| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Commentary |
1 School of Medicine, University of California, San Diego, UCSD Center for
Functional MRI, 9500 Gilman Dr., Mail Code 0677, La Jolla, CA
92093-0677.
2 Department of Radiology, University of Washington and Harborview Medical
Center, Seattle, WA 98104.
Received January 8, 2007; accepted after revision May 23, 2007.
Address correspondence to O. Leontiev.
OBJECTIVE. The purpose of this article is to examine the value of coronary artery calcium scoring as a screening tool for identifying individuals at risk of experiencing a catastrophic coronary event.
CONCLUSION. Because of the noninvasive nature of CT and the large societal burden of coronary atherosclerosis, there is great interest in developing CT-based techniques for detection of coronary artery calcium, a known marker of underlying atherosclerosis. However, important questions remain including the role of calcium in the process of acute plaque rupture, the utility of this technique in the asymptomatic population, and the statistical distribution of calcium scores and hard cardiac events in the general population.
Keywords: calcium scoring • coronary artery calcium scoring • coronary artery disease • CT
Since the introduction of electronbeam CT (EBCT) and the Agatston scoring system, coronary calcium scoring has become an increasingly popular test for cardiovascular risk stratification. Several population studies have shown that coronary artery calcium is a good predictor of cardiovascular events (acute myocardial infarction [MI], sudden death, coronary revascularization) that may factor independently from cardiac risk factors described in the Framingham study [1–3]. When conventional epidemiologic measures such as relative risk and odds ratio are used to identify sectors of the population at risk, the results of these studies suggest that there is a graded relation between the extent of coronary artery calcium and the incidence of future cardiovascular events [4, 5].
However promising, certain key issues remain, preventing the widespread application of coronary artery calcium scoring as a screening tool for identifying individuals at risk of experiencing a catastrophic coronary event. The majority of acute coronary events occur in the setting of a fairly unpredictable sequence of events leading to thrombus formation and acute plaque rupture. These generally are arteriographically mild coronary lesions that undergo significant progression to severe stenosis or total occlusion over a period of a few months [6]. This consensus has suggested the fearful notion that nonstenotic, hemodynamically insignificant plaques may rupture, precipitating occlusive thrombus, myo cardial infarction, or death, often in asymptomatic patients. For this reason, it is imperative that the utility of calcium scoring be properly evaluated in the asymptomatic, low- to medium-risk population.
The role of calcium in the process of acute plaque rupture is an active area of research and has not been sufficiently elucidated by the scientific community. It is well established that coronary artery calcium is well correlated with total plaque volume or atherosclerotic burden [7]. This alone can explain the high risk of developing a cardiac event associated with a high calcium score. However, calcium is found infrequently in the culprit lesions of ruptured plaques. In a postmortem study of human aortas, calcium detected by CT was not significantly related to histologic evidence of plaque rupture [8]. Shemesh et al. [9] in 2003 studied underlying calcified atherosclerotic lesions in acute (acute MI, unstable angina, acute ischemia, sudden cardiac death) versus chronic (severe stable angina, angiographically identified disease, disease requiring angioplasty, disease requiring bypass surgery) coronary events using dual-sector helical CT. Acute coronary events were found to be associated with a significantly lower coronary artery calcium score than chronic events (906 chronic vs 63 acute and scored according to Shemesh et al. [10]).
Similarly, Detrano et al. [11] measured a low coronary artery calcium score (mean, 44) in asymptomatic subjects with risk factors compared with a mean coronary artery calcium score of 335 in symptomatic subjects. Teleologically, it can be hypothesized that coronary artery calcium is a process that exists to protect threatened myocardium by strengthening weakened atherosclerotic plaques. From a biomechanical standpoint, calcified lesions are stiffer than cellular lesions and are more resistant to rupture [12], and the results of the Familial Atherosclerosis Treatment Study (FATS) indicate that the primary benefit of lipid-lowering drugs is stabilization of existing atherosclerotic lesions by lowering their lipid content [13]. Likewise, for carotid sonography studies, it has been shown that asymptomatic patients with detectable calcium are much less likely to suffer from cerebrovascular events than those without carotid calcium deposits [14]. A common conclusion drawn from these studies is that although calcification is a feature of advanced obstructive atherosclerosis, it cannot be said that calcium per se contributes to the development of acute coronary syndromes.
Many earlier studies supporting the use of coronary artery calcium scoring
report substantially higher levels of risk of morbidity or mortality
associated with higher calcium scores compared with lower coronary artery
calcium scores [1,
2,
4]. This risk is typically
quantified by such conventional measures as relative risk (risk of an event
relative to exposure) or odds ratio (the odds of an event occurring in one
group to the odds of it occurring in another group). Because both calculations
are sensitive to the total number of individuals in the groups of calcium
score being compared and because only a very small portion of the population
shows a high calcium score, the high relative risk and odds ratio for
individuals in the upper percentiles of coronary artery calcium score in many
prospective population studies can be explained simply by the fact that
coronary events are evenly distributed across ranges of calcium scores
[15]. In Wong et al.
[3], one third of hard coronary
events occurred in those with mild coronary artery calcium (
80). In
Doherty et al. [16], 52% of
coronary events occurred in patients in the bottom two thirds of calcium
scores. Secci et al. [17]
found that hard coronary events were not significantly more frequent in higher
quartiles of coronary artery calcium score. An important conclusion of these
findings is that an individual suffering a coronary event has a roughly equal
probability of belonging to any coronary artery calcium score quartile.
A critical question for any screening test is what is the distribution of the measured variable in the population? For the data available in the asymptomatic population, the distribution of calcium scores is clearly nongaussian and varies considerably between different studies and between men and women. Also, the distributions are significantly skewed toward lower calcium scores, have large SD values, and show a strong disparity between mean and median values. For example, in Kondos et al. [4] the mean calcium score was 137 ± 376 in men (59 ± 263 in women), whereas the median score was < 10 for both sexes. In Raggi et al. [18] the mean calcium score across both sexes was 101 ± 254, whereas the median was 3.1. Both studies used EBCT, used the calcium scoring method described by Agatston et al. [19], and included a cohort of asymptomatic low- to intermediate-risk patients.
Therefore, because of the variability of reported data and the large variance of calcium scores in the asymptomatic population, a clear definition of a coronary artery calcium score threshold for application to the general population is problematic. For a cutoff threshold in the 100–200 range of Agatston score, most population studies have found that although the negative predictive value is very high (0.95–0.99), the positive predictive value is rather low (0.02–0.13) [2, 18, 20]. Furthermore, the values reported for calculated risk in many studies have to be interpreted with caution. Most population studies use coronary revascularization as a measurable end point. However, the degree to which therapeutic intervention is guided by calcium score remains unknown.
An important prerequisite for the application of any screening test is that the measurement error should be exceeded by the natural variance in the general population. To that end, high reproducibility is a key requirement for coronary artery calcium scoring because the application has been suggested to monitor the progression of atherosclerotic plaque burden. In early EBCT reproducibility studies, the reported interexamination variability was found to exceed the normal progression of coronary artery calcium scores per year or the accelerated progression seen with significant coronary disease with considerable loss of reproducibility for patient subgroups with low to mild calcification [21]. However, more recent studies using early diastolic or end-systolic triggering show improved interscan variability [22]. Retrospective gating with MDCT, however, suffers from higher radiation exposure to patients and feasibility issues including poorer interreviewer reproducibility and markedly increased interpretation times. Moreover, significant interscan variability of low coronary artery calcium scores continues to be an important problem despite advancements in cardiac gating techniques [23].
The use of CT-based coronary artery calcium scoring techniques as a general screening tool for identification of individuals at risk for acute coronary events, although promising, is hampered by limitations in scientific backing, adequate reproducibility of low calcium scores, epidemiological and statistical considerations, and conclusive data in the asymptomatic population. The population data, so far, support the idea that patients with greater amounts of coronary calcification are more likely to suffer coronary events compared with patients without calcification or lesser amounts. However, these results may simply reflect a skewed distribution of coronary artery calcium scores in the general population and do not necessarily imply a causal relationship between the extent of calcification and the process of acute plaque rupture. Moreover, the paucity of data in women and members of ethnic minorities, the majority of whom have been shown to have lower calcium scores, may reduce the conclusiveness of these findings. With advances in CT technology and increased availability of thin-slice CT, reproducibility in the low range of coronary artery calcium scores is likely to improve over time. Until then, the prediction of unstable angina and acute coronary syndromes will likely continue to be unsuccessful when based on calcium screening alone.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Rademaker, H. Schoder, N. S. Ariaratnam, H. W. Strauss, J. Yahalom, R. Steingart, and K. C. Oeffinger Coronary Artery Disease After Radiation Therapy for Hodgkin's Lymphoma: Coronary CT Angiography Findings and Calcium Scores in Nine Asymptomatic Patients Am. J. Roentgenol., July 1, 2008; 191(1): 32 - 37. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E J HALL and D J BRENNER Cancer risks from diagnostic radiology Br. J. Radiol., May 1, 2008; 81(965): 362 - 378. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
