Author + information
- Received May 10, 2001
- Revision received October 15, 2001
- Accepted November 1, 2001
- Published online February 6, 2002.
- Matthew J Budoff, MD∗,* (, )
- Ted P Yang, MD∗,
- Robert M Shavelle, PhD, MBA†,
- Daniel H Lamont, MBA∗ and
- Bruce H Brundage, MD‡
- ↵*Reprint requests and correspondence:
Dr. Matthew Budoff, Division of Cardiology, Harbor-UCLA Research and Education Institute, 1124 West Carson Street, Bldg. RB-2, Torrance, California 90502-2064, USA.
Objectives The study was done to evaluate whether ethnic differences exist in the prevalence of coronary artery calcification (CAC), and to determine whether differences in calcification correlate with the degree of coronary obstruction.
Background Electron beam tomography (EBT) can be used to quantitate the amount of CAC and assist in prognostication of future cardiac events. It is unclear whether ethnic differences in coronary mortality are related to differences in the prevalence of coronary obstruction and CAC.
Methods A total of 782 symptomatic subjects underwent both EBT and angiography. A 50% luminal narrowing defined an angiographic obstruction.
Results We observed substantial ethnic differences in prevalence of both CAC and angiographic stenosis. In whites (n = 453), prevalence of CAC (score >0) was 84%, and significant obstruction on angiogram was 71%. Compared with whites, blacks (n = 108) had a significantly lower prevalence of CAC (62%, p < 0.001) and angiographic disease (49%, p < 0.01). Hispanics (n = 177) also had a lower prevalence of CAC (71%, p < 0.001) and angiographic obstruction (58%, p < 0.01). Asians (n = 44) were not significantly different in regard to CAC (73%, p = 0.06) or angiographic stenosis (64%, p = 0.30). These ethnic differences remained after simultaneously controlling (by use of multiple logistic regression) for age, gender and cardiac risk factors.
Conclusions As compared with whites, blacks and Hispanics had significantly lower prevalence of CAC and obstructive coronary disease. Ethnic differences in risk-factor profiles do not explain these differences. This study demonstrated that whites have a higher atherosclerotic burden than blacks and Hispanics, independent of risk-factor differences among symptomatic patients referred for angiography.
Coronary artery disease (CAD) is a major cause of mortality and morbidity in all ethnic and racial subsets (1,2). Often, the first indication of coronary atherosclerosis is acute myocardial infarction (MI) or sudden coronary heart death. Identification of those subjects with a very high probability of suffering coronary death and/or MI, therefore, is of great interest, particularly because coronary disease is an increasingly treatable pathologic entity in both its early and later stages. Both electron beam tomography (EBT) and measurement of coronary artery calcification (CAC) have been found useful as sensitive markers of atherosclerosis (3,4)and are currently used as methods for early detection of coronary disease.
However, prevalence of coronary disease varies among different ethnicities (5–7). This variability may be due in part to ethnic differences in the prevalence of coronary atherosclerosis and its risk factors (8). We studied 782 symptomatic patients using angiography and EBT to examine whether a difference in the prevalence of CAC exists among different ethnicities.
The study sample consisted of 782 patients undergoing coronary angiography for clinical indications, the vast majority of whom were suspected to have CAD (angina). Patients underwent EBT scanning within three months of angiography to evaluate for CACs as part of the research protocol to compare atherosclerosis in patients undergoing angiography. Patients were excluded from the study when EBT scans were performed more than three months after the angiogram (n = 4). All subjects signed a written consent form approved by the Institutional Review Board at Harbor-UCLA Medical Center prior to entry into the study or performing any part of the study protocol.
The coronary angiograms were analyzed by an experienced reader without knowledge of the EBT results. Each coronary vessel (left main, left anterior descending, circumflex and right coronary artery) was assessed and the visual estimation of the percent luminal reduction for each lesion was reported. Multiple projections were acquired to discern the maximal coronary artery luminal narrowing. Investigators recorded the maximum stenosis in each vessel in one of five categories: none, luminal irregularities (<50% stenosis), 50% to 75%, 75% to 99% or 100% occlusion. A 50% luminal narrowing in any coronary vessel defined a significant angiographic obstruction. A medical resident reviewed medical records of the visit before angiography for the presence or absence of the following risk factors: age, gender, hypertension, high cholesterol, diabetes, tobacco use or family history of premature CAD.
