Calcium deposition in the coronary arteries is a marker of intimal atherosclerosis. Unlike calcification elsewhere in the arterial system that can be related to medial sclerosis, coronary artery calcification (CAC) is, with rare exceptions, associated only with intimal disease. This fact was recognized by Blankenhorn and Stern (1) in 1959, when they first proposed using CAC as a way to noninvasively detect coronary artery disease (CAD) by using fluoroscopy (1).
The fluoroscopic detection of CAC, despite its shortcomings, was considered clinically useful by some cardiologists and, when carefully performed, has a high specificity for CAD (2,3). Fluoroscopic detection of CAC was shown to be a powerful and independent predictor of mortality risk in a population of CAD patients undergoing coronary angiography (4). Pathologic studies, to be discussed in more detail later, confirmed the association of CAC with both the presence and extent of atherosclerosis. Though digital fluoroscopy may have some value, the use of fluoroscopy to detect CAC has become negligible in the practice of clinical cardiology. Coronary artery calcification was considered by most physicians as part of the normal aging process.
Computed tomography can detect CAC with a much higher sensitivity than fluoroscopy, though motion artifact prevented accurate and reproducible measurement of CAC by conventional CT. Electron beam computed tomography (EBCT), which became clinically available in the early 1980s, scanned the heart at speeds that eliminated most motion artifact. Initially, EBCT used a low resolution mode with an 8-mm slice thickness and 4-mm gaps between slices, which precluded accurate quantification of CAC. When a high-resolution radiology mode (3 mm) became available, EBCT became capable of accurately measuring CAC. In 1988, our group at Mount Sinai Medical Center of Miami Beach started performing high-resolution EBT of the coronary arteries, with the aim of quantifying the amount of calcium present (5,6). It was thought that quantification of calcium would have a clinical usefulness beyond that of previous studies that detected calcium qualitatively or at most semiquantitatively (7). As evidenced by the increasing interest in CAC and the large body of literature that has been published in the years since then, this view has been proved correct. Recent advances in helical computed tomography (CT) technology allow calcium quantification to be performed by using helical scanners, with results that appear to be comparable to those of EBCT studies. Cardiac imaging and CAC quantification by helical CT may become more common than EBCT studies as the major manufacturers increase their research, development, and marketing efforts in this area.
PATHOLOGIC STUDIES OF CORONARY ARTERY CALCIUM
CAC invariably is associated with intimal atherosclerosis. Calcification occurs within lipid-laden plaque by several mechanisms. De novo calcification may take place within nonruptured soft plaque. The mechanisms of calcium formation appear to be similar to the calcification that occurs within bone. A more important mechanism is the deposition of calcium within complex plaque. Recent theories of plaque progression postulate that soft lipid-laden plaque is vulnerable to rupture or ulceration. These ruptured or ulcerated plaques contain highly thrombogenic material, and clot forms at the site of the rupture. In most instances, these clots are partially lysed and partially incorporated into the plaque with the resultant formation of complex plaque that, with time, undergoes calcification. In most instances, plaque rupture and healing is an ongoing process with no clinical consequence. Over an extended period (decades), plaque builds up to the point that thrombus formation at a site of plaque rupture occludes flow in the vessel and the patient has a clinical event, which may be an infarct, angina, or, in approximately one third of patients, sudden death. Most culprit lesions occur at sites that are not severe enough to be flow-limiting. In most patients, at the time of their first event there is extensive atherosclerosis throughout the coronary arteries. Some patients may have an increase in thrombogenic risk factors that cause larger clots to be formed at the site of plaque rupture, and events can occur in them with minimal atherosclerotic changes present. Angiograms performed after thrombolytic therapy can show normal or minimal atherosclerosis in these patients.
In individual lesions, the amount of calcium increases as the severity of stenosis increases (8). Elegant studies at the Mayo Clinic have shown excellent correlation of CAC, measured by EBCT, with calcium in ashed specimens of the coronary arteries. They also have shown an excellent correlation of CAC by EBCT with severity of stenosis in 3-mm-thick pathologic specimens. It has been shown that calcified plaque represents approximately 20% of total plaque burden (9,10).
