Journal of the American College of Cardiology Vol. 59, No. 4, 2012© 2012 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00Published by Elsevier Inc. doi:10.1016/j.jacc.2011.10.862

EDITORIAL COMMENT

Coronary ComputedTomography AngiographyThe Challenge of Coronary Calcium*

Steven E. Nissen, MD

Cleveland, Ohio

In this issue of the Journal, Arbab-Zadeh et al. (1) report onthe influence of calcification and the pre-test likelihood ofdisease on the diagnostic accuracy of coronary computedtomography angiography (CTA) in a small multicenterstudy (CORE-64 [Coronary Artery Evaluation Using 64-Row Multidetector Computed Tomography Angiogra-phy]). Their report provides valuable insights into theimportance of coronary calcification as an unresolved limi-tation of CTA. The original CORE-64 report pre-specifiedexclusion of patients with a calcium score �600 Agatstonunits, a criterion that eliminated 89 of 405 patients (22%)from the analysis (2). In the current paper (1), the authorshave appropriately included such patients, documenting animportant limitation in the diagnostic accuracy of CTAwhen applied in an unselected population. For patients withAgatston score �600, the performance of CTA was unfa-vorable. Regardless of pre-test likelihood of disease, thenegative predictive value was 0.50 for quantitative assess-ment of CAD and 0.63 for visual assessment.

See page 379

The current study also demonstrates that lesser degrees ofcoronary calcification can impair the diagnostic accuracy ofcoronary CTA. Among patients with any calcification score�0, the negative predictive value of CTA in patients with ahigh pre-test probability of disease ranged from approxi-mately 50% to 60%. Similar deterioration in performancewas observed for patients with coronary calcification and anintermediate probability of disease, particularly if the Agat-ston score exceeded 100. The careful analysis provided bythe current study will be valuable to practitioners consider-ing how to optimally apply coronary CTA in routine clinicalpractice. These findings suggest that CTA, using current

*Editorials published in the Journal of the American College of Cardiology reflect theviews of the authors and do not necessarily represent the views of JACC or theAmerican College of Cardiology.

From the Cleveland Clinic Foundation, Cardiovascular Medicine, Cleveland,

Ohio. The author has reported he has no relationships relevant to the contents of thispaper to disclose.

technology, probably should not be used for diagnosticpurposes in patients with substantial coronary calcification.

Understanding the exclusion criteria in any imaging studyis critical to interpretation of the findings. The originalCORE-64 study report, like several similar studies, wasdesigned to evaluate coronary CTA under reasonably idealconditions. The list of exclusion criteria included priorcardiac surgery, creatinine levels �1.5 mg/dl, atrial fibrilla-tion, class III or IV heart failure, coronary interventionwithin the past 6 months, intolerance to beta-blockers, anda body mass index �40 kg/m2. It is useful to consider theprofile of patients undergoing catheterization for suspectedcoronary disease in the United States. The median bodymass index of patients undergoing catheterization in recentstudies exceeds 30 kg/m2, with substantial numbers exceed-ing 40 kg/m2. The prevalence of atrial fibrillation in theU.S. population is substantial and increases with age,reaching nearly 10% by 80 years of age (3). Approximately11% of American over the age of 60 years have a creatininelevel �1.6 mg/dl (4). The prevalence of chronic obstructivepulmonary disease is also high among patients undergoingcatheterization, suggesting that intolerance to beta-blockerswill result in exclusion of additional patients.

For ethical reasons, the CORE-64 study population onlyincluded patients for whom cardiac catheterization wasdeemed clinically indicated. Ninety-eight patients (26%)actually had known coronary disease. Such patients repre-sent a high-risk subset of those seeking medical care forchest pain of suspected cardiac origin, which likely explainsthe relatively high prevalence of coronary calcification. Thesymptoms exhibited by these patients were sufficient towarrant the risk and expense of cardiac catheterization. Inthis population, we must consider whether the results ofCTA would allow deferral of coronary angiography. In thecurrent analysis (1), the negative predictive value of coronaryCTA for patients with known CAD was approximately0.50. In the group with an intermediate probability ofdisease, the sensitivity and specificity were more acceptable,with false positive and false negative rates in the 10% to 15%range. Therefore, as documented in the current study (1),error rates remain too high to recommend use of CTA as analternative to cardiac catheterization or imaging stress testsin patients with a high probability of CAD.

