.01; 95% CI, 1.67-2.42), smoking (B = 0.36; OR = 1.44; 95% CI, 1.25-1.65), hypercholesterolemia (B = 0.28; OR = 1.32; 95% CI, 1.15-1.51), hypertension (B = 0.24; OR = 1.28; 95% CI, 1.1-1.47), family history of CAD (B = 0.2; OR = 1.23; 95% CI, 1.07-1.4) and age (B = 0.03; OR = 1.03; 95% CI, 1.027-1.04).

When the researchers added CAC score to the multivariate model, it became the most significant predictor of coronary stenosis > 50% (B = 1.25; OR = 3.48; 95% CI, 3.19-3.81), and was true regardless of how stenosis was assessed. Male sex, diabetes, smoking, family history and age remained significant predictors, but hypercholesterolemia and hypertension did not, they wrote.

When they analyzed the 3,106 patients for whom obesity data was available, they determined that obesity was not a significant predictor of coronary stenosis > 50% (P = .68).

Sensitivity of risk factors and/or CAC score for prediction of coronary stenosis > 50% was higher when assessed by conventional angiogram than by CT angiogram, but specificity was higher when measured by CT angiogram, Nicoll and colleagues wrote.

Cross-sectional analysis

They wrote that accuracy of CT angiogram for predicting coronary stenosis > 50% using the CAC score only was higher (area under the curve = 0.85) than the accuracy of conventional angiogram for predicting coronary stenosis > 50% using CAC score and conventional risk factors (AUC = 0.81).

“This supports the current application of using CT [coronary angiography] for diagnosing patients at intermediate risk with significant stenosis,” Nicoll and colleagues wrote. “These results support those of others and show that the CAC score is highly predictive for significant stenosis. It should therefore maintain its position in the screening process of symptomatic patients, particularly those at intermediate risk, with the caveat that 5.5% of our patients with zero CAC nevertheless had significant stenosis.” by Erik Swain

Disclosure: The researchers report no relevant financial disclosures.

In symptomatic patients, coronary artery calcium score predicted coronary stenosis of more than 50% more accurately than conventional risk factors, according to results of the Euro-CCAD study.

For the retrospective study, researchers analyzed 5,515 symptomatic patients (mean age, 60 years; 60.6% men) from seven centers in six countries. All were assessed for conventional risk factors and had a coronary artery calcium (CAC) score taken using local CT protocol. All were assessed for presence and amount of coronary artery stenosis by conventional or CT coronary angiography.

Rachel Nicoll, BSc, a PhD candidate from the department of public health and clinical medicine, Umeå University and Heart Centre, Umeå, Sweden, and colleagues determined that 28% of patients had significant coronary stenosis, defined as > 50%; 64% had CAC score > 0; median CAC score was 21 and mean number of risk factors was 2.6 per patient.

Predictors of major coronary stenosis

In a multivariate analysis of the conventional risk factors, the following variables predicted coronary stenosis > 50%: male sex (B = 1.07; OR = 2.92; 95% CI, 2.52-3.39), diabetes (B = 0.7; OR = 2.01; 95% CI, 1.67-2.42), smoking (B = 0.36; OR = 1.44; 95% CI, 1.25-1.65), hypercholesterolemia (B = 0.28; OR = 1.32; 95% CI, 1.15-1.51), hypertension (B = 0.24; OR = 1.28; 95% CI, 1.1-1.47), family history of CAD (B = 0.2; OR = 1.23; 95% CI, 1.07-1.4) and age (B = 0.03; OR = 1.03; 95% CI, 1.027-1.04).

When the researchers added CAC score to the multivariate model, it became the most significant predictor of coronary stenosis > 50% (B = 1.25; OR = 3.48; 95% CI, 3.19-3.81), and was true regardless of how stenosis was assessed. Male sex, diabetes, smoking, family history and age remained significant predictors, but hypercholesterolemia and hypertension did not, they wrote.

When they analyzed the 3,106 patients for whom obesity data was available, they determined that obesity was not a significant predictor of coronary stenosis > 50% (P = .68).

Sensitivity of risk factors and/or CAC score for prediction of coronary stenosis > 50% was higher when assessed by conventional angiogram than by CT angiogram, but specificity was higher when measured by CT angiogram, Nicoll and colleagues wrote.

Cross-sectional analysis

They wrote that accuracy of CT angiogram for predicting coronary stenosis > 50% using the CAC score only was higher (area under the curve = 0.85) than the accuracy of conventional angiogram for predicting coronary stenosis > 50% using CAC score and conventional risk factors (AUC = 0.81).

“This supports the current application of using CT [coronary angiography] for diagnosing patients at intermediate risk with significant stenosis,” Nicoll and colleagues wrote. “These results support those of others and show that the CAC score is highly predictive for significant stenosis. It should therefore maintain its position in the screening process of symptomatic patients, particularly those at intermediate risk, with the caveat that 5.5% of our patients with zero CAC nevertheless had significant stenosis.” by Erik Swain

Disclosure: The researchers report no relevant financial disclosures.

    Perspective
    Michael J. Blaha

    Michael J. Blaha

    Consistent with data on the superior risk predictive value of CAC vs. risk factors alone, this study demonstrates that a direct measure of coronary atherosclerotic burden is superior to risk factors alone for predicting a significant coronary artery stenosis. 

    This study doesn’t tell us much about CT angiograms vs. invasive coronary angiography. It was not designed to compare these two modalities, and most patients in the study had only one of these tests. However, we can infer from this study that the CAC was slightly more accurate in predicting significant coronary stenosis on CTA vs. coronary angiography, which is not surprising since both CAC and CTA use CT technology.

    These results will not really change clinical practice. However, it will remind practitioners how poor risk factors are for predicting significant stenosis, and will reiterate the need to incorporate subclinical coronary atherosclerosis data in risk assessment when it is available. It may also remind practitioners how common coronary stenoses are in patients with CAC scores above 400.

    Further research needs to be done on patients who have a low pre-test probability of coronary stenoses. There, a CAC score of zero, for example, may be sufficient to exclude a coronary stenosis. However, more research needs to be done to support this argument.

    Additionally, more research needs to be done to understand the appropriate course of action in minimally symptomatic patients with CAC > 400.  When is a stress test, for example, indicated given the high likelihood of significant stenosis?

    • Michael J. Blaha, MD, MPH
    • Director, clinical research, the Ciccarone Center for the Prevention of Heart Disease Assistant professor of medicine, Johns Hopkins University School of Medicine

    Disclosures: Blaha reports no relevant financial disclosures.