One of the largest pathology studies of coronary atherosclerosis was the International Atherosclerosis Project, published in 1968, which looked at 6,319 autopsy specimens (11). A subgroup of that study looked at 1,242 autopsy specimens in New Orleans (12). In this subgroup, specific measurements of CAC were performed by specimen radiography with densitometric measurement of the extent of coronary surface area with coronary artery calcium. Four subgroups were examined: subjects dying of coronary heart disease, those with related conditions, accidental deaths, and other natural deaths. Data for prevalence of CAC and surface area involved with CAC were examined in the different disease groups according to age. It was shown that the prevalence of CAC was very high in patients dying atherosclerotic deaths in all age-groups from 30 to 69 years. In addition, the extent of atherosclerosis, as evidenced by the surface area involved with CAC, was much higher in the atherosclerosis group than in the other groups. This was true whether a subject died in their 30s or 60s. From these data, it was apparent that the extent of atherosclerosis was an important indicator of risk of death from CAD and that patients dying of CAD at younger ages had similar amounts of atherosclerosis to older CAD patients. Younger patients had an accelerated atherosclerotic process, but no other quantitative difference.
A mathematical analysis of the data also suggested that there was a threshold of extent of atherosclerosis above which the risk of death rose significantly (13). Conversely, if the extent of atherosclerosis could be measured noninvasively, patients at a high risk could be identified, and intervention begun, before the onset of clinical disease. Because we know that coronary atherosclerosis exists for decades before the onset of clinical disease, there is a window of opportunity to detect early atherosclerosis before the patient has an event. From the Framingham data, we know that conventional risk factors, such as lipid profiles, do not detect most patients destined to develop clinical disease. The window of opportunity provided to intervene early in the coronary atherosclerotic process provided by the detection of preclinical disease has become of greater importance with the development of effective therapeutic regimens to decrease or reverse atherosclerosis. Multiple primary and secondary prevention studies have shown the effectiveness of risk factor intervention, mainly lipid-lowering therapy, to decrease coronary events whether hard or soft end points are evaluated. The problem with primary prevention has always been the selection of those subjects in whom the high cost of primary prevention can be justified. The use of CAC to select high-risk individuals for early intervention will probably be the most clinically useful application of this technology.
MEASUREMENT OF CAC BY COMPUTED TOMOGRAPHY
Measurement of CAC by CT is a fairly simple process if cardiac motion artifact can be minimized. First described for EBCT, the Agatston-Janowitz scoring system attempted to derive a calcium score (CS) that correlated with the amount of calcium in the coronary arteries. Although this scoring system was arbitrarily derived, it was standardized early in the process by the few centers performing CAC studies and was incorporated into the EBCT software and has become a de facto standard. Most published studies of CAC have used it with minimal modification. Other scoring systems have been evaluated that have shown no clinical advantage over the original CS. Recently, a volumetric-based scoring system has been proposed that has decreased interscan variability, making it useful in serial studies (14). Several different quantitative measurements may be useful, depending on the clinical question being asked: extent of disease, risk of event, and response to therapy, each of which may correlate best with different quantitative parameters. The inherently quantitative nature of CT allows various scores to be generated with little or no additional analysis time.
Although most publications concerning CAC have used EBCT, there is increasing interest in the use of helical CT to measure coronary calcium. Conventional, ungated CT at 1 second per slice is too slow to accurately quantitate CAC because of motion artifact. The speed of helical CT scanners has increased significantly in recent years. Single-slice scanners using retrospective or prospective cardiac gating and segmented reconstruction are now available that allow helical CT images to be obtained with minimal motion artifact and permit measurement of CAC (15). New multislice helical scanners using retrospective gating from several heartbeats provide effective scan speeds of approximately 100 milliseconds per slice at a slice thickness of 1.5 mm. Limited data suggest that helical CAC quantification has excellent correlation with EBCT, though there is a paucity of published CAC data using helical CT, and extrapolation of EBCT literature must be done with caution, recognizing the possibility that differences in sensitivity and accuracy may exist.
