Electron Beam CT and Coronary Calcium Score
To the Editors:
I read with enthusiasm the recent article by Secci et al.1 This prospective study using electron beam CT (EBCT) and coronary calcium score was done in 326 mostly elderly (mean age, 66±8 years) men (82%). Hypertension (50%) and family history of coronary disease (44%) were common, but lipids were average (LDL 144±37 mg/dL, HDL 54±14 mg/dL). The 10-year Framingham risk was 19±9%. Each was followed up for 32 months after EBCT or until documentation of a hard (death, infarction) or a soft (need for revascularization) event. Dividing results into quartiles of EBCT calcium score, Secci et al found a clear trend for more total events in those with scores above the median. When hard and soft events were separated, there were still significantly more soft events when the score was above the median. However, despite a greater total number of events for subjects with calcium scores in the highest quartile, especially compared with those with scores in the lowest quartile, there was no significant trend for hard events alone.
My comments relate to four important issues not raised in the discussion. First, the amount of calcified plaque correlates, albeit as an underestimation, with the total atherosclerotic plaque burden (lipid-rich, fibrotic, and calcified fibrotic plaques) as shown by histological2 3 and ultrasonic4 5 studies. The discussion by Secci et al of calcified plaque and acute coronary syndromes totally misses the point regarding total plaque burden and vulnerable plaques, calcification, and inflammation.6 Second, the data7 given in their reference 16 states an average 10-year risk for men between 60 and 70 years old of 21% to 30%. This would suggest that the cohort currently under discussion was likely average and not necessarily high risk, since age and male gender are particularly important determinants using the Framingham equations. In fact, the calcium scores given in Table 3 of the Secci study (3-mm protocol) are similar to those previously published in similar age groups by Janowitz for “average” asymptomatic adults.8 Third, the study is significantly underpowered (n=326) for hard cardiac events, and from the outset, a potentially weak but not necessarily significant relationship to calcium score would have been anticipated. By combining hard and soft events, they likely improved the power and were able to demonstrate a correlation to higher calcium scores (above the median and especially in the highest quartile). Fourth, the follow-up was actually short, less than 3 years, and arbitrary at best. Follow-up of a minimum of 5 years is the common time period for most prospective studies examining cardiac events. They need to continue follow-up for a more extended time period. Rather than decrying this as such a “negative” study, I find the results provocative, consistent with prior studies9 10 11 in younger adults, and supportive of the notion that the EBCT calcium score, as a surrogate to total atherosclerotic plaque burden and thus extent of coronary disease, has the potential to become one of the most if not the most powerful noninvasive predictors of cardiac mortality and morbidity in adults at risk, irrespective of age and gender.
- Copyright © 1998 by American Heart Association
Secci A, Wong N, Tang W, Wang S, Doherty T, Detrano R. Electron beam computed tomographic coronary calcium as a predictor of coronary events: comparison of two protocols. Circulation. 1997;96:1122–1129.
Rumberger JA, Simons DB, Fitzpatrick LA, Sheedy PF, Schwartz RS. Coronary artery calcium areas by electron beam computed tomography and coronary atherosclerotic plaque area: a histopathologic correlative study. Circulation. 1995;92:2157–2162.
Baumgart D, Schmermund A, Goerge G, Haude M, Ge J, Adamzik M, Sehnert C, Altmaier K, Groenemeyer D, Seibel R, Erbel R. Comparison of electron beam computed tomography with intracoronary ultrasound and coronary angiography for detection of coronary atherosclerosis. J Am Coll Cardiol. 1997;30:57–64.
Berliner JA, Navab M, Fogelman AM, Frank JS, Demer LL, Edwards PA, Watson AD, Lusis AJ. Atherosclerosis: basic mechanisms—oxidation, inflammation, and genetics. Circulation. 1995;91:2488–2496.
Anderson K, Wilson P, Odell P, Kannel W. An updated coronary risk profile. Circulation. 1991;83:356–363.
