Computed tomographic (CT) imaging for coronary artery calcium (CAC) began in the late 1980s with electron beam computed tomography (EBCT), the first computed tomography with sufficient temporal resolution to image a moving target such as the heart. In 1990, Agatston et al ( ) introduced a scoring method, calculated as the product of the area of calcification by a co-factor based on the peak attenuation within the plaque (1 for attenuation of 130–199 HU, 2 for 200–299 HU; 3 for attenuation of 300–400, and 4 for attenuation >400 HU). The authors suggested that this method had acceptable inter-reader variability, and it clearly separated those with coronary artherosclerosis from those without and risk factors for it. Numerous publications followed in the next 18 years that clearly demonstrated that the Agatston score is prognostically linked with hard cardiovascular events as well as all-cause mortality ( ); furthermore, it bears a modest relationship with the presence of obstructive coronary artery disease ( ). Recent American ( ) and European guidelines ( ) endorsed CAC screening as a method to reclassify risk in patients at intermediate risk by traditional multifactorial scores such as the Framingham and Procam algorithms. However, it became obvious soon after its introduction that the Agatston score had a high interscan variability ( ). Some authors tried to obviate to this problem by acquiring thicker slices ( ), but this reduced the sensitivity of the method ( ). Others developed alternative scoring methods, such as the volume score, with reportedly better reproducibility ( ).
What are the main factors affecting CAC score reproducibility? Motion is without a doubt the most important factor and it is obviously dependent on the temporal resolution of the computed tomographic scanner. Additionally, patients’ cooperation with breathing instructions and the presence of cardiac arrhythmias contribute substantially to motion artifacts. The next two very important factors affecting reproducibility are partial volume averaging and spatial resolution of the CT scanner. With the introduction of multidetector CT (MDCT) scanners, improvements came along with new problems. Undoubtedly, MDCT has superior spatial resolution compared to EBCT, but in the majority of cases it has inferior temporal resolution. Furthermore, although EBCT scanners are produced by a single manufacturer under a single license and with identical technical specifications for all scanners sold internationally, MDCT scanners are now produced by four different manufacturers and come in “different flavors” with very different technical characteristics. The field of cardiac computed tomography has evolved in just a few years from 4- to 64-row MDCT, dual source 64-row MDCT, and soon-to-come 320-slice MDCT, although the most commonly used systems are 16- and 64-row MDCT scanners. This has created a large number of choices for the consumer, but it also reduced the uniformity of the imaging protocols. It is therefore very appropriate to wonder whether the reproducibility of CAC scores is the same with the most frequently used MDCT scanners, and, along with it, whether the radiation dose varies according to the imaging protocol and the MDCT scanner used.
Horiguchi et al did exactly that in a laboratory experiment published in this issue of the Journal ( ). The investigators used four simulated coronary plaques (1, 3, 5, and 10 mm in length) with different attenuation profiles (∼240, ∼600, and ∼1000 HU) and embedded them in plastic cylinders of 4-mm diameter to simulate coronary arteries. The plaques caused on average 80% luminal stenosis of the plastic cylinders that were ultimately loaded on a heart phantom capable of in-plane and some through-plane motion. Using both 16- and 64-row MDCT scanners and both retrospective and prospective protocols, the investigators scanned the phantom three times, with five heart rate sequences, introducing intermittent variations in heart rate. The slice thickness was set at 2.5 mm with 1.25-mm overlapping reconstruction on the retrospective protocols to reduce partial volume averaging. To account for similar exposure, the tube current time products were set to similar values for all four protocols (32.7 mAs for 64-slice prospective and retrospective scans, 33.3 mAs for 16-slice prospective and retrospective scans). Electrocardiographic modulation technique during retrospective gating to control tube current was available only for the 64-row MDCT. Overall, CAC scores (Agatston, volume, or mass score) were not different between the four protocols. However, the retrospective electrocardiographically-gated protocol with 64-slice MDCT yielded the lowest score variability (11%) along with the highest radiation exposure (3.7 mSv) despite dose modulation, whereas prospective gating with 16-slice MDCT showed the highest score variability (20%) with six times lower radiation exposure (0.6 mSv). Interestingly, prospective electrocardiographically-gated 64-row MDCT showed acceptable Agatston score variability comparable to that of retrospective electrocardiographically-gated 16-row MDCT (16% for each) and a very low radiation dose (0.5 mSv compared to 3.5 mSv for the retrospective 16-slice MDCT protocol).