A total of 782 patients at Harbor-UCLA Medical Center underwent EBT using an Imatron C-150XL Ultrafast computed tomographic scanner (Imatron Inc., South San Francisco, California). Coronary artery visualization was obtained without contrast medium injection, and 30 consecutive images were obtained at 3-mm intervals beginning 1 cm below the carina and progressing caudally to include the proximal and midcoronary arteries. Exposure time was 100 ms/image slice, and total skin irradiation dose was <600 mrad/scan. Electrocardiographic (ECG) triggering was used and adjusted such that image acquisition occurred at the same point in the cardiac cycle during diastole, at 80% of the RR interval. Proximal coronary artery visualization was obtained without contrast medium injection, and 30 consecutive images were obtained at 3-mm intervals beginning 1 cm below the carina and progressing caudally to include the proximal coronary arteries.
A computed tomographic threshold of 2 pixels and 130 Hounsfield units (Hu) was utilized for identification of a calcific lesion. Each focus exceeding the minimum criteria was scored using the algorithm developed by Agatston et al. (9)and calculated by multiplying the lesion area by a density factor derived from the maximal Hu within this area. The density factor was assigned in the following manner: 1 for lesions whose maximal densities were 130 to 199 Hu, 2 for lesions 200 to 299 Hu, 3 for lesions 300 to 399 Hu and 4 for lesions >400 Hu. A total coronary calcium score (CCS) was determined by summing individual lesion scores from each of four anatomic sites (left main, left anterior descending, circumflex and right coronary arteries). The EBT scoring was performed by a cardiologist blinded to the clinical, ECG and angiographic information.
Categorical data are presented as number (percent), and continuous data as mean value ± SD. Either the Fisher exact test or the chi-square test analysis was used to compare categorical variables. All tests of significance were two-tailed, and significance was defined at the ≤0.05 level. Sensitivity, specificity and predictive values were calculated using standard formulas. Accuracy was defined as the number of true positives plus true negatives divided by total number. Comparisons of continuous variables between two groups were performed using two-tailed ttests. Data were analyzed using SAS for PC (Version 6.12, SAS Institute, Cary, North Carolina). Multiple linear regression and multiple logistic regression analyses were performed. Coronary calcium scores were transformed by taking the natural log of (1 + CCS) for inclusion in a secondary analysis.
Hypertension, tobacco use, diabetes, age, gender, family history of premature heart disease, hypercholesterolemia and the utilization of statin therapy were evaluated as independent variables using one-way analysis of variance. Multivariate logistic regression analysis was used to develop statistical models for measuring the presence of both CAC and obstructive CAD, based upon ethnicity. These analyses were performed to adjust for the possible interaction of any individual risk factor.
A total of 782 patients were evaluated by EBT and angiography. There were 453 white patients, 108 black patients, 177 Hispanics and 44 Asian-Americans. Of the patients enrolled, 69% were men, 505 (65%) had obstructive coronary disease and 606 (78%) had detectable CAC. Table 1shows the average age and the distribution of coronary risk factors by race. No significant differences existed in the prevalence of any risk factor for any ethnic group.
Substantial ethnic differences in prevalence of both CAC and angiographic stenosis were noted (Table 2). Prevalence of atherosclerosis in whites (n = 453) by CAC was 84% and by obstruction on angiogram was 71%. As compared with whites, black patients (n = 108) had a significantly lower prevalence of CAC (62%, p < 0.001) and angiographic disease (49%, p < 0.01). Hispanics (n = 177) also had a lower prevalence of CAC (71%, p < 0.001) and angiographic obstruction (58%, p < 0.01). Asians had less CAC and angiographic stenosis than whites, but these differences were not statistically significant (73%, p = 0.06 and 64%, p = 0.30, respectively). Controlling for age, gender and risk factors, white patients had a higher risk of angiographic disease (odds ratio [OR], 1.65) than all other races separately or combined (p = 0.004). Compared with whites, blacks had 0.48 times the risk (p = 0.0002), Hispanics 0.68 (p = 0.07) and Asians 0.66 (p = 0.25) (Table 3). The same conclusions are reached if the EBT score is included in the model, if risk factors are excluded or if the model contains only race variables.
Controlling for age, gender and risk factors, whites had a higher average EBT score (using the transformed variable ln[score + 1]) than the other races (p = 0.0001). Blacks, Hispanics and Asians had average scores not significantly different from one another (p > 0.10), but significantly different from whites (p = 0.002, 0.0005 and 0.04, respectively). Ethnic differences were also found in CAC by using multivariate logistic regression, with CAC as the dependent variable, and age, gender and other cardiac risk factors as independent variables (Table 4).