Arad Y, Spadaro LA, Goodman K, Liedo-Perez A, Sherman S, Lerner G, Guerci AD. Predictive value of electron beam computed tomography of the coronary arteries: 19-month follow-up of 1173 asymptomatic subjects. Circulation. 1996;93:1951–1953.
Agatston AS, Janowitz WR, Kaplan GS, Lee D, Prashad R, Viamonte M, Lamas GA. Electron beam CT coronary calcium predicts future coronary events. Circulation. 1996;94(suppl I):I-360. Abstract.
Rumberger responds to our report entitled “Electron Beam Computed Tomographic Coronary Calcium as a Predictor of Coronary Events: Comparison of Two Protocols.” This report treated a preliminary study of the 326 members of our South Bay Heart Watch cohort of 1309 high-risk subjects. These 326 subjects were scanned multiple times using two different electron beam CT scanning protocols within a half-hour period in 1991. All 1309 subjects have now been followed up clinically for 3 years. The preliminary report to which Rumberger refers states three conclusions:
1. Calcium quantities from the 3-mm and more reproducible 6-mm scans are equally accurate for predicting coronary events.
2. Calcium is a weak predictor of coronary death and infarction.
3. The predictive accuracy of calcium for predicting revascularization is greater than that for predicting death or infarction.
Rumberger does not disagree with our conclusions but comments on their interpretation in our discussion. Specifically, he states that we “missed the point” regarding the monotonically increasing relationship between the amount of calcified plaque and the “total plaque burden” that he and others have noted in autopsied hearts.
We agree that the amount of calcific plaque, quantitated on radiographic studies, roughly reflects the total sum of atherosclerotic areas in segments of the coronary tree. However, what determines the probability of a coronary catastrophe is not only the amount of atherosclerosis but the propensity of individual plaque segments to rupture and collect blood elements that obstruct the arterial lumen.
The following mathematical diversion clarifies this point.
Here, P(t) denotes the probability of plaque rupture somewhere in the coronary tree at time t, and dl is an infinitesimal segment length in that tree. P(t) will be related to the plaque areas as follows:
The argument between the integral sign and dl is a product of two factors. The plaque area, A(l,t), changes with location, l, in the coronary tree and also changes with time (progresses). This plaque area is what Rumberger has found to be roughly and directly related to the amount of calcium. The second factor in the argument, p(l,t), is the plaque vulnerability function. This is the probability that plaque at location l will rupture at time t. This probability increases with the size of the lipid core.R1 It decreases with the thickness of the fibrous capR1 and probably with the amount of calcium deposited in the plaque.R2 Thus, this factor will contribute to an inverse relationship between the probability of coronary death or infarction and the measured amount of calcification.
Cities are occasionally buried by volcanic eruptions. The chance that a city will be buried depends not only on the number of nearby peaks but also on the volcanic activity of each one. Indeed, Vesuvius is the only large mountain near Pompeii.
These considerations explain why calcification, even if it correlated perfectly with the quantity of atherosclerosis, might not be a good predictor of coronary events. The relation between the amount of calcium and the probability of coronary events is much more complex than we might have desired.
Rumberger believes that as the power of our study grows with time and with the inclusion of all subjects in the cohort, we will see a relationship between coronary calcium amount and hard events. Indeed, we agree and have already seen this relationship.R3 However, the relationship remains, as Rumberger has supposed, “weak.” In fact, the discriminatory power of coronary calcium remains similar to that of the serum cholesterol, which is cheaper to obtain and can be modified.
Mann J, Davies M. Vulnerable plaque: relation of characteristics to degree of stenosis in human coronary arteries. Circulation. 1996;94:928–931.
Cheng GC, Loree HM, Kamm RD, Fishbein MC, Lee RT. Distribution of circumferential stress in ruptured and stable atherosclerotic lesions: structural analysis with histopathological correlation. Circulation. 1993;87:1179–1187.
Detrano R, Wong ND, Tang W, Doherty TM. Determining coronary event risk in asymptomatic high risk subjects: a risk factor versus an anatomic approach. Circulation. 1997;96(suppl):I-404. Abstract.