CLINICAL STUDIES OF CAC
Many studies have looked at various clinical aspects of CAC, and several recent review articles have summarized the published data (16,17). Population studies using EBCT have shown that the prevalence and extent of CAC closely track the autopsy studies of CAC discussed earlier (18). In asymptomatic subjects, the prevalence of CAC increases with age, as does the mean CS. Expected gender differences are seen, with women lagging behind men in prevalence and CS by approximately 10 years, reflecting the known lower incidence of CHD in women before menopause. There are large differences in CS scores in patients with and without known CAD. The sensitivity of CAC to detect patients with obstructive disease is approximately 95% when CS > 0 is used as the threshold for abnormal, though specificity is low at approximately 50% (19). Specificity in this situation refers to the presence of obstructive disease, and patients with CAC considered “false positive” actually have nonobstructive coronary atherosclerosis, whose extent is reflected in the magnitude of the CS, and are not normal. There has been much confusion in the cardiology community concerning the so-called low specificity of CAC measurements, and it is important that this distinction be made. Angiographic studies have shown that the prognosis for patients with nonobstructive disease is closer to that of patients with obstructive disease as opposed to patients with normal coronary arteries, who have a much better prognosis. Use of higher CS thresholds improve specificity. However, to achieve a significant improvement in specificity, the sensitivity for obstructive disease falls to approximately 85% (20). Several conclusions regarding CAC can be drawn: 1) the CS correlates with the extent of atherosclerosis (19,21); 2) severity of disease (presence of obstructive disease) correlates with CS as well as the number of vessels with coronary calcium (19); 3) a CS of 0 is associated with a very low likelihood of significant coronary stenosis and a low risk of a coronary event (16); 4) the sensitivity in the presence of obstructive disease is approximately 95% ( for CS > 0 ); and 5) the risk of a coronary event is significantly higher in patients with a high CS (22,23,24,25).
USE OF CAC SCANNING IN SYMPTOMATIC PATIENTS
A formidable array of diagnostic tests are available to detect obstructive CAD. Treadmill or pharmacologic stress testing with radionuclide imaging or echocardiography are commonly used procedures for both diagnostic and prognostic information in patients with suspected or known obstructive CAD. Years of experience and follow-up have led to considerable clinical reliance on the results of these tests. Although a respectable case can be presented showing that CAC quantification is an accurate and cost-effective predictor of obstructive CAD (16), it adds little to existing diagnostic studies unless it is used in a carefully organized diagnostic strategem. There is little proven value at this time to performing CAC studies in patients with known obstructive CAD. In symptomatic patients (patients with chest pain), without documented CAD, CAC scanning can play an important role.
The most valuable use of CAC scanning in symptomatic patients is to exclude CAD in those patients with a negative test. If sensitivity is maintained at a high level (>95%), a CS of 0 excludes CAD with an accuracy greater than any other noninvasive test. Further cardiac evaluation can be avoided in most patients with considerable cost savings. This could decrease the normal cardiac catherization rate, which in many hospitals is approximately 40%.
Using the cost-effectiveness model previously described by Patterson et al. (26), if CAC screening is used as the first diagnostic procedure, there are improvements in cost/dx and cost/ΔQual (a measure of quality of care) in patients with pretest probabilities of disease of less than 0.4, compared with a strategy of using radionuclide imaging as the first procedure (27). This finding is, however, only true in populations in which the specificity of CAC scanning is greater than 0.50. Other studies have suggested that using EBCT with a higher threshold for positivity, increasing specificity, followed by angiography, is a cost-effective protocol. (28)
USE OF CAC SCANNING IN ASYMPTOMATIC PATIENTS
Considerable controversy exists on the role of CAC scanning as a screening procedure for asymptomatic individuals. The desirability of a screening test for CAD is obvious. In ⅓ to ½ of patients, the first symptom of CAD is sudden death or myocardial infarction. Other noninvasive tests and risk factors do not identify most patients destined to develop CAD before the development of a hemodynamically significant stenosis. As opposed to risk factors such as lipid profiles, hypertension, smoking, and so forth, the presence of CAC is pathologic evidence of the atherosclerotic process in the coronary arteries. It is a reflection of the integrative effect of a lifetime of exposure to all known and unknown risk factors as well as the protective or deleterious effect of an individual's genetic makeup, of which we are largely ignorant. It is an indicator of preclinical disease and as such is a better predictor of disease. With newer therapeutic options capable of significantly altering risk factors, and clinical trials showing the ability of aggressive risk factor alteration to lower cardiovascular mortality, there are evident reasons to implement effective screening programs to detect early coronary atherosclerosis.
The requirements for an ideal screening test for coronary atherosclerosis were discussed in an editorial in the New England Journal of Medicine(29). These requirements are that the test 1) be safe and inexpensive, 2) be noninvasive or minimally invasive, 3) be reliable and reproducible, 4) correlates with the extent of atherosclerosis and 5) has a high positive and negative predictive values for clinical events. CAC scanning appears to meet these criteria according to all available data, with some qualifications, as discussed later.