The sensitivity of EBT-determined CAC, compared with angiography in detecting obstructive coronary disease, ranged from 89% to 98% among the different racial groups (Table 5). Specificity of EBT CAC ranged from 50% to 69%. No significant differences existed in sensitivity or specificity of EBT for detecting obstructive angiographic disease among any of the ethnic groups (Table 5).
In symptomatic subjects, we found significant ethnic differences in the prevalence of CAC. Whites and Asian-Americans had a significantly higher prevalence of CAC when compared with blacks and Hispanics. Because this difference in prevalence may depend on differences in risk factors between different ethnic groups, we used logistic regression to control for differences in cardiac risk factors. These differences persisted even after controlling for standard cardiac risk factors. Whites not only had a higher prevalence of CAC but the mean CCSs were significantly higher than the other races. The higher prevalence and degree of calcification correlated with a higher prevalence of obstruction on angiography.
Racial differences in coronary calcium
Racial differences in CAC have been found in autopsy studies. On 1,242 necropsies performed in Louisiana in 1963, whites were found to have a higher prevalence of CAC when compared with blacks (10,11). Other studies have demonstrated more severe angiographic disease in whites. In a postmortem study of 169 hearts, a distinctly higher proportion of hearts having three or four vessels with 75% stenotic lesions was observed in white men (70%) compared with white women (34%). In contrast, 58% of the hearts of both black men and black women had three or four vessels with ≥75% stenoses (12). Radiographically, whites have also been found to have a higher prevalence of CAC as detected by fluoroscopy (13). These studies demonstrated black race to be a significant, independent and inverse predictor of the presence of coronary calcium (OR, 0.47; 95% confidence interval [CI], 0.29, 0.75).
We similarly documented a lower prevalence of coronary calcium among black patients (p < 0.001 as compared with whites). However, our study is the first to document this difference in both coronary calcium and angiographic burden. We showed that the lower prevalence of coronary calcium is probably entirely due to a lower prevalence of obstructive disease. After multivariate analysis, blacks had 0.48 times the risk of obstructive disease as compared with whites (p = 0.003).
Many epidemiologic reports also suggest that the risks of cardiovascular disease differ between blacks and whites (14). However, results are controversial and conflicting. Some studies (1)show that blacks are protected against coronary heart disease but have a much greater risk of stroke, whereas others (such as the Atherosclerosis Risk in Communities Study) show that because risk factors such as diabetes and hypertension seem to be more prevalent among blacks than among whites (8), blacks were not at higher risk for coronary disease independent of these risk factors (6).
The Charleston Heart Study, begun in 1960, is the largest epidemiologic study used to identify differences in cardiovascular disease in black and white subjects. In 30 years of follow-up, this study showed no significant racial differences in the rate ratios for death from coronary disease (7). However, data from studies in Charleston, South Carolina (7), and Evans County, Georgia, indicated that black men experience lower coronary death rates than white men (15). Furthermore, data from the First National Health and Nutritional Examination Survey (16), and pooled follow-up data from both the first and second survey (17), showed that total incidence of coronary disease and coronary death were lower in black men as compared with white men. Furthermore, the Multiple Risk Factor Intervention Trial also demonstrated adjusted relative risk of coronary heart disease death for blacks compared with whites to be 0.71 (95% CI, 0.53 to 0.95) (18). Thus, increased CAC in whites may translate into increased cardiovascular mortality. Contrary to this observation of increased coronary heart disease incidence and mortality in whites, sudden death due to arrhythmia has been shown to be a more important factor in blacks than in whites (19). Recent large studies demonstrated excess risk of cardiac arrest and sudden cardiac death in blacks compared with whites (19–21).
One other study (22)of coronary calcium and ethnicity has been reported, demonstrating a lower prevalence of coronary calcium in blacks, yet a similar or higher coronary event rate. The investigators postulated that differences in the pathobiology of atherosclerosis and calcium metabolism between blacks and whites were responsible for this finding. However, there was no angiography performed or other measure of disease in that trial (22), and the numbers of black subjects were excessively small.