In determining when to use coronary CTA in clinicalpractice, we must also consider the potential harm producedby coronary CTA imaging as currently practiced. The dosesof radiation are substantial, although gradually falling atsophisticated centers with technical improvements in studymethods. The median radiation dose in a large multicenterstudy of CTA was 12 mSv, equivalent to 600 chestradiographs (5). The CORE-64 study authors reportslightly higher doses, ranging from 12 to 15 mSv, with a toplimit of 20 mSv. By comparison, most studies of conven-tional diagnostic coronary angiography report a dose range

from 2 to 6 mSv. Accordingly, the radiation dosage from

389JACC Vol. 59, No. 4, 2012 NissenJanuary 24, 2012:388–9 Coronary CT Angiography

CTA is equivalent to 3 to 7 diagnostic catheterizations. Thelikely effect of such doses has been debated, but it isgenerally accepted that there is no safe dosage of radiationand that exposure to current dose levels will predictablyincrease rates of malignancy. Many patients will likelyreceive numerous CTA studies and other radiographic andnuclear imaging procedures for cardiac and noncardiacindications. Accordingly, the patient care benefits of anyimaging procedure that exposes patients to ionizing radia-tion must clearly exceed the hazards.

In addition to the problem of coronary calcificationdemonstrated by the current study, it must also be recog-nized that the presence of a coronary stenosis per se does notaccurately determine the physiological consequences of suchlesions. The presence or absence of a 50% stenosis in thecoronary arteries does not define whether revascularizationis appropriate or inappropriate. Symptoms and the presenceof reversible ischemia are the pivotal factors determining theappropriateness of coronary interventions. Accordingly, inthe assessment of patients with an intermediate probabilityof CAD, coronary CTA remains less useful than imagingstress tests. For such patients, we are less interested in thepresence of stenosis than evidence documenting whetherobstructions are ischemia-producing. Accordingly, an ab-normal CTA may require another imaging procedure inmany patients to determine the physiologic impact ofobserved stenoses.

In future studies, what proof of benefit should wedemand of CTA? To justify widespread usage, the tech-nique must be shown to improve relevant patient outcomes,including survival, improve the accuracy of diagnoses, orsubstantially reduce health care costs. None of these benefitshas yet been demonstrated convincingly for coronary CTA.Nonetheless, CT coronary imaging continues to grow inutilization, often fueled by weekend “training” coursesdesigned to achieve certification for participants. Somepractitioners have purchased devices to enable in-officeimaging of patients, a situation that creates incentives forover-use. Unfortunately, for some patients, an abnormal CTscan rather than symptoms or evidence of ischemia, are theprimary driver for invasive procedures, including coronaryinterventions.

A case recently reported in the Archives of InternalMedicine poignantly illustrates the risks of decision making

overly influenced by the presence of a stenosis in the

presence of coronary calcification (6). We describe a 52-year-old nurse with atypical chest pain and a low pre-testprobability of disease, who underwent CTA to “reassure”her. The presence of a difficult to visualize calcified leftanterior descending artery lesion led to cardiac catheteriza-tion, which resulted in a catastrophic dissection of the leftmain coronary artery, eventually leading to heart transplan-tation. This case dramatically highlights the importance ofunderstanding how coronary calcification limits the accuracyof CTA.

The current study reported by Arbab-Zadeh et al. (1)helps us understand what must be expected in subsequentstudies examining the clinical utility of CTA. Future studiesmust evaluate important clinical outcomes, not just theextent of stenoses, in a wide spectrum of patients notselected because they represent ideal candidates for CTA.Pending such evaluation, coronary imaging using CTAshould be used sparingly, with full recognition of theradiation burdens and risks of misdiagnosis.

Reprint requests and correspondence: Dr. Steven E. Nissen,Cleveland Clinic Foundation, Cardiovascular Medicine, 9500Euclid Avenue, Desk F-15, Cleveland, Ohio 44195-0001. E-mail:nissens@ccf.org.

REFERENCES

1. Arbab-Zadeh A, Miller JM, Rochitte CE, et al. Diagnostic accuracy ofcomputed tomography coronary angiography according to pre-testprobability of coronary artery disease and severity of coronary arterialcalcification: the CORE-64 (Coronary Artery Evaluation Using 64-Row Multidetector Computed Tomography Angiography) interna-tional multicenter study. J Am Coll Cardiol 2012;59:379–87.

2. Miller JM, Rochitte CE, Dewey M, et al. Diagnostic performance ofcoronary angiography by 64-row CT. N Engl J Med 2008;359:2324–36.

3. Roger VL, Go AS, Lloyd-Jones DM, et al. American Heart Associa-tion Statistics Committee and Stroke Statistics Subcommittee. Heartdisease and stroke statistics—2011 update: a report from the AmericanHeart Association (erratum in: Circulation 2011;123:e240). Circulation2011;123:e18–209.

4. Coresh J, Wei GL, McQuillan G, et al. Prevalence of high bloodpressure and elevated serum creatinine level in the United States:findings from the third National Health and Nutrition ExaminationSurvey (1988–1994). Arch Intern Med 2001;161:1207–16.

5. Hausleiter J, Meyer T, Hermann F, et al. Estimated radiation doseassociated with cardiac CT angiography. JAMA 2009;301:500–7.

6. Becker MC, Galla JM, Nissen SE. Left main trunk coronary arterydissection as a consequence of inaccurate coronary computed tomo-graphic angiography. Arch Intern Med 2011;171:698–701.

Key Words: angiography y cardiac imaging y coronary disease.