Safe and Inexpensive
The only risk associated with CAC quantification is the radiation exposure associated with the CT study. Although not negligible, the risk is low and is consistent with other diagnostic studies. The exposure is the same as for chest CT, though a smaller volume is irradiated because the entire chest is not scanned. The charges for a study are approximately $300 to 500, similar to a treadmill stress test that is often used to screen for obstructive disease. At high patient volumes, charges could be considerably less because the costs of running a scanner are essentially fixed, with very few variable costs.
Noninvasive or Minimally Invasive
CAC scanning is totally noninvasive, with the only concern being that of radiation exposure.
Reliable and Reproducible
The study is extremely reliable, with visualization of the coronary arteries in close to 100% of subjects. Problems may be encountered in patients who are unable to hold their breath or obese patients, in whom image quality may be poor. Interobserver reproducibility is very high (6). Some studies have reported interscan variability of up to 30% in patients studied twice in one setting. This has primarily been a problem in patients with low CS, in whom small differences in absolute values may represent a large percentage difference, though in absolute terms there is little clinical significance between these two scores. Interscan variability is most important in studies looking at serial scans or in therapeutic intervention studies, in which small changes in plaque burden are being measured and compared. A scoring system using an interpolated volume score has less interscan variability and may be better for following serial changes in plaque burden (14). One recent study was able to show significant changes in CAC in patients aggressively treated with statins compared with untreated patients or those in whom the low-density lipoprotein level did not fall below 120 mg/dl (30). For a single measurement of CAC for assessment of CAD, the variability seen in the CS is, in most instances, not clinically important.
Correlates with the Extent of Atherosclerosis
As discussed previously, abundant data show both pathologically and angiographically that the CS closely correlates with both the extent and severity of atherosclerosis (7,9,10,19–21). The volume of calcified plaque is approximately 20% of the total plaque burden. More severely stenotic lesions have a higher percentage of calcium, and the greater the number of vessels with CAC, the higher the probability of obstructive CAD. The absence of CAC is associated with a high probability of no significant CAD, with most patients with obstructive disease and no CAC having single-vessel disease. Patients with a calcium score of greater than 400 have a high incidence (50%) of obstructive disease (31).
High Positive and Negative Predictive Value for Clinical Events
This is probably the one area where CAC screening is most vulnerable to criticism given the paucity of long-term clinical studies and lack of randomized trials. Data from several studies have looked at the cardiac event rates in asymptomatic subjects who had undergone CAC scanning (22,25). In the largest series of asymptomatic subjects reported by Arad et al. (24), followed-up for a mean of 3.6 years and using a CS greater than 160 as the cutoff, CAC scanning had a sensitivity of 80% a specificity of 85%, a negative predictive value of 0.992, and an odds ratio of 23 for predicting cardiovascular death, myocardial infarction, or revascularization. Similar values were seen in the Mount Sinai study. Predictive value for cardiac events were less in another study which followed a high-risk population of patients (32). These patients likely were already at such a high risk because of other comorbid conditions, primarily diabetes, that there was little incremental value added to risk assessment by CAC scanning (33). These investigators also used a protocol for CAC quantification that was significantly different from that used by most other centers, which has a decreased sensitivity to detect CAC.
A recently published study followed-up 632 patients for a mean of 32 ± 7 months after being scanned by EBCT (25). Twenty-seven hard events (8 cardiac deaths and 19 acute infarcts) occurred in this group, and 96% had CAC at the time of the scan. Seventy percent of the patients who had a hard event were in the top quartile in CS. Although the annualized event rate increased with increasing CS from 0.11% to those with a 0 score to 4.8% for patients with a CS over 400, more patients with events were identified by using the top quartile.
From the available data, it would appear that CAC scanning is a powerful predictor of future coronary events when compared with traditional risk factors. Given the large resources that would be required to implement a screening program directed at asymptomatic individuals, it is prudent to wait for additional studies to confirm these findings before public health recommendations advocating CAC screening are promulgated. The National Institute of Health is funding a multicenter study that uses both EBCT and helical CT to answer some of these questions. In individual patients, treatment decisions can be facilitated from a knowledge of the CAC score.