To try and put these findings into perspective, one would have to look at the universe of CAC imaging today. The number of systems being sold is very large, and the number of cardiac computed tomographies being performed is increasing. Whether sequential CT imaging for assessment of CAC score change is appropriate in clinical practice remains questionable, because randomized trials have not shown that currently utilized medical therapies slow CAC progression ( ), with the exception of patients with end-stage renal disease ( ). Furthermore, as demonstrated in the work by Horiguchi et al ( ), the radiation exposure can be substantial with protocols that provide more accurate CAC scores. Nonetheless, continued progression of CAC has being linked with an unfavorable outcome in several reports ( ). Additionally, several companies are interested in assessing the effect of investigational drugs for atherosclerosis with computed tomography, and CAC appears to be an obvious and easy target of imaging. In view of Horiguchi’s article ( ), therefore, how should we advise the potentially interested party? What generation MDCT scanners should be included in a multicenter study? What protocol should be used to reduce the “cost” of radiation exposure? Retrospective or prospective? Should we advise the use of β blockers to maximize the advantage of dose modulation while perfecting image acquisition? I think it would behoove us, from my reading of this research report, to sacrifice some score reproducibility (from 11% to 16%) to reduce the radiation dose substantially (from 3.7 to 0.5 mSv) by advising scanning with 64-slice MDCT and prospective gating. The risk of excessive radiation exposure, even with electrocardiographic modulation, may not justify the pursuit of high score reproducibility employing retrospective gating on either 64- or 16-row MDCT. For research purposes, prospective triggering with 16-slice MDCT appears to be the least desirable due to high interscan variability despite a very low radiation dose, whereas it may be acceptable in the clinical setting. Of paramount importance, a research scan should be repeated with the same equipment and the same imaging protocol and parameters to obtain follow-up data that are reliable as possible. Nonetheless, there were obvious limitations to the report by Horiguchi et al ( ) that do not allow complete confidence in issuing an opinion on sequential MDCT imaging: this was an in vitro study and the coronary arteries were simulated with 4-mm diameter plastic cylinders; actual coronary arteries can vary in diameter from greater than 5 to smaller than 2 mm and they have a diminishing diameter as they grow centrifugally from the ostia. The human heart moves simultaneously in several directions in space, whereas the main direction of motion in the model used in the experiment was in-plane. Furthermore, patients do tend to involuntarily move and breathe during imaging. Although the investigators used plaques of different length, the plaques caused approximately 80% stenosis of the lumen of the artificial coronary arteries and this is not the typical situation in native coronary arteries showing calcification. The investigators chose to reconstruct images or to time the prospective gating at 80% of the RR interval, and this may not be optimal for all heart rates. Scatter and image noise is highly dependent on patients’ body size and body composition, factors that could not have been modeled in this study. All of these factors and others will affect the reproducibility of CAC scores when sequential CT imaging is applied to assess change in CAC over time. In the end, a conscientious physician will have to weigh quality (high reproducibility) versus risk (radiation exposure) and reach a compromise as the perfect solution to this dilemma is far from being at hand. Sobering as it is, the advisability of sequential CAC scanning in clinical practice remains to be demonstrated.
References
1. Agatston A.S., Janowitz W.R., Hildner F.J., Zusmer N.R., Viamonte M., Detrano R.: Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990; 15: pp. 827-832.
2. Raggi P., Shaw L.J.: Epidemiologic guidance with coronary artery calcium scoring. Curr Cardiol Rep 2008; 10: pp. 60-66.
3. Budoff M.J., Diamond G.A., Raggi P., et. al.: Continuous probabilistic prediction of angiographically significant coronary artery disease using electron beam tomography. Circulation 2002; 105: pp. 1791-1796.
4. Budoff M.J., Achenbach S., Blumenthal R.S., et. al.: Assessment of coronary artery disease by cardiac computed tomography: a scientific statement from the American Heart Association Committee on Cardiovascular Imaging and Intervention, Council on Cardiovascular Radiology and Intervention, and Committee on Cardiac Imaging, Council on Clinical Cardiology. Circulation 2006; 114: pp. 1761-1791.
5. Oudkerk M., Stillman A.E., Halliburton S.S., et. al.: Coronary artery calcium screening: current status and recommendations from the European Society of Cardiac Radiology and North American Society for Cardiovascular Imaging. Int J Cardiovasc Imaging 2008; in press
6. Wang S., Detrano R.C., Secci A., et. al.: Detection of coronary calcification with electron-beam computed tomography: evaluation of interexamination reproducibility and comparison of three image-acquisition protocols. Am Heart J 1996; 132: pp. 550-558.
7. Wexler L., Brundage B., Crouse J., et. al.: Coronary artery calcification: pathophysiology, epidemiology, imaging methods, and clinical implications. Circulation 1996; 94: pp. 1175-1192.
8. Callister T., Janowitz W., Raggi P.: Sensitivity of two electron beam tomography protocols for the detection and quantification of coronary artery calcium. AJR Am J Roentgenol 2000; 175: pp. 1743-1746.
9. Callister T.Q., Cooil B., Raya S.P., Lippolis N.J., Russo D.J., Raggi P.: Coronary artery disease: improved reproducibility of calcium scoring with an electron-beam CT volumetric method. Radiology 1998; 208: pp. 807-814.
10. Horiguchi J K.M., Fujioka C., et. al.: Variability of repeated coronary artery calcium scoring and radiation dose on 64-slice and 16-slice CT by prospective electrocardiographically-triggered axial and retrospective electrocardiographically-gated spiral CT: A phantom study. Acad Radiol 2008;
11. Raggi P., Davidson M., Callister T.Q., et. al.: Aggressive versus moderate lipid-lowering therapy in hypercholesterolemic postmenopausal women: Beyond Endorsed Lipid Lowering with EBT Scanning (BELLES). Circulation 2005; 112: pp. 563-571.
12. Schmermund A., Achenbach S., Budde T., et. al.: Effect of intensive versus standard lipid-lowering treatment with atorvastatin on the progression of calcified coronary atherosclerosis over 12 months: a multicenter, randomized, double-blind trial. Circulation 2006; 113: pp. 427-437.
13. Block G.A., Spiegel D.M., Ehrlich J., et. al.: Effects of sevelamer and calcium on coronary artery calcification in patients new to hemodialysis. Kidney Int 2005; 68: pp. 1815-1824.
14. Chertow G.M., Burke S.K., Raggi P.: Sevelamer attenuates the progression of coronary and aortic calcification in hemodialysis patients. Kidney Int 2002; 62: pp. 245-252.
15. Raggi P., Callister T.Q., Shaw L.J.: Progression of coronary artery calcium and risk of first myocardial infarction in patients receiving cholesterol-lowering therapy. Arterioscler Thromb Vasc Biol 2004; 24: pp. 1272-1277.
16. Raggi P., Cooil B., Ratti C., Callister T.Q., Budoff M.: Progression of coronary artery calcium and occurrence of myocardial infarction in patients with and without diabetes mellitus. Hypertension 2005; 46: pp. 238-243.