(Circulation. 1995;92:632-636.)
© 1995 American Heart Association, Inc.
Articles |
From the Division of Cardiology, University of California, Irvine, and the St Johns Cardiovascular Research Center, Harbor-UCLA Medical Center.
Correspondence to Dr Nathan D. Wong, Preventive Cardiology Program, C240 Medical Sciences I, University of California, Irvine, CA 92717.
Key Words: atherosclerosis coronary disease tomography calcium
| Introduction |
|---|
|
|
|---|
There has been a dramatic increase in the use of electron beam computed tomography to screen for coronary calcium phosphate deposits. This radiographic procedure, available at increasingly numerous centers in the United States and abroad, can detect small amounts of these deposits. The test costs about $400 (technical fee, including professional interpretation), takes 15 minutes to perform, involves little radiation (similar to that of a barium enema or chest radiograph series), and is noninvasive, not requiring injections or drugs to perform the procedure and obtain results.
Advocates of universal coronary calcium screening state that since coronary calcium signifies atherosclerosis, a potentially deadly disease, its detection is important, and a positive screen is reason for aggressive management of risk factors or even further evaluation with exercise testing or angiography. They argue that asymptomatic persons whose coronary status is unknown would benefit from the test, since those with positive scans can be treated more aggressively than those with negative scans and thereby be prevented from suffering myocardial infarctions or cardiac death.
Others are not as convinced regarding the value of this new procedure. A recent science advisory published by the American Heart Association1 concluded that the "clinical use of ultrafast computed tomographic imaging to screen patients for coronary artery disease is not justified at this time." Major reasons given were (1) the lack of a precise correlation between the degree of atherosclerosis and coronary calcium and (2) the unknown prognostic significance of calcification detected by this technology.
We summarize in this review the strength of the evidence linking coronary calcium to atherosclerosis and coronary events. We then propose a paradigm on which the future development of this technique can be modeled. On the basis of these arguments, we try to explain the varying consensus concerning the implications of the findings on coronary calcium scans as well as that concerning indications for screening.
| Evidence Linking Coronary Calcium to Atherosclerosis |
|---|
|
|
|---|
Angiographic studies also support an association between calcium and luminal stenoses in symptomatic populations. Gianrossi and colleagues8 performed a meta-analytic review of the relation between fluoroscopically detectable coronary calcium and angiographic disease. The studies reviewed used conventional image intensifier fluoroscopy in symptomatic subjects undergoing angiography for clinical indications. It was found that accuracies for cardiac fluoroscopy approached those of exercise ECG and exercise thallium scintigraphy. Similar investigations comparing electron beam tomography for calcium and coronary angiography for stenoses in symptomatic subjects have produced sensitivities and negative predictive values approaching 100% but specificities and positive predictive values of lower magnitude (ie, 47% and 62%, respectively).9 10 11 A high percentage of arteries with significant compared with mild stenoses are seen to contain calcium,12 and those patients with symptomatic coronary artery disease are most likely to show coronary calcium.13 Although specificity may be enhanced by a more stringent definition for defining a positive test (eg, a greater area, score threshold, or number of calcified vessels) by electron beam computed tomography, this comes at the expense of reduced sensitivity.14 15 16
| Coronary Calcium as a Predictor of Coronary Events in Asymptomatic Populations |
|---|
|
|
|---|
For better or worse, coronary angiographic luminal narrowings are
accepted standards of pathology. Loecker et al17 performed
a prospective fluoroscopic evaluation of 613 asymptomatic aircrew
members undergoing conventional coronary fluoroscopy and coronary
angiography. These authors found that angiographic narrowing was
present in 17% of subjects, and found a sensitivity of 66% and a
specificity of 78% for the ability of fluoroscopy positive for calcium
to predict at least one angiographic luminal narrowing of at least
50%. We know from epidemiological data that at least 95% of healthy
men can expect to remain free of coronary events for
10
years.18 Even so, as many as 20% had fluoroscopic
coronary calcium, and 17% of those undergoing angiography had at least
one obstruction >50% of the normal luminal diameter. Their findings
suggest that although fluoroscopically detectable coronary calcium or a
positive angiogram signifies the presence of atherosclerosis, the mere
presence of either of these findings may not be a sufficient predictor
of prognosis.
The prevalence of coronary artery calcium detected by electron beam computed tomography is even higher and rises dramatically with age; >60% of asymptomatic women and 80% of asymptomatic men >60 years old have at least some detectable calcium by this method.19 20 Since a much smaller proportion of these individuals will eventually develop clinical coronary events,18 effective risk stratification requires that a certain threshold amount of calcium be present for a calcium screen to indicate "high risk." Sensitivity and specificity with respect to subsequent events in asymptomatic populations, who have recently been targeted for calcium screening, remain unknown, but the high prevalence of simple "positive" scans21 suggests that specificity is low for disease that is likely to become clinically manifest. The amount of disease that can be detected is so small as to greatly limit the predictive value of the simple detection of coronary calcium. For this reason, we suggest that accurate quantification of coronary calcium will be necessary to find useful thresholds above and below which the detection of calcium is of prognostic value.