CAC scores clearly correlate with the extent and severity of atherosclerosis as well as with the risk of a future event. Patients and physicians should not be unduly alarmed by the presence of CAC and should try to place the patients' CS in context with published normative data. Extensive additional testing should be avoided. Most experienced investigators who use CAC quantification will only consider performing stress tests that look for obstructive disease in asymptomatic subjects with a CS over 400. More aggressive therapy aimed at risk factor alteration is probably justified in patients in the top quartile of CS adjusted for age and gender. Results of CAC studies also can be used as a motivating factor to get patients to comply with recommendations dealing with lifestyle changes, because an image showing CAC in a patients' heart is much more convincing than abstract discussions about preventing heart disease. A consensus report was recently published that provides recommendations concerning the integration of CAC scores into clinical practice (34). Although these recommendations are still considered preliminary, they represented the opinions of some of the most experienced investigators in the field of CAC quantification. There is little doubt that CAC quantification will become increasingly used as a tool to assess CAD. Although its exact role has not yet been determined, the ability the use both EBCT and the larger installed base of helical CT scanners should greatly expand the current use of CAC quantification.
1. Blankenhorn DH, Stern D. Calcification of the coronary arteries. Am J Roentgenol 1959; 81:772–7.
2. Loecker TH, Schwartz RS, Cotta CW, Hickman Jr. JR Fluoroscopic coronary artery calcification and associated coronary disease an asymptomatic young men. J Am Coll Cardiol 1992; 19:1167–72.
3. Detrano R, Froelicher V. A logical approach to screening for coronary artery disease. Ann Intern Med 1987; 106:456–69.
4. Margolis JR, Chen JT, Kong Y, Peter RH, Behar BS, Kisslo JA. The diagnostic and prognostic significance of coronary artery calcification: a report of 800 cases. Radiology 1980; 137:609–16.
5. Janowitz WR, Agatston AS, Zusmer N, Viamonte M. High resolution ultrafast CT of the coronary arteries: a new technique for visualization of coronary anatomy. Radiology 1988; 169 (p):345.
6. Agatston AS, Janowitz WR, Hildner F, Zusmer NR, Viamonte M, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990; 15:827–32.
7. Tannenbaum SR, Kondos GT, Veselik KE, Pendergast MR, Brundage BH, Chomka EV. Detection of calcific deposits in coronary arteries by ultrafast computed tomography and correlation with angiography. Am J Cardiol 1989; 63:870–72.
8. Kragel AH, Reddy SG, Whittes JT, Roberts WC. Morphometric analysis of the composition of atherosclerotic plaques in the four major epicardial coronary arteries in acute myocardial infarction and in sudden coronary death. Circulation 1989; 80:1747–56.
9. Simons DB, Schwartz RS, Edwards WD, et al. Noninvasive definition of anatomic coronary artery disease by ultrafast computed tomographic scanning: a quantitative pathologic comparison study. J Am Coll Cardiol 1992; 20:1118–26.
10. Rumberger JA, Simons B, Fitzpatrick LA, et al. Coronary artery calcium area by electron-beam computed tomography and coronary atherosclerotic plaque area: a histopathologic correlative study. Circulation 1995; 92:2157–62.
11. Strong JP, Solberg LA, Restrepo C. Atherosclerosis in persons with coronary heart disease. Lab Invest 1968; 18:527–37.
12. Eggen DA, Strong JP, McGill HC. Coronary calcification: relationship to clinically significant coronary lesions and race, sex and topographic distribution. Circulation 1965; 32:948–55.
13. Gofman JW. The quantitative nature of the relationship of coronary artery atherosclerosis and coronary heart disease risk. Cardiol Dig
14. Callister TQ, Cooil B, Raya SP, et al. Coronary artery disease: improved reproducibility of calcium scoring with an electron-beam CT volumetric method. Radiology 1998; 208:807–14.
15. Woodhouse CE, Janowitz WR, Viamonte Jr. M Coronary arteries: retrospective cardiac gating technique to reduce cardiac motion artifact at spiral CT. Radiology 1997; 204:566–9.
16. Wexler L, Brundage B, Crouse J, Detrano R, Fuster V, Maddahi J, Rumberger J, Stanford W, White R, Taubert K. Coronary artery calcification: pathophysiology, epidemiology, imaging methods, and clinical implications. A statement for health professionals from the American Heart Association. Circulation 1996; 94:1175–92.
17. Fiorino AS. Electron-beam computed tomography, coronary artery calcium, and evaluation of patients with coronary artery disease. Ann Intern Med 1998; 128:839–47.
18. Janowitz WR, Agatston AS, Kaplan G, Viamonte Jr. M Differences in prevalence and extent of coronary artery calcium detected by ultrafast computed tomography in asymptomatic men and women. Am J Cardiol 1993; 72:247–54.
19. Budoff MJ, Georgiou D, Brody A, et al. Ultrafast computed tomography as a diagnostic modality in the detection of coronary artery disease: a multicenter study. Circulation 1996; 93:898–904.
20. Rumberger JA, Sheedy PF, Breen JF, Schwartz RS. Electron beam computed tomographic coronary calcium score cutpoints and severity of associated angiographic lumen stenosis. J Am Coll Cardiol 1997; 29:1542–8.
21. Agatston AS, Janowitz WR, Kaplan G, Gasso J, Hildner F, Viamonte M. Ultrafast computed tomography detected coronary calcium reflects the angiographic extent of coronary arterial atherosclerosis. Am J Cardiol 1994; 74:1272–4.
22. Arad Y, Spadaro LA, Goodman K, et al. Predictive value of electron beam computed tomography of the coronary arteries: 19 month follow-up of 1173 asymptomatic subjects. Circulation 1996; 93:1951–1953.
23. Janowitz, WR, Agatston AS, Kaplan G, Viamonte Jr. M Electron beam CT of coronary calcium predicts risk of future coronary events. Am J Cardiac Imag 1996; 10(Suppl 1):9.
24. Arad Y, Spadaro LA, Goodman K, Liedo A, Sherman S, Guerci AD. 3.6 years follow-up of 1136 asymptomatic adults undergoing electron beam CT (EBT) of the coronary arteries. J Am Coll Cardiol 1998; 31(Suppl A):210A.
25. Raggi P, Callister TQ, Cooil B, et al. Identification of patients at increased risk of first unheralded acute myocardial infarction by electron-beam computed tomography. Circulation 2000; 101:850–5.
26. Patterson RE, Eisner RL, Horowitz SF. Comparison of cost-effectiveness and utility of exercise ECG, single photon emission computed tomography, positron emission tomography, and coronary angiography for diagnosis of coronary artery disease. Circulation 1995; 91:54–65.
27. Janowitz WR, Ziffer JA. Incremental value of electron beam CT prior to perfusion PET or SPECT imaging. J Nucl Med 1997; 38(Suppl):75P.
28. Rumberger JA, Behrenbeck T, Breen JF, Sheedy PF. Coronary calcification by electron beam computed tomography and obstructive coronary artery disease: a model for cost and effectiveness of diagnosis as compared with conventional cardiac testing methods. J Am Coll Cardiol 1999; 33:453–62.
29. Celermajer DS. Noninvasive detection of atherosclerosis. N Engl J Med 1998; 329:2014–5.
30. Callister TQ, Raggi P, Cooil B, Lippolis NJ, Russo DJ. Effect of HMG-CoA reductase inhibitors on coronary artery disease as assessed by electron-beam computed tomography. N Engl J Med 1998; 339:1972–8.
31. He Z, Hedrick TD, Pratt CM, Verani MS, Aquino V, Roberts R, Mahmarian JJ. Severity of coronary artery calcification by electron beam computed tomography predicts silent myocardial ischemia. Circulation 2000; 101:244–251.
32. Detrano RC, Wong ND, Doherty TM, Shavelle RM, Tang W, Ginzton LE, Budoff MJ, Narahara KA. Coronary calcium does not accurately predict near-term future coronary events in high-risk adults. Circulation 1999; 99:2633–8.
33. Guerci AD, Arad Y, Agatston A. Predictive value of EBCT scanning (letter). Circulation 1998; 97:2583–4.
34. Rumberger JA, Brundage BF, Rader D, Kondos GT. Electron beam computed tomographic coronary calcium scanning: a review and guidelines for use in asymptomatic persons. Mayo Foundation for Medical Education and Research 1999; 74:243–52.
Editor: Jeffrey S. Klein
Associate Editors: Ann Leung, MD
David Lynch, MD, Jung-Gi Im, MD
Michio Kono, MD, Charles White, MD
Guest Editor: William Stanford, M.D.