Author + information
- Michael J. Blaha, MD, MPH* (, )
- Roger S. Blumenthal, MD and
- Khurram Nasir, MD, MPH
- ↵*Johns Hopkins Ciccarone Preventive Cardiology Center, Blalock 524C Division of Cardiology, 600 North Wolfe Street, Baltimore, Maryland 21287
We read with interest the article by Gottlieb et al. (1) on the utility of coronary artery calcium (CAC) for excluding obstructive coronary artery disease (CAD) within the CORE64 (Coronary Evaluation Using Multi-Detector Spiral Computed Tomography Angiography Using 64 Detectors) study. In this high-risk group of patients suspected of having obstructive CAD on clinical grounds, a CAC score of 0 markedly reduces the likelihood of >50% stenosis (19% vs. 71% for CAC >10), but not sufficiently to use CAC to definitely exclude important CAD. The authors rightfully emphasize the high pre-test probability in their symptomatic cohort, and differentiate their results from the excellent prognosis seen with CAC = 0 in the asymptomatic screening population. We would add that, in accordance with Bayes' theorem, there likely will neverbe a test that conclusively excludesobstructive CAD in symptomatic patients with a clinical picture that warrants invasive angiography.
The accompanying editorial (2) makes much broader conclusions about the utility of CAC, perhaps confusing potential applications for CAC testing. In clinical practice, CAC is used most commonly as part of global risk stratification in asymptomatic patients to guide selection of appropriate pharmacotherapy. The editorial discusses the use of CAC “to decide who should be referred for [coronary angiography],” and attributes to our iJACCpaper (3) on mortality in asymptomatic patients the conclusion that “the absence of coronary calcification was a reliable predictor of the absence of angiographic CAD.”
Guidelines do not support the use of CAC to determine who needs angiography. No test should prevent referral for angiography that is indicated based on clinical criteria, such as quality and duration of chest pain and accompanying risk factors. Prior American College of Cardiology and American Heart Association statements and the recent landmark MESA (Multi-Ethnic Study of Atherosclerosis) (4) indicate that it is reasonable to use CAC for risk stratification in intermediate-risk asymptomatic patients. Our study followed >44,000 asymptomatic patients referred for CAC scanning for the occurrence of all-cause mortality, showing that CAC = 0 portends an excellent prognosis with estimated 10-year mortality of ∼1%. The confusion of asymptomatic versus clearly symptomatic patients, and clinical events versus angiographic coronary artery disease, hampers the unbiased assessment of the utility of CAC in the medical literature.
It is important to note that the conclusion in the editorial that the Gottlieb et al. (1) paper presents a “starkly contrasting picture” to a prior systematic review is based on a statistical error. In the small Gottlieb et al. (1) study, which considered CAC = 0 as a “positive study,” the positive predictive value (PPV) and negative predictive value (NPV) are calculated in exactly the opposite way from the large systematic review of over 10,000 patients by Sarwar et al. (5). When calculated the same way, the NPV of 81% and PPV of 68% is not much different (accounting for population differences, including exclusion of high CAC scores in CORE64) than the NPV of 93% and PPV of 68% presented in the systematic review.
Once again, Bayes' theorem is critical. Although CAC = 0 may not definitively exclude important CAD in patients referred for coronary angiography, there may be potential applications in lower-risk patients presenting with atypical chest pain features.
- American College of Cardiology Foundation
- Gottlieb I.,
- Miller J.M.,
- Arbab-Zadeh A.,
- et al.
- Redberg R.F.
- Blaha M.,
- Budoff M.J.,
- Shaw L.J.,
- et al.
- Sarwar A.,
- Shaw L.J.,
- Shapiro M.D.,
- et al.