7/27/2019 hecht.pdf

http://slidepdf.com/reader/full/hechtpdf 1/5

Questions in Cardiovascular CT 

How much calcium is too much calcium for coronarycomputerized tomographic angiography?

Harvey S. Hecht, MD, FACC*, Tandeep Bhatti, DO

 Lenox Hill Heart and Vascular Institute, 130 E 77th Street, New York, NY 10021 USA

Abstract. The coronary artery calcium (CAC) score above which it is recommended that coronary

computerized tomographic angiography (CTA) not be performed has been steadily increasing. Cur-

rently, calcium scoresϾ 1000 are thought to prohibit CTA accurate interpretation. However, a reasoned

approach suggests that there is no absolute upper limit that applies to all patients and imaging centers.

To anticipate the problems posed by calcium, a CAC scan must be obtained before CTA. Understanding

the clinical goals of the CTA and the source and recognition of CAC-based imaging artifacts can enable

accurate clinical CTA examinations even in the setting of high calcium scores.

© 2008 Society of Cardiovascular Computed Tomography. All rights reserved.

KEYWORDS:Artifacts;

Coronary calcium;

CTA

Introduction

The coronary artery calcium (CAC) score above which itis recommended that coronary computerized tomographic

angiography (CTA) not be performed has been steadily

increasing. Currently, calcium scoresϾ 1000 are thought to

prohibit CTA accurate interpretation. However, a reasoned

approach suggests that there is no absolute upper limit that

applies to all patients and imaging centers.

To anticipate the problems posed by calcium, a CAC

scan must be obtained before CTA. Most centers perform

CAC scanning before CTA; many do so only on request.

Although routine CAC scanning is advisable, for centers in

which is not the clinical routine, heavy calcification will be

apparent on the scout chest scan performed before contrastinjection and should alert personnel to the potential prob-

lem.

General considerations

CTA objectives

There are 2 broad goals of CTA. The first goal is detection

of obstructive disease. The quantitative approach to stenosis

measurement depends on the ability to clearly separate contrast

from calcified plaque, whether lesion severity is determined

from qualitatively “eyeballing,” or a quantitative measurement,

or from cross-sectional analysis and derivation of minimum

luminal area. However, it is critical to realize that a clinically

effective CTA scan result does not require the ability to accu-

rately analyze every millimeter of the vessel. Rather, the cru-

cial issue is to arrive at the appropriate clinical recommenda-

tion. Consequently, if in most cases there is readily detectiblesignificant obstructive disease secondary to evaluable, less-

calcified plaque, the inability to evaluate heavily calcified areas

is rendered irrelevant. For example, in Figure 1, an 88-year-old

man with a CAC score of 4253 underwent CTA. There was

clearly significant left main and right coronary stenosis sec-

ondary to almost entirely noncalcified plaque, in addition to a

significant left anterior descending stenosis resulting from

mixed plaque. The presence of adjacent densely calcified non-

evaluable areas did not interfere with arriving at the correct

diagnosis.

Conflict of interest: Dr. Hecht discloses that he received grant support

and honoraria from Philips Medical Systems. Dr. Bhatti reports no conflicts

of interest.

* Corresponding author.

E-mail address: hhecht@aol.com

The online version of this article contains supplementary data.

Submitted April 9, 2008. Accepted for publication April 16, 2008.

1934-5925/$ -see front matter © 2008 Society of Cardiovascular Computed Tomography. All rights reserved.

doi:10.1016/j.jcct.2008.04.003

Journal of Cardiovascular Computed Tomography (2008) 2, 183–187

7/27/2019 hecht.pdf

http://slidepdf.com/reader/full/hechtpdf 2/5

In addition, data suggest that the more severe stenoses are

associated with less-calcified plaques, minimizing the impor-

tance of the densely calcified areas.1 The increased positive

remodeling associated with extensively calcified plaque pre-

serves the lumen, despite the large plaque burden. Nonetheless,

there will be cases with extensive calcified plaque that will

prevent evaluation of those sites, without clear obstructive

disease elsewhere. In these situations, it is appropriate to state

in the report that “There is extensive calcified plaque at whichlocation significant stenosis cannot be excluded”.

The second goal is assessment of plaque burden. Char-

acterization and quantitation of both noncalcified and cal-

cified plaque offers a more complete assessment of risk than

does the calcified component alone. In general, the degree

of vulnerability is thought to be directly related to the

noncalcified component. A quantitative assessment of 

calcified plaque and a semiquantitative evaluation of non-

calcified plaque enable stratification of the extent of ath-

erosclerosis, which in some centers is considered of equalor greater importance than evaluation of obstructive dis-

Figure 1 An 88-year-old man presented with chest pain. The CAC score was 4253 with extensive calcified plaque in all vessels. CTA

showed readily apparent, significant left main (LM) (A) and right coronary artery (RCA) (C) stenoses associated with almost exclusivelynoncalcified plaque, as well as significant left anterior descending (LAD) coronary artery lesions secondary to mixed calcified and

noncalcified plaque. The findings were confirmed by invasive angiography (B and D).

184 Journal of Cardiovascular Computed Tomography, Vol 2, No 3, May/June 2008

7/27/2019 hecht.pdf

http://slidepdf.com/reader/full/hechtpdf 3/5

ease. Clearly, this analysis is not hindered by calcified

plaque.

Calcium artifacts

Technical problems are inherent in CT imaging regionsadjacent to calcified plaque of varying densities.

Partial volume effect

Calcific densities may create artifacts because densities

are greatly influenced by the “company they keep.” En-

hancement of the Hounsfield units of adjacent contrast by

the partial volume effect of the denser calcium (blooming)

may make the calcified region appear larger and thus over-estimate the degree of luminal obstruction. Broadening the

Figure 2 A 64-year-old woman underwent coronary CTA for evaluation of chest pain. CAC score was 1962. CTA of the left main and

left anterior descending (LAD) arteries showed extensive calcified plaque. Increasing the window width from 806 HU (A) to 3970 HU(B) dramatically decreased the apparent size of the calcified plaques. Analysis of cross-sectional images obtained from the straightened

multiplanar reconstruction (D) showed no change in Hounsfield units (E and F) despite the dramatic reduction in size and appearance of 

the calcified plaque. Invasive angiography showed no significant disease (C).

185Hecht and Bhatti How much is too much calcium for CTA?

7/27/2019 hecht.pdf

http://slidepdf.com/reader/full/hechtpdf 4/5

gray scale by increasing the width of the image windowmay visually shrink the size of the calcified plaque and

appear to make it less space occupying. However, this is

merely an optical illusion; the Hounsfield units do not

change (Fig. 2), and this technique should be used cau-

tiously. After window widening to evaluate a large calcified

plaque that appears to occupy the entire lumen, the least

dense area may still be calcium and should not be mistak-

enly identified as contrast. The use of a sharper reconstruc-

tion filter, as would be used for stent analysis, may decrease

the blooming and facilitate interpretation of highly calcified

segments.

Shadowing

If the calcium density is very high, eg, Ͼ1000

Hounsfield units (HUs), the contrast may appear hypodense

as the result of photopenia from the “shadowing” effect of 

the calcium, even in the absence of any significant stenosis.

Unfortunately, there is no absolute level of calcium

Hounsfield units above which there will be shadowing and

below which there will be a partial volume effect. In gen-

eral, the greater the Hounsfield unit gradient between the

calcified plaque and the adjacent hypodense area, the more

likely is the contribution of shadowing to the hypodensity.The CTA of the left circumflex in a 65-year-old man with a

totally occluded left anterior descending coronary artery isshown in Figure 3. The apparent severe ostial stenosis

secondary to mixed plaque was, in reality, the result of 

shadowing of the lumen by the dense calcification, mimick-

ing the appearance of a critical lesion.

Motion

Motion artifacts, whether secondary to rapid heart rate,

arrhythmias, patient motion, or respiration, are exaggerated

in calcified compared with noncalcified areas. Conse-

quently, the use of ␤-blockers for heart rate reduction to

Ͻ60 beats/min is recommended to minimize image blur-ring.

Patterns of calcification

Equally high calcium scores may result from numerous

small nonobstructive plaques that will not interfere with

image interpretation, or from several large calcified plaques

that make interpretation difficult. In reality, even a CAC

score of 100 may pose a problem if there is a single dense

plaque. In Figure 4, a single moderately large calcified

plaque obscures the lumen (Fig. 4A) in an angiographically

minimally narrowed vessel (Fig. 4B), whereas multiplesmaller calcified plaques do not obscure 3 significant focal

Figure 3 CTA of the left circumflex (LCx) coronary artery in a 65-year-old man with a totally occluded left anterior descending coronary

artery showed a severe ostial stenosis secondary to mixed calcified and noncalcified plaque (A). Subsequent invasive angiography

(B) showed no obstructive disease. The apparent narrowing was the result of shadowing of the lumen by the dense calcification, mimicking

the appearance of a critical lesion.

186 Journal of Cardiovascular Computed Tomography, Vol 2, No 3, May/June 2008

7/27/2019 hecht.pdf

http://slidepdf.com/reader/full/hechtpdf 5/5

stenoses (Fig. 4C), confirmed on invasive angiography

(Fig. 4D).

Recommendations

In summary, there can be no arbitrary, absolute calcium

score above which coronary CTA should not be performed.

Rather, each case must be individualized with respect to the

primary purpose of the examination (evaluation of obstruc-

tive disease versus plaque burden), the pattern of calcifica-

tion, and the expertise of the interpreting physician. A CAC

score Ͼ1000 or a pattern of isolated large, dense calcifica-

tions even with CAC Ͻ1000 should alert personnel to

potential interpretation problems and should not be fol-

lowed by CTA if the interpreting physician does not have

extensive experience in dealing with these issues.

After increasing experience, at a point to be determinedby the comfort level of each practitioner preferably with the

guidance of a mentor, physicians may authorize the perfor-

mance of CTA in patients with these CAC characteristics, to

be analyzed with full understanding of the principles de-

scribed above. There is no absolute minimum number of 

cases to acquire the requisite expertise.

Reference

1. Lubarsky L, Prakash M, Jelnin V, Panogopoulos G, Hecht HS: Relation

of plaque composition to degree of stenosis evaluation by 64 detectorCTA. J Cardiac Computed Tomogr. 2007;1(suppl):S34.

Figure 4 A patient with a CAC score of 135 in which a single moderately large calcified plaque in the left anterior descending (LAD)

coronary artery obscures the lumen (A) and poses interpretive problems in an angiographically minimally narrowed vessel ( B). In sharp

contrast, in a patient with a CAC score of 2341, multiple smaller calcified plaques in the left anterior descending coronary artery do not

obscure 3 significant focal stenoses (C), confirmed on invasive angiography (D).

187Hecht and Bhatti How much is too much calcium for CTA?