Our study, by use of angiography, suggests that the lower calcium prevalence is due to the lower prevalence of coronary atherosclerosis in blacks. We have demonstrated that white patients were more likely to have significant CAD than black patients. The finding in our study is similar to other studies of angiographic disease prevalence rates between the races (23), and it supports the hypothesis that the lower CCSs parallel the lower angiographic prevalence of disease in blacks. Increased myocardial mass, left ventricular hypertrophy and more severe hypertension without atherosclerosis most likely explains the higher black mortality in a study performed by Doherty et al. (22). This high prevalence of left ventricular hypertrophy and severe hypertension most likely explains why the findings of that study diverges significantly from the published literature in regards to cardiac event rates and calcium scores (3,24–27). Larger prospective studies of the natural history and progression of subclinical atherosclerosis in different ethnic groups (such as the Multi-Ethnic Study of Atherosclerosis) will provide more definitive answers to some of these questions.
Coronary events and race
Census results (28)indicate that Hispanics have lower mortality rates from coronary disease when compared with whites, and Asian-Americans have been found to have similar cardiac mortality rates as whites (29). These findings correlated well with our findings of less angiographic disease and CAC in Hispanics, and similar angiographic disease in Asian-Americans, when compared with whites.
The sensitivity of EBT in detecting significant angiographic stenosis was 89% to 98% and did not differ among the different racial groups. Specificity of EBT for obstructive disease ranged from 50% to 69% (Table 5). These values are similar to previous studies on the sensitivity and specificity of EBT in detecting significant angiographic stenosis (30). Detection of CAC in symptomatic subjects is a sensitive marker for atherosclerotic coronary disease (31). Our study showed that whites and Asian-Americans have a significantly higher prevalence of CAC and angiographic disease than blacks or Hispanics.
Our study, like other studies of symptomatic persons who had their CAD status verified using coronary angiography, is subject to verification bias. This verification bias is a consequence of the preferential referral of positive test responders to angiography and of negative test responders away from angiography, resulting in an overestimation of sensitivity and an underestimation of specificity (32). Also, this was a study of symptomatic individuals; thus, any extrapolation to the asymptomatic population cannot be made. We purposely performed a single-center study to minimize selection bias as much as possible. Our facility has a single group of cardiologists referring for angiography; thus, our study was not biased by socioeconomic status differences as our hospital is a county facility that generally serves a low socioeconomic class. Other study limitations could include: possible bias due to exclusion of cases; missing data or measurement error; possible confounding by variables not controlled (as this was an observational study); possible chance findings due to multiple comparisons and wide confidence intervals within some strata; and limitations of generalizability to asymptomatic cohorts.
In a symptomatic population, whites and Asian-Americans have a higher burden of atherosclerosis, both angiographically and by EBT, when compared with blacks and Hispanics. The sensitivity and specificity of EBT in detecting significant angiographic obstruction did not differ among different ethnic groups despite a different prevalence in CAC. The EBT-detected CAC correlated well with extent and prevalence of angiographic CAD across all ethnicities, demonstrating a considerable similarity in calcifying stenotic lesions.
- coronary artery calcification
- coronary artery disease
- coronary calcium score
- confidence interval
- electron beam tomography
- Hounsfield units
- myocardial infarction
- odds ratio
- Received May 10, 2001.
- Revision received October 15, 2001.
- Accepted November 1, 2001.
- American College of Cardiology
- American Heart Association
- Arad Y.,
- Spadaro L.A.,
- Goodman K.,
- et al.
- Budoff M.,
- Georgiou D.,
- Brody A.,
- et al.
- Hutchinson R.G.,
- Romm F.,
- Watson R.L.,
- et al.
- Agatston A.S.,
- Janowitz W.R.,
- Hildner F.J.,
- et al.
- Strong J.P.,
- McGill H.C.
- Perper J.A.,
- Kuller L.H.,
- Cooper M.
- Gillum R.F.
- Vaccaro O.,
- Stamler J.,
- Neaton J.D.
- Liao Y.,
- Cooper R.S.,
- McGee D.L.,
- et al.
- Doherty T.M.,
- Tang W.,
- Detrano R.C.
- Raggi P.,
- Callister T.Q.,
- Cooil B.,
- et al.
- Arad Y.,
- Sparado L.A.,
- Goodman K.,
- et al.
- Detrano R.,
- Tzung H.,
- Wang S.,
- et al.
- ↵(1997) Mortality from coronary heart disease and cardiovascular disease among adult U.S. Hispanics: findings from the National Health Interview Survey. Atherosclerosis 30:1200–1204.
- ↵Marmot MG, Syme SL, Kagan A. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii, and California: prevalence of coronary and hypertensive heart disease and associated risk factors. Am J Epidemiol 102;6:514–25
- Wexler L.,
- Brundage B.H.,
- Crouse J.,
- et al.
- O’Rourke R.A.,
- Brundage B.H.,
- Froelicher V.F.,
- et al.