There is preliminary direct evidence of an association between coronary artery calcium, risk factors, and prognosis. The presence (or number) of selected coronary risk factors, including reported hypertension, smoking, diabetes, male sex, and hyperlipidemia, corresponds to a higher prevalence (or quantity, based on total calcium score) of calcium among asymptomatic populations.19 20 22 23 After adjustment for coronary risk factors, the presence of coronary artery calcium is an independent indicator of prior history of coronary artery disease24 as well as angiographically obstructive disease.25 Furthermore, preliminary findings have linked calcium detected by fluoroscopy26 or electron beam tomography27 to an increased risk of future coronary events or revascularization.
| A Paradigm for Future Research |
|---|
|
|
|---|
If these three conditions are met, reliable thresholds for a positive test can be found such that specificity and predictive value are high enough to be clinically useful.
Accuracy in Quantifying Calcium Mass
Although limited work
has been done to validate calcium
measurements in coronary arteries, an arbitrary scoring system has been
used for several years.28 Important physical limitations
of this scoring algorithm have been identified, and work toward the
development of more accurate methods, including those that accurately
correspond to total calcium mass, is in
progress.29 30 31 32 A
recent report12 showed that both calcium score and area of
calcified deposits assessed by electron beam computed tomography
correlated highly (r=.95 to .96) with histomorphometric
calcium area. Investigators have also begun to demonstrate high
reliability for identifying calcified foci of a given area on repeat
scanning,33 and concordance of a positive test based on a
given number of calcified pixels34 has been reported by
some investigators.
Relation of Mass to Pathology
Recent
studies21 35 show an imperfect relation
between the amount of calcium and atherosclerosis. Whereas the absence
of coronary calcium at any site is highly specific for the absence of
obstructive disease, nonobstructive disease (<75% stenosis) is still
common in most such segments. Others show a closer correlation between
calcified plaque volume (r=.84 for individual coronary
arteries and r=.94 for whole-heart calcium volumes) and
histological plaque volume.36 The percentage of coronary
artery surface area with calcified plaque from autopsy subjects with
accidental deaths also tracks closely with coronary calcium prevalence
in a large, asymptomatic sample.19 In addition, the mass
of calcium may track progression or stabilization of
atherosclerosis,37 and such progression is measurable and
may predict prognosis.38 Also, retest reliability may be
sufficient for tracking of coronary calcium or atherosclerotic
progression, particularly in research studies.34
Pathological comparison studies and further prognostic data are needed
to confirm these reports, however. More importantly, the relation
between clinical plaque stability and amount, morphology, or changes in
coronary calcium is not clear. More research, including studies
evaluating longitudinal changes in coronary calcium with
atherosclerosis assessed by methods such as coronary angiography and
intravascular ultrasound, is needed to examine the relation of calcium
mass and morphology with volume, morphology, progression, and stability
of atherosclerotic plaque.
Relation of Pathology to Risk
Although it is plausible that a
direct and strong relation between
amount of atherosclerosis and risk exists, the available evidence to
support this is limited. Neither has it been proved that calcified
atherosclerosis is more malignant than noncalcified
atherosclerosis,39 40 41 although persons
with identified
coronary artery calcium appear to have a graver
prognosis,26 27 possibly because such persons have
correspondingly more atherosclerosis in general. There is, however,
some evidence that calcified atherosclerosis may be associated with
greater
stability.39 40 41 42 43
One study40 showed
that calcified plaques may be less likely to rupture. Studies using
intravascular ultrasound before and after
interventions42 43 also have shown that calcified
plaques
are less likely to be associated with arterial restenosis than are
noncalcified plaques.
Reasons for the Controversy
Electron beam computed tomography
is a noninvasive tool that can
detect even small amounts of coronary calcium (which may or may not be
associated with clinically significant lumen stenosis). However,
atherosclerosis is detected in many individuals who may never suffer
from its clinical consequences. This is not the only example of the
dilemma of widespread potentially malignant pathology that, if left
unknown and untouched, usually runs a benign course. Harach et
al44 estimated that almost everyone >50 years old would
have a diagnosis of thyroid carcinoma if sufficiently thorough biopsies
were performed, although thyroid carcinoma is a relatively uncommon
clinical malignancy. Fortunately, routine biopsy of the entire
population has not been seriously proposed. The advantages and
disadvantages of coronary calcium screening of large segments of the
asymptomatic population, in particular, which could also uncover
pathology likely to remain clinically nonmanifest, needs to be weighed
in a similar light.
At present, identification of coronary calcium may serve to confirm the presence, or even relative extent, of atherosclerosis, but not necessarily the clinical severity of disease. For the large number of asymptomatic individuals with coronary calcium, however, there are no established guidelines for referral for either an exercise treadmill test, angiogram, or even more aggressive medical treatment. Therefore, at present, acceptable medical decisions may be difficult to make on the basis of the results of coronary calcium screening. The presence of coronary calcium, however, might alert the patient to a potential emerging problem and encourage him or her to better manage his or her risk factors by making important lifestyle changes. This is currently under investigation as a potential benefit of coronary artery screening by electron beam tomography. However, this supposed beneficial effect must be balanced against the potentially detrimental effects of negative labeling, anxiety, and the cost to society for additional and potentially unwarranted diagnostic and invasive procedures.
There are, however, possible scenarios in which electron beam computed tomography may be helpful in clinical decision making. For example, we must evaluate whether an abnormal exercise treadmill test in an asymptomatic subject with a negative coronary calcium screen might assist the clinician in determining whether the exercise test is a false-positive due to hyperventilation,45 medications, or an estrogen effect.46 Conversely, whether a positive exercise test might confirm the significance of a positive coronary calcium screen as an aid to effective risk stratification47 needs consideration. Also, a positive screen in an individual with chest pain could provide evidence that such chest pain is in fact linked to coronary atherosclerosis. Furthermore, persons in extremely high-risk occupations, such as airline pilots and emergency response workers, might benefit from coronary calcium screening in that a negative result, together with a negative history, might indicate a low likelihood of a cardiac catastrophe. One must, however, use caution in interpreting positive results, and one must be willing to remove from these high-risk positions some able persons whose likelihood of cardiac events is not much higher than that of their colleagues with negative coronary scans. Investigations are also needed to determine whether asymptomatic persons with elevated risk factors who have significant coronary calcium are at greater risk of coronary heart disease events and could benefit from further diagnostic evaluation or intervention beyond risk factor control.
| Summary |
|---|
|
|
|---|
Despite lack of validation, this test has widespread appeal, both to the public as a means of being able to find out the condition of their coronary arteries "without injections or dye" and to hospitals and private medical groups who view this both as an innovation in cardiovascular diagnosis and as a potentially profitable diagnostic procedure. Physicians asked to explain the results of the test must clearly delineate to patients the meaning of their results. Since the clinical value and cost-effectiveness of identifying individuals at risk over and above the existing methods of risk-factor evaluation are not known, it is not possible to recommend universal screening at this time. However, active research and emerging findings by many investigators may soon make it possible to recommend whether specific risk subgroups of individuals might benefit from coronary calcium screening.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. J. Pletcher, J. A. Tice, M. Pignone, and W. S. Browner Using the Coronary Artery Calcium Score to Predict Coronary Heart Disease Events: A Systematic Review and Meta-analysis Arch Intern Med, June 28, 2004; 164(12): 1285 - 1292. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Schurgin, S. Rich, and T. Mazzone Increased Prevalence of Significant Coronary Artery Calcification in Patients With Diabetes Diabetes Care, February 1, 2001; 24(2): 335 - 338. [Abstract] [Full Text] |
||||
![]() |
P. Raggi, T. Q. Callister, B. Cooil, Z.-X. He, N. J. Lippolis, D. J. Russo, A. Zelinger, and J. J. Mahmarian Identification of Patients at Increased Risk of First Unheralded Acute Myocardial Infarction by Electron-Beam Computed Tomography Circulation, February 29, 2000; 101(8): 850 - 855. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. G. Hundley, C. A. Hamilton, G. D. Clarke, L. D. Hillis, D. M. Herrington, R. A. Lange, R. J. Applegate, M. S. Thomas, J. Payne, K. M. Link, et al. Visualization and Functional Assessment of Proximal and Middle Left Anterior Descending Coronary Stenoses in Humans With Magnetic Resonance Imaging Circulation, June 29, 1999; 99(25): 3248 - 3254. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. F Muldoon and M. H Criqui The emerging role of statins in the prevention of coronary heart disease BMJ, December 13, 1997; 315(7122): 1554 - 1555. [Full Text] |
||||
![]() |
A. Schmermund, D. Baumgart, G. Gorge, R. Seibel, D. Gronemeyer, J. Ge, M. Haude, J. Rumberger, and R. Erbel Coronary Artery Calcium in Acute Coronary Syndromes : A Comparative Study of Electron-Beam Computed Tomography, Coronary Angiography, and Intracoronary Ultrasound in Survivors of Acute Myocardial Infarction and Unstable Angina Circulation, September 2, 1997; 96(5): 1461 - 1469. [Abstract] [Full Text] |
||||
![]() |
L. Wexler, B. Brundage, J. Crouse, R. Detrano, V. Fuster, J. Maddahi, J. Rumberger, W. Stanford, R. White, and K. Taubert Coronary Artery Calcification: Pathophysiology, Epidemiology, Imaging Methods, and Clinical Implications: A Statement for Health Professionals From the American Heart Association Circulation, September 1, 1996; 94(5): 1175 - 1192. [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |