- Email: [email protected]
Conventional Coronary Artery Disease Risk Factors and Coronary Artery Calcium Detected by Electron Beam Tomography in 30,908 Healthy Individuals
Conventional Coronary Artery Disease Risk Factors and Coronary Artery Calcium Detected by Electron Beam Tomography in 30,908 Healthy Individuals JULIE ANNE HOFF, PHD, RN, MARTHA L. DAVIGLUS, MD, PHD, EVA V. CHOMKA, MD, ANDREW J. KRAINIK, MPH, ALEXANDER SEVRUKOV, MD, AND GEORGE T. KONDOS, MD
PURPOSE: Electron beam tomography (EBT) is a noninvasive measure of coronary artery calcium (CAC), a marker for atherosclerosis. In this study we examined the association between conventional risk factors for coronary artery disease (CAD) and CAC. METHODS: EBT CAC screening was performed on 30,908 asymptomatic individuals aged 30 to 90 years. Prior to EBT screening, individuals provided demographic and CAD risk factor information. EBT utilized a C-100 EBT scanner, and the amount of CAC was determined using the Agatston scoring method. RESULTS: The results of this study demonstrate that for both men and women, all conventional risk factors were significantly associated with the presence of any detectable CAC, and the mean CAC score increased in proportion to the number of CAD risk factors. In age-adjusted (multivariable) logistic regression analysis, cigarette use, histories of hypercholesterolemia, diabetes, and hypertension were each significantly associated with mild to extensive CAC scores (10.0). CONCLUSION: CAD risk factors are associated with higher atherosclerotic plaque burden in both men and women. The odds ratios associated with each risk factor relative to the extent of CAC are similar to those reported for the development of clinical CAD, suggesting the existence of an association between CAC (subclinical disease) and CAD (clinical disease). Ann Epidemiol 2003;13:163–169. © 2003 Elsevier Science Inc. All rights reserved. KEY WORDS: Computed Tomography (CT), Electron Beam, Calcium, Coronary Disease, Coronary Disease Risk Factors.
INTRODUCTION Numerous pathologic studies have correlated the extent of calcification in the coronary arteries with the presence and degree of atherosclerotic plaque development (1, 2). Only recently, however, has the clinical utility of these post-mortem observations been fully understood. Electron beam tomography (EBT) is a noninvasive, anatomical method of quantifying coronary artery calcium (CAC), allowing clinicians to accurately assess an individual’s plaque burden (3– 6). Previous studies have demonstrated that the amount of CAC detected by EBT correlates well with the histologic From the Department of Medicine, Section of Cardiology, University of Illinois at Chicago College of Medicine, Chicago, IL (J.A.H., E.V.C., A.J.K., A.S., G.T.K.); Department of Medical Surgical Nursing, University of Illinois at Chicago College of Nursing, Chicago, IL (J.A.H.); and Department of Preventive Medicine, Northwestern University Medical School, Chicago, IL (M.L.D.). Address correspondences to: Julie Anne Hoff, RN, PhD, Department of Medicine, Section of Cardiology, 840 South Wood Street (M/C 787), Chicago, Illinois 60612, USA. Tel.: (312) 996-6730; Fax: (312) 996-2948; E-mail: [email protected] Received June 19, 2001; revised April 15, 2002; accepted April 24, 2002. This research was supported with internal funding from the University of Illinois, Department of Medicine, Section of Cardiology. © 2003 Elsevier Science Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010
plaque volume (7–9). There is also evidence that CAC is a sensitive marker for angiographically defined coronary artery disease (CAD), and that the amount of CAC is strongly associated with maximal stenosis of the coronary arteries (10, 11). Hence, determination of the extent of coronary atherosclerosis by EBT may allow the establishment of relative risks for the development of clinical CAD, and associated subsequent cardiovascular events. The coronary artery calcium score as determined by EBT (12) has been shown to have prognostic value in determining the course of disease in asymptomatic individuals (13– 15). Among high-risk individuals, the amount of CAC detected is correlated with both the number and severity of CAD risk factors (16). The objective of this study was to examine the association between CAC and CAD risk factors in a large cohort of asymptomatic men and women.
METHODS Study Sample From 1993 through 1999, 32,481 individuals aged 30 to 90 years underwent EBT CAC screening at the UIC Medical 1047-2797/03/$–see front matter PII S1047-2797(02)00277-6
164
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
Selected Abbreviations and Acronyms EBT electron beam tomography CAC coronary artery calcium CAD coronary artery disease EKG electrocardiography ANOVA analysis of variance
Center and a complete medical history was also obtained. Of these individuals, 1573 were excluded from these analyses for the following reasons: a history of angina, diagnostic coronary angiography, catheter-based intervention, coronary artery bypass surgery, and/or myocardial infarction. Thus, the study sample consisted of 22,190 men and 8718 women. Risk Factor Acquisition Immediately prior to EBT screening, each individual was asked to complete a questionnaire that elicited information regarding personal demographics, past medical history, and conventional risk factors for CAD, which included cigarette use, family history of CAD, hypercholesterolemia, diabetes and hypertension. A positive history for cigarette use was defined as any history of cigarette use, past or current. Individuals were considered to have a positive family history for CAD when they reported physician-diagnosed CAD or myocardial infarction in a parent, grandparent or sibling occurring before the age 65. Hypercholesterolemia was defined as a self-reported total cholesterol level greater than 200mg/dL, or the use of lipid-lowering agents. Subjects were considered diabetic if they reported using oral hypoglycemic agents or insulin and hypertensive if they reported a history of high blood pressure or the use of medications to lower blood pressure. To facilitate the interpretation of age as a risk factor, the variable was dichotomized into 55 and 55 years. The validity of CAD risk factor information obtained using a self-administered questionnaire was examined in a peripheral study (17). The results of that study exhibited high correlation between self-reported and clinically measured CAD risk factor data. Variables validated by this study included weight, height, histories of hypercholesterolemia, diabetes and hypertension. EBT Imaging Procedures EBT CAC scans were obtained with 100-msec acquisition time and 3-mm scan width using a C-100 EBT scanner (Imatron, South San Francisco, CA). Using electrocardiographic triggering at 80% of the EKG RR interval, 40 images were acquired during one to two breath-holding sequences. To ensure identification of CAC in the proximal coronary arteries, an overlap scan consisting of additional 20 images starting at the base of the heart was obtained in the same manner. Images were reconstructed to a 512 512 pixels matrix with a 26-cm field of view.
AEP Vol. 13, No. 3 March 2003: 163–169
EBT software allows for quantification of calcium area and density (12). A calcified plaque was considered present if at least four adjacent pixels with a CT number of at least 130 Hounsfield units (an area of 1mm2) were identified. An attenuation factor for each lesion was determined by the peak CT number. The calcium score for each lesion was the product of the attenuation factor multiplied by the area of the lesion in square millimeters. A minimum positive CAC score in this study was equal to 1.0. The total CAC score was the sum of all the lesions scored within the left main, left anterior descending, left circumflex , and right coronary arteries. The overlap scan of the proximal coronary arteries was scored using the same protocol. On an artery-by-artery basis, the higher of the two scores was used to calculate the total CAC score for each study. Reproducibility of the CAC results is considered moderate to excellent depending on the EBT screening laboratory (technician/physician experience) and the amount of calcium present (18). In this study, all EBT CAC screening studies were analyzed by one of two experienced observers. To facilitate data interpretation, the CAC score was classified into the following categories: 0.0, 1.0 to 9.9, 10.0 to 99.9, 100 to 399.9, and 400 (no identifiable plaque, minimal plaque, mild plaque, moderate plaque, and extensive atherosclerotic plaque burden, respectively). These categories of the CAC score have been used to differentiate between very low, low, moderate, moderately high, and high cardiovascular risk (3). Additionally, positive CAC scores were categorized using gender-specific quartiles. An overall risk score for each individual was computed by assigning a score of 1 to each reported conventional CAD risk factor (cigarette use, family history of CAD, hypercholesterolemia, hypertension and diabetes). The purpose of this study was to examine the association between five major risk factors for CAD (smoking, family history of CAD, hypercholesterolemia, hypertension, and diabetes) and the extent of CAC in a large cohort of asymptomatic men and women. Statistical Analysis Data analyses were performed using SPSS statistical software (version 10.0; SPSS, Chicago, IL). Descriptive statistics were used to summarize the demographic characteristics, CAD risk factor information and the CAC scores. The differences in mean ages and CAC scores in relation to the overall risk score were examined separately for men and women using ANOVA and Bonferroni post-hoc tests. Chisquare analysis was used to determine significance when comparing the prevalence of each CAD risk factor across the CAC score categories. The association between risk factors for clinical CAD and the extent of subclinical atherosclerosis as measured by CAC was examined using multivariable logistic regression. The logistic models were created using block entry of all six independent variables
165
AEP Vol. 13, No. 3 March 2003: 163–169
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
(age 55 years, smoking, family history of CAD, hypercholesterolemia, diabetes and hypertension).
ANOVA. For men, the number of risk factors increased as the mean age increased. The mean age was similar among women who had 0, 1, or 2 risk factors. However, women with three or more risk factors were older compared with those in all lower risk score categories. For both men and women, the mean CAC score increased in proportion to the number of CAD risk factors. Post-hoc comparisons demonstrated that the differences in mean CAC scores for men were significant. The mean CAC scores were significantly different between women with 1 versus 2, and 2 versus 3 risk factors while the mean CAC score of women with 3 or more risk factors was significantly higher compared with all lower risk score categories. Results of the ANOVA analysis are summarized in Table 2. Using the predetermined CAC score categories (0.0, 1.0 to 9.9, 10.0 to 99.9, 100 to 399.9, and 400), chi-square analysis was used to test whether the proportion of men and women positive for each CAD risk factor differed when stratified by varying degrees of CAC. Among men, 55 years, cigarette use, and histories of diabetes and hypertension were more prevalent in individuals with CAC scores at or above each cut-point compared with those with less or no CAC. The prevalence of family history of CAD reported by men who had any amount of CAC was significantly higher compared with men with no detectable CAC (p 0.001) whereas the prevalence figures were similar across the categories of positive CAC score above 10.0 (52%). The prevalence of hypercholesterolemia among
RESULTS Demographics The study sample was comprised of 30,908 individuals, of whom 22,190 (72%) were men with a mean age of 50 years (SD 9 years), and 8,718 (28%) were women with a mean age of 54 years (SD 9 years). Differences in baseline characteristics between men and women are summarized in Table 1. The sample was primarily composed of whites, with education and income levels above the national average. A large percentage of the sample reported seeing a family physician, following a general diet, and exercising regularly. Risk Factors and Coronary Artery Calcium Differences in the prevalence of the conventional CAD risk factors and EBT CAC scores among men and women are summarized in Table 1. Men reported cigarette use more frequently than women (p 0.001). However, family history of CAD, hypercholesterolemia, and hypertension were reported more often by women (p 0.001). The prevalence of diabetes did not differ by gender. The mean ages and CAC scores were then compared across the overall risk score categories (0, 1, 2, or 3 risk factors) separately for men and women using
TABLE 1. Baselilne characteristics for men and women undergoing electron beam tomography coronary artery calcium screening Characteristics Mean Age ( SD), years Age 55 years (%) White (%) Marital Status Single (%) Married (%) Divorced (%) Education 12 years (%) Annual household income $50,000 (%) Diet General (%) Low salt (%) Low Fat (%) Vegetarian (%) Regular exercise (%) Family physician (%) Cigarette use (ever/current) (%) Family history of CAD (%) Hypercholesterolemia (%) Diabetes (%) Hypertension (%) Presence of CAC (%) Mean CAC score ( SD) Median CAC score CAC Coronary Artery Calcium.
Men (n 22,190)
Women (n 8,718)
p
50 ( 9) 27 93
54 ( 9) 42 91
0.001 0.001 0.001
7 86 6 84 82
6 80 8 70 67
0.05 0.001 0.001 0.001 0.001
82 4 10 1.9 46 80 50 52 36 3.4 21 77 135 ( 381) 6
72 7 15 2.4 42 87 45 56 39 3.8 23 54 55 ( 235) 1
0.001 0.001 0.001 0.06 0.001 0.001 0.001 0.001 0.001 0.08 0.001 0.001 0.001 0.001
166
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
AEP Vol. 13, No. 3 March 2003: 163–169
TABLE 2. Mean age and coronary arery calcium score in men and women with increasing number of conventional CAD risk factors Overall Risk Score Men
0
1
2
3
2,683 48 ( 9) 77 ( 292)
7,159 49 ( 10) 96 ( 291)
7,446 50 ( 9) 144 ( 397)
4,283 52 ( 9) 217 ( 489)
0
1
2
3
950 53 ( 10) 32 ( 207)
2,681 53 ( 10) 36 ( 162)
3,024 54 ( 9) 53 ( 218)
1,807 56 ( 9) 99 ( 341)
a
n Mean Age, years ( SD) Mean CAC Score ( SD) Womenb n Mean Age, years ( SD) Mean CAC Score ( SD)
An overall risk score for each individual was computed by assigning a score of 1 to each reported conventional CAD risk factor. For Men, ANOVA (with Bonferroni post hoc test) showed that mean ages among each of the 4 risk score categories were significantly different (0.001 P 0.01); mean CAC scores among each of the 4 risk score categories were significantly different (P 0.001) except when comparing mean CAC scores of men with 0 versus 1 risk factor. b For Women, ANOVA (with Bonferroni post hoc test) showed that mean ages among women with 0, 1, and 2 risk factors were not significantly different while mean ages of women with 0, 1, and 2 risk factors differed significantly from those with 3 or more risk factors ( P 0.001); mean CAC scores among women with 0, 1, and 2 risk factors differed significantly from those with 3 or more risk factors (P 0.001), and mean CAC scores also differed between women with 1 versus 2, and 2 versus 3 or more risk factors (0.001 P 0.03). CAC Coronary Artery Calcium. a
men with CAC scores in the range of 1.0 to 99.9 was greater compared with men with no CAC (31% versus 33%, p 0.01). However, the prevalence figures did not differ among men with CAC scores above 100 (42%). Among women, age and prevalence of hypercholesterolemia, hypertension and diabetes were higher in each subsequent CAC score category compared with lower scores or no CAC. The prevalence of cigarette use was lower in women with CAC scores in the range of 1.0 to 9.9 compared with women with no CAC (42% versus 46%, p 0.006), whereas in women with CAC scores exceeding 10.0, the prevalence figures increased in proportion to the extent of CAC (51% versus 53%, versus 59%, p 0.001). Among women with CAC scores in the range of 1.0 to 9.9, the family history of CAD was reported more frequently
compared with women with no CAC (59% versus 55%, p 0.001). However, the prevalence figures among women with CAC scores above 10.0 were similar compared with women with no CAC. Table 3, summarizes the proportion of CAD risk factors by each category of the CAC score. The association between conventional CAD risk factors and the presence as well as the extent of CAC was examined using multivariable logistic regression. The results of the regression analysis are summarized in Table 4. For both men and women, age greater than 55 years was the strongest predictor for the presence of any detectable CAC. After age, the strongest predictors for CAC in both genders were diabetes and hypertension. For all CAD risk factors, with the exception of family CAD history, all of the associated odds ratios increased in proportion to the magnitude of
TABLE 3. Age and other conventional CAD risk factors in men and women with increasing coronary artery calcium scores Coronary artery calcium score
Men n (%) Age 55 years (%) Cigarette use (ever/current) (%) Family history of CAD (%) Hypercholesterolemia (%) Diabetes (%) Hypertension (%) Women n (%) Age 55 years (%) Cigarette use (ever/current) (%) Family history of CAD (%) Hypercholesterolemia (%) Diabetes (%) Hypertension (%)
100 to 399.9
400
p
5,011 (23) 31 54 52 40 3.8 23
3,163 (14) 49 61 52 42 5.2 28
2,043 (9) 70 65 52 42 8.6 37
0.001 0.001 0.001 0.001 0.001 0.001
1,338 (15) 60 51 54 49 4.9 31
691 (8) 78 53 57 51 8.0 35
292 (3) 84 59 54 52 9.6 46
0.001 0.001 0.003 0.001 0.001 0.001
0
1.0 to 9.9
10.0 to 99.9
5,020 (23) 11 43 47 31 1.5 12
6,952 (31) 14 47 55 33 2.1 18
4,042 (46) 28 46 55 33 2.1 16
2,355 (27) 40 42 59 38 4.1 25
167
AEP Vol. 13, No. 3 March 2003: 163–169
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
CAC. It was noted, however, that a history of hypercholesterolemia in both genders and diabetes in men were not associated with minimal CAC scores (1.0 to 9.9). Also notable is the fact that trends in the associations between CAD risk factors and the extent of CAC were similar when CAC was expressed as gender specific quartiles. Several differences between men and women were noted in the odds ratios from the previous logistic regression models. Among women, histories of hypercholesterolemia and diabetes posed greater risk for any detectable CAC whereas age, cigarette use, family history of CAD and hypertension posed greater risk in men.
assessment for the prediction of angiographic CAD, reported age, the ratio of total cholesterol to high-density lipoprotein cholesterol, and the CAC score to be significantly associated with the presence of CAD. However, a family history of heart disease and a personal history of hypertension were not reported to be significantly associated with CAD. All risk factors examined in this study were associated with the presence of CAC for both men and women. The odds ratios were significant, but apart from age, only histories of diabetes and hypertension in both men and women were found to confer twice the risk for CAC. Past studies have shown that men and women share many of the known CAD risk factors including age, cigarette use, a family history of CAD, and histories of hypercholesterolemia, diabetes and hypertension (26–30). Despite these similarities, gender differences exist. It is possible that these differences are due to estrogen although the cardioprotective effects of estrogen have not yet been confirmed in prospective clinical trials (31–33). The magnitude of risk associated with specific CAD risk factors also varies between men and women. Data from the Framingham study demonstrated that the relative risks for CAD associated with cigarette use, hypercholesterolemia and hypertension are similar in men and women, but the relative risks associated with age and diabetes vary by gender (31). The results of the current study concur with these findings. Age 55 years posed a significantly greater risk for CAC in men compared with women. In contrast, diabetes posed a greater risk for CAC among women. Certain limitations must be considered before forming any conclusions. First, despite the fact that the study population is relatively large, it remains fairly homogeneous.
DISCUSSION The results of this study demonstrate that for both men and women, all conventional risk factors for CAD were significantly associated with the presence of any detectable CAC, and the mean CAC score increased with increasing numbers of CAD risk factors. Overall, the odds ratios for CAD risk factors and the CAC score (Table 4) are similar in magnitude to those reported by large epidemiologic studies examining CAD risk factors in relation to the development of symptomatic CAD (19–23). These similarities in risk estimates further demonstrate the link between subclinical CAD (as measured by CAC) and clinical CAD. The findings from this study support the presence of an association between CAD risk factors and the presence of CAC among asymptomatic individuals as reported by a previous study (24). Another study (25) comparing the effect of EBT-detected CAC and conventional CAD risk factor
TABLE 4. Association of age and other conventional CAD risk factors with coronary artery calcium scores Coronary artery calcium score 0
Men Age 55 years Cigarette use (ever/current) Family history of CAD Hypercholesterolemia Diabetes Hypertension Women Age 55 years Cigarette use (ever/current) Family history of CAD Hypercholesterolemia Diabetes Hypertension
1.0 to 9.9
10.0 to 99.9
400
100 to 399.9
Odds Ratio
95% CI
Odds Ratio
95% CI
Odds Ratio
95% CI
Odds Ratio
3.62* 1.38* 1.36* 1.24* 1.83* 1.95*
3.28–3.99 1.29–1.47 1.28–1.46 1.16–1.33 1.42–2.34 1.77–2.14
1.35* 1.14* 1.35* 1.05 1.18 1.59*
1.20–1.52 1.05–1.23 1.25–1.45 0.97–1.13 0.88–1.57 1.43–1.78
3.47* 1.40* 1.32* 1.39* 1.87* 2.01*
3.10–3.89 1.29–1.52 1.21–1.44 1.28–1.52 1.40–2.49 1.79–2.25
7.34* 1.78* 1.43* 1.44* 2.17* 2.28*
2.84* 1.13† 1.24* 1.31* 2.17* 1.79*
2.59–3.12 1.03–1.23 1.13–1.35 1.19–1.44 1.66–2.82 1.60–2.00
1.67* 0.89† 1.27* 1.08 1.80* 1.57*
1.49–1.87 0.80–0.99 1.14–1.41 0.97–1.21 1.32–2.44 1.38–1.79
3.44* 1.34* 1.09 1.53* 2.01* 1.85*
3.01–3.95 1.18–1.53 0.95–1.25 1.34–1.75 1.41–2.87 1.59–2.16
8.78* 1.51* 1.41* 1.49* 3.42* 1.87*
95% CI
Odds Ratio
95% CI
6.50–8.28 1.60–1.97 1.29–1.58 1.29–1.60 1.58–2.97 2.00–2.60
17.24* 2.02* 1.59* 1.45* 3.24* 2.83*
14.96–19.85 1.78–2.31 1.39–1.83 1.26–1.66 2.28–4.60 2.41–3.32
7.15–10.76 1.26–1.82 1.17–1.69 1.24–1.79 2.26–5.17 1.53–2.29
11.55* 1.79* 1.27 1.40 † 3.66* 2.93*
8.26–16.15 1.37–2.34 0.97–1.66 1.07–1.83 2.13–6.29 2.23–3.85
The multivariable logistic regression models were created for each coronary artery calcium score category using block entry of all CAD risk factor variables. For each model, the reference group was comprised of negative scores. The analyses were performed separately for men and women. † 0.01 p 0.05, * p 0.001.
168
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
AEP Vol. 13, No. 3 March 2003: 163–169
This is primarily a result of two factors: (i) the study’s reliance on self-referred individuals who tend to be highly motivated to assess their CAD risk and (ii) the individual’s ability to afford the expense associated with the EBT CAC scan. As mentioned previously, the population is predominantly white (92%) with higher than average socioeconomic status compared with the general population. According to the 1990 census information, 75% of the US adult population completed at least 12 years of education and the median income was $30,056 (US Department of Commerce, Economics and Statistics Administration, 1993). In this study, almost all of the men and women completed at least 12 years of education and the majority reported an annual income $50,000. These findings raise the possibility that this study population differs from the general population in their use of preventive health care, screening procedures, and other preventive measures. The lack of substantial representation from other races and socioeconomic classes also makes it difficult to generalize the findings from this study to the overall population. Furthermore, women were underrepresented in this study (72% male versus 28% female). However, despite the high socioeconomic status of the study sample and the under-representation of women and minorities, the prevalence of CAD risk factors in our study sample are similar to estimates for the general US population using data from the National Health and Nutrition Survey (NHANES) and the Atherosclerosis Risk in Communities (ARIC) Study (34, 35). The findings of this study have important clinical and research implications. CAC, a specific marker for atherosclerosis, can be detected and quantified with EBT, and this study demonstrates that the extent of CAC increases with increasing numbers of CAD risk factors in both men and women. It is important to note, however, that in this sample of healthy asymptomatic individuals, a small portion of men (8%) and women (5%) had CAC in the absence of conventional CAD risk factors. Finally, the association of conventional CAD risk factors with subclinical CAD (as measured by CAC) reinforces the importance of early CAD risk factor identification followed by risk factor modification in unassuming asymptomatic men and women.
guidelines for use in asymptomatic persons. Mayo Clin Proc. 1999;74: 243–252. O’Rourke RA, Brundage BH, Froelicher VF, Greenland P, Grundy SM, Hachamovitch R, et al. ACC/AHA expert consensus document on electron-beam computed tomography for the diagnosis and prognosis of coronary artery disease. Circulation. 2000;102: 126–140. Wexler L, Brundage B, Crouse J, Detrano R, Fuster V, Maddahi J, et al. Coronary artery calcium: pathophysiology, epidemiology, image methods and clinical implications—A scientific statement from the American Heart Association. Circulation. 1996;94:1175–1192. Stanford W, Thompson BH. Imaging of coronary artery calcification: its importance in assessing atherosclerotic disease. Radiol Clin North Am. 1999;37:257–272. Mautner SL, Mautner GC, Froehlich J, Feuerstein IM, Proschan MA, Roberts WC, et al. Coronary artery disease: prediction with in vitro electron beam CT. Radiology. 1994;192:625–630. Rumberger JA, Simons B, Fitzpatrick LA, Sheedy PF, Schwartz RS. Coronary artery calcium area by electron-beam computed tomography and coronary atherosclerotic plaque area. Circulation. 1995;92:2157– 2162. Sangiorgi G, Rumberger JA, Severson A, Edwards WD, Gregoire J, Fitzpatrick LA, et al. Arterial calcification and not lumen stenosis is highly correlated with atherosclerotic plaque burden in humans: a histologic study of 723 coronary artery segments using nondecalcifying methodology. J Am Coll Cardiol. 1998;31:126–133. Kajinami K, Takekoshi N. Noninvasive prediction of coronary atherosclerosis by quantification of coronary artery calcification using electron beam computed tomography: comparison with electrocardiographic and thallium exercise stress test results. J Am Coll Cardiol. 1995;25:626–632. Guerci AD, Spadaro LA, Popma JJ, Goodman KJ, Brundage BH, Lerner G, et al. Relation of coronary calcium score by electron beam computed tomography to arteriographic findings in asymptomatic and symptomatic adults. Am J Cardiol. 1997;79:128–133. Agatston AS, Janowitz WR, Hildner FJ, Zustmer NR, Viamonte M, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol. 1990;15:827–832. Arad Y, Spadaro LA, Goodman K, Newstein D, Guerci A. Prediction of coronary events with electron beam tomography. J Am Coll Cardiol. 2000;36:1253–1260. Raggi P, Callister TQ, Cooil B, He ZX, Lippolis NJ, Russo DJ, et al. Identification of patients at increased risk of first unheralded acute myocardial infarction by electron-beam computed tomography. Circulation. 2000;101:850–855. Wong ND, Hsu JC, Detrano RC, Diamond G, Eisenberg H, Gardin JM. Coronary artery calcium evaluation by electron beam computed tomography and its relation to new cardiovascular events. Am J Cardiol. 2000;86:495–498. Secci A, Wong N, Tang W, 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. Hoff JA. Coronary artery calcium screening using electron beam tomography. Thesis. University of Illinois at Chicago, 1999. Devries S, Wolfkiel C, Shah V, Chomka E, Rich S. Reproducibility of the measurement of coronary artery calcium with ultrafast computed tomography. Am J Cardiol. 1995;75:973–975. Myers RH, Kiely DK, Cupples A, Kannel WB. Parental history is an independent risk factor for coronary artery disease: The Framingham Study. Am Heart J. 1990;120:963–969. Stamler J, Wentworth D, Neaton JD (MRFIT Research Group). Is the relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? JAMA. 1986;256: 2823–2828.
REFERENCES 1. Eggen DA, Strong JP, McGill HC. Coronary calcification: relationship to clinically significant coronary lesions and race, sex, and topographic distribution. Circulation. 1965;32:948–955. 2. Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insull W Jr, et al. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis: a report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation. 1995;92:1355–1374. 3. Rumberger JA, Brundage BH, Rader DJ, Kondos GT. Electron beam computed tomographic coronary calcium scanning: a review and
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17. 18.
19.
20.
169
AEP Vol. 13, No. 3 March 2003: 163–169
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
21. Dawber TR. The Framingham Study: The Epidemiology of Atherosclerotic Disease. Cambridge: Harvard University Press; 1980. 22. Kannel WB, Castelli WP, Gordon T, McNamara PM. Serum cholesterol, lipoproteins, and the risk of coronary heart disease: The Framingham Study. Ann Intern Med. 1971;74:1–12. 23. The Pooling Project Research Group. Relationship of blood pressure, serum cholesterol, smoking habit, relative weight and ECG abnormalities to incidence of major coronary events: Final report of the Pooling Project. J Chron Dis. 1978;31:201–306. 24. Wong ND, Kouwabunpat D, Vo AN, Detrano RC, Eisenberg H, Goel M, et al. Coronary calcium and atherosclerosis by ultrafast computed tomography in asymptomatic men and women: relation to age and risk factors. Am Heart J. 1994;127:422–430. 25. Guerci AD, Spadaro LA, Goodman KJ, Lledo-Perez A, Newstein D, Lerner G, et al. Comparison of electron beam computed tomography scanning and conventional risk factor assessment for the prediction of angiographic coronary artery disease. J Am Coll Cardiol. 1998;32: 673–679. 26. Eaker ED, Packard B, Wenger NK, Clarkson TB, Tyroler HA. Coronary artery disease in women. Am J Cardiol. 1988;61:641–644. 27. Wenger NK. Coronary heart disease: an older woman’s major health risk. BMJ. 1997;315:1085–1090.
28. Kannel WB. Metabolic risk factors for coronary heart disease in women: perspective from the Framingham Study. Am Heart J. 1987; 114:413–419. 29. Kuhn FE, Rackley CE. Coronary artery disease in women: Risk factors, evaluation, treatment and prevention. Arch Intern Med. 1993; 153:2626– 2636. 30. Murdaugh CL, O’Rourke RA. Coronary heart disease in women: special considerations. Curr Probl Cardiol. 1988;13:73–156. 31. Anderson KM, Wilson PW, Odell PM, Kannel WB. An updated coronary risk profile: a statement for health professionals. Circulation. 1991;83: 356–362. 32. Kannel WB. Nutrition and the occurrence and prevention of cardiovascular disease in the elderly. Nutr Rev. 1988;46:68–78. 33. Dobson AJ, Alexander HM, Heller RF, Lloyd DM. Passive cigarette use and the risk of heart attack or coronary death. Med J Aust. 1991; 154:793–797. 34. Garg R, Madans JH, Kleinman JC. Regional variation in ischemic heart disease incidence. J Clin Epidemiol. 1992;45:149–156. 35. Howard G, Manolio TA, Burke GL, Wolfson SK, O’Leary DH. Does the association of risk factors and atherosclerosis change with age? An analysis of the combined ARIC and CHS cohorts. The Atherosclerosis Risk in Communities (ARIC) and Cardiovascular Health Study (CHS) investigators. Stroke. 1997;28:1693–1701.
PURPOSE: Electron beam tomography (EBT) is a noninvasive measure of coronary artery calcium (CAC), a marker for atherosclerosis. In this study we examined the association between conventional risk factors for coronary artery disease (CAD) and CAC. METHODS: EBT CAC screening was performed on 30,908 asymptomatic individuals aged 30 to 90 years. Prior to EBT screening, individuals provided demographic and CAD risk factor information. EBT utilized a C-100 EBT scanner, and the amount of CAC was determined using the Agatston scoring method. RESULTS: The results of this study demonstrate that for both men and women, all conventional risk factors were significantly associated with the presence of any detectable CAC, and the mean CAC score increased in proportion to the number of CAD risk factors. In age-adjusted (multivariable) logistic regression analysis, cigarette use, histories of hypercholesterolemia, diabetes, and hypertension were each significantly associated with mild to extensive CAC scores (10.0). CONCLUSION: CAD risk factors are associated with higher atherosclerotic plaque burden in both men and women. The odds ratios associated with each risk factor relative to the extent of CAC are similar to those reported for the development of clinical CAD, suggesting the existence of an association between CAC (subclinical disease) and CAD (clinical disease). Ann Epidemiol 2003;13:163–169. © 2003 Elsevier Science Inc. All rights reserved. KEY WORDS: Computed Tomography (CT), Electron Beam, Calcium, Coronary Disease, Coronary Disease Risk Factors.
INTRODUCTION Numerous pathologic studies have correlated the extent of calcification in the coronary arteries with the presence and degree of atherosclerotic plaque development (1, 2). Only recently, however, has the clinical utility of these post-mortem observations been fully understood. Electron beam tomography (EBT) is a noninvasive, anatomical method of quantifying coronary artery calcium (CAC), allowing clinicians to accurately assess an individual’s plaque burden (3– 6). Previous studies have demonstrated that the amount of CAC detected by EBT correlates well with the histologic From the Department of Medicine, Section of Cardiology, University of Illinois at Chicago College of Medicine, Chicago, IL (J.A.H., E.V.C., A.J.K., A.S., G.T.K.); Department of Medical Surgical Nursing, University of Illinois at Chicago College of Nursing, Chicago, IL (J.A.H.); and Department of Preventive Medicine, Northwestern University Medical School, Chicago, IL (M.L.D.). Address correspondences to: Julie Anne Hoff, RN, PhD, Department of Medicine, Section of Cardiology, 840 South Wood Street (M/C 787), Chicago, Illinois 60612, USA. Tel.: (312) 996-6730; Fax: (312) 996-2948; E-mail: [email protected] Received June 19, 2001; revised April 15, 2002; accepted April 24, 2002. This research was supported with internal funding from the University of Illinois, Department of Medicine, Section of Cardiology. © 2003 Elsevier Science Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010
plaque volume (7–9). There is also evidence that CAC is a sensitive marker for angiographically defined coronary artery disease (CAD), and that the amount of CAC is strongly associated with maximal stenosis of the coronary arteries (10, 11). Hence, determination of the extent of coronary atherosclerosis by EBT may allow the establishment of relative risks for the development of clinical CAD, and associated subsequent cardiovascular events. The coronary artery calcium score as determined by EBT (12) has been shown to have prognostic value in determining the course of disease in asymptomatic individuals (13– 15). Among high-risk individuals, the amount of CAC detected is correlated with both the number and severity of CAD risk factors (16). The objective of this study was to examine the association between CAC and CAD risk factors in a large cohort of asymptomatic men and women.
METHODS Study Sample From 1993 through 1999, 32,481 individuals aged 30 to 90 years underwent EBT CAC screening at the UIC Medical 1047-2797/03/$–see front matter PII S1047-2797(02)00277-6
164
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
Selected Abbreviations and Acronyms EBT electron beam tomography CAC coronary artery calcium CAD coronary artery disease EKG electrocardiography ANOVA analysis of variance
Center and a complete medical history was also obtained. Of these individuals, 1573 were excluded from these analyses for the following reasons: a history of angina, diagnostic coronary angiography, catheter-based intervention, coronary artery bypass surgery, and/or myocardial infarction. Thus, the study sample consisted of 22,190 men and 8718 women. Risk Factor Acquisition Immediately prior to EBT screening, each individual was asked to complete a questionnaire that elicited information regarding personal demographics, past medical history, and conventional risk factors for CAD, which included cigarette use, family history of CAD, hypercholesterolemia, diabetes and hypertension. A positive history for cigarette use was defined as any history of cigarette use, past or current. Individuals were considered to have a positive family history for CAD when they reported physician-diagnosed CAD or myocardial infarction in a parent, grandparent or sibling occurring before the age 65. Hypercholesterolemia was defined as a self-reported total cholesterol level greater than 200mg/dL, or the use of lipid-lowering agents. Subjects were considered diabetic if they reported using oral hypoglycemic agents or insulin and hypertensive if they reported a history of high blood pressure or the use of medications to lower blood pressure. To facilitate the interpretation of age as a risk factor, the variable was dichotomized into 55 and 55 years. The validity of CAD risk factor information obtained using a self-administered questionnaire was examined in a peripheral study (17). The results of that study exhibited high correlation between self-reported and clinically measured CAD risk factor data. Variables validated by this study included weight, height, histories of hypercholesterolemia, diabetes and hypertension. EBT Imaging Procedures EBT CAC scans were obtained with 100-msec acquisition time and 3-mm scan width using a C-100 EBT scanner (Imatron, South San Francisco, CA). Using electrocardiographic triggering at 80% of the EKG RR interval, 40 images were acquired during one to two breath-holding sequences. To ensure identification of CAC in the proximal coronary arteries, an overlap scan consisting of additional 20 images starting at the base of the heart was obtained in the same manner. Images were reconstructed to a 512 512 pixels matrix with a 26-cm field of view.
AEP Vol. 13, No. 3 March 2003: 163–169
EBT software allows for quantification of calcium area and density (12). A calcified plaque was considered present if at least four adjacent pixels with a CT number of at least 130 Hounsfield units (an area of 1mm2) were identified. An attenuation factor for each lesion was determined by the peak CT number. The calcium score for each lesion was the product of the attenuation factor multiplied by the area of the lesion in square millimeters. A minimum positive CAC score in this study was equal to 1.0. The total CAC score was the sum of all the lesions scored within the left main, left anterior descending, left circumflex , and right coronary arteries. The overlap scan of the proximal coronary arteries was scored using the same protocol. On an artery-by-artery basis, the higher of the two scores was used to calculate the total CAC score for each study. Reproducibility of the CAC results is considered moderate to excellent depending on the EBT screening laboratory (technician/physician experience) and the amount of calcium present (18). In this study, all EBT CAC screening studies were analyzed by one of two experienced observers. To facilitate data interpretation, the CAC score was classified into the following categories: 0.0, 1.0 to 9.9, 10.0 to 99.9, 100 to 399.9, and 400 (no identifiable plaque, minimal plaque, mild plaque, moderate plaque, and extensive atherosclerotic plaque burden, respectively). These categories of the CAC score have been used to differentiate between very low, low, moderate, moderately high, and high cardiovascular risk (3). Additionally, positive CAC scores were categorized using gender-specific quartiles. An overall risk score for each individual was computed by assigning a score of 1 to each reported conventional CAD risk factor (cigarette use, family history of CAD, hypercholesterolemia, hypertension and diabetes). The purpose of this study was to examine the association between five major risk factors for CAD (smoking, family history of CAD, hypercholesterolemia, hypertension, and diabetes) and the extent of CAC in a large cohort of asymptomatic men and women. Statistical Analysis Data analyses were performed using SPSS statistical software (version 10.0; SPSS, Chicago, IL). Descriptive statistics were used to summarize the demographic characteristics, CAD risk factor information and the CAC scores. The differences in mean ages and CAC scores in relation to the overall risk score were examined separately for men and women using ANOVA and Bonferroni post-hoc tests. Chisquare analysis was used to determine significance when comparing the prevalence of each CAD risk factor across the CAC score categories. The association between risk factors for clinical CAD and the extent of subclinical atherosclerosis as measured by CAC was examined using multivariable logistic regression. The logistic models were created using block entry of all six independent variables
165
AEP Vol. 13, No. 3 March 2003: 163–169
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
(age 55 years, smoking, family history of CAD, hypercholesterolemia, diabetes and hypertension).
ANOVA. For men, the number of risk factors increased as the mean age increased. The mean age was similar among women who had 0, 1, or 2 risk factors. However, women with three or more risk factors were older compared with those in all lower risk score categories. For both men and women, the mean CAC score increased in proportion to the number of CAD risk factors. Post-hoc comparisons demonstrated that the differences in mean CAC scores for men were significant. The mean CAC scores were significantly different between women with 1 versus 2, and 2 versus 3 risk factors while the mean CAC score of women with 3 or more risk factors was significantly higher compared with all lower risk score categories. Results of the ANOVA analysis are summarized in Table 2. Using the predetermined CAC score categories (0.0, 1.0 to 9.9, 10.0 to 99.9, 100 to 399.9, and 400), chi-square analysis was used to test whether the proportion of men and women positive for each CAD risk factor differed when stratified by varying degrees of CAC. Among men, 55 years, cigarette use, and histories of diabetes and hypertension were more prevalent in individuals with CAC scores at or above each cut-point compared with those with less or no CAC. The prevalence of family history of CAD reported by men who had any amount of CAC was significantly higher compared with men with no detectable CAC (p 0.001) whereas the prevalence figures were similar across the categories of positive CAC score above 10.0 (52%). The prevalence of hypercholesterolemia among
RESULTS Demographics The study sample was comprised of 30,908 individuals, of whom 22,190 (72%) were men with a mean age of 50 years (SD 9 years), and 8,718 (28%) were women with a mean age of 54 years (SD 9 years). Differences in baseline characteristics between men and women are summarized in Table 1. The sample was primarily composed of whites, with education and income levels above the national average. A large percentage of the sample reported seeing a family physician, following a general diet, and exercising regularly. Risk Factors and Coronary Artery Calcium Differences in the prevalence of the conventional CAD risk factors and EBT CAC scores among men and women are summarized in Table 1. Men reported cigarette use more frequently than women (p 0.001). However, family history of CAD, hypercholesterolemia, and hypertension were reported more often by women (p 0.001). The prevalence of diabetes did not differ by gender. The mean ages and CAC scores were then compared across the overall risk score categories (0, 1, 2, or 3 risk factors) separately for men and women using
TABLE 1. Baselilne characteristics for men and women undergoing electron beam tomography coronary artery calcium screening Characteristics Mean Age ( SD), years Age 55 years (%) White (%) Marital Status Single (%) Married (%) Divorced (%) Education 12 years (%) Annual household income $50,000 (%) Diet General (%) Low salt (%) Low Fat (%) Vegetarian (%) Regular exercise (%) Family physician (%) Cigarette use (ever/current) (%) Family history of CAD (%) Hypercholesterolemia (%) Diabetes (%) Hypertension (%) Presence of CAC (%) Mean CAC score ( SD) Median CAC score CAC Coronary Artery Calcium.
Men (n 22,190)
Women (n 8,718)
p
50 ( 9) 27 93
54 ( 9) 42 91
0.001 0.001 0.001
7 86 6 84 82
6 80 8 70 67
0.05 0.001 0.001 0.001 0.001
82 4 10 1.9 46 80 50 52 36 3.4 21 77 135 ( 381) 6
72 7 15 2.4 42 87 45 56 39 3.8 23 54 55 ( 235) 1
0.001 0.001 0.001 0.06 0.001 0.001 0.001 0.001 0.001 0.08 0.001 0.001 0.001 0.001
166
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
AEP Vol. 13, No. 3 March 2003: 163–169
TABLE 2. Mean age and coronary arery calcium score in men and women with increasing number of conventional CAD risk factors Overall Risk Score Men
0
1
2
3
2,683 48 ( 9) 77 ( 292)
7,159 49 ( 10) 96 ( 291)
7,446 50 ( 9) 144 ( 397)
4,283 52 ( 9) 217 ( 489)
0
1
2
3
950 53 ( 10) 32 ( 207)
2,681 53 ( 10) 36 ( 162)
3,024 54 ( 9) 53 ( 218)
1,807 56 ( 9) 99 ( 341)
a
n Mean Age, years ( SD) Mean CAC Score ( SD) Womenb n Mean Age, years ( SD) Mean CAC Score ( SD)
An overall risk score for each individual was computed by assigning a score of 1 to each reported conventional CAD risk factor. For Men, ANOVA (with Bonferroni post hoc test) showed that mean ages among each of the 4 risk score categories were significantly different (0.001 P 0.01); mean CAC scores among each of the 4 risk score categories were significantly different (P 0.001) except when comparing mean CAC scores of men with 0 versus 1 risk factor. b For Women, ANOVA (with Bonferroni post hoc test) showed that mean ages among women with 0, 1, and 2 risk factors were not significantly different while mean ages of women with 0, 1, and 2 risk factors differed significantly from those with 3 or more risk factors ( P 0.001); mean CAC scores among women with 0, 1, and 2 risk factors differed significantly from those with 3 or more risk factors (P 0.001), and mean CAC scores also differed between women with 1 versus 2, and 2 versus 3 or more risk factors (0.001 P 0.03). CAC Coronary Artery Calcium. a
men with CAC scores in the range of 1.0 to 99.9 was greater compared with men with no CAC (31% versus 33%, p 0.01). However, the prevalence figures did not differ among men with CAC scores above 100 (42%). Among women, age and prevalence of hypercholesterolemia, hypertension and diabetes were higher in each subsequent CAC score category compared with lower scores or no CAC. The prevalence of cigarette use was lower in women with CAC scores in the range of 1.0 to 9.9 compared with women with no CAC (42% versus 46%, p 0.006), whereas in women with CAC scores exceeding 10.0, the prevalence figures increased in proportion to the extent of CAC (51% versus 53%, versus 59%, p 0.001). Among women with CAC scores in the range of 1.0 to 9.9, the family history of CAD was reported more frequently
compared with women with no CAC (59% versus 55%, p 0.001). However, the prevalence figures among women with CAC scores above 10.0 were similar compared with women with no CAC. Table 3, summarizes the proportion of CAD risk factors by each category of the CAC score. The association between conventional CAD risk factors and the presence as well as the extent of CAC was examined using multivariable logistic regression. The results of the regression analysis are summarized in Table 4. For both men and women, age greater than 55 years was the strongest predictor for the presence of any detectable CAC. After age, the strongest predictors for CAC in both genders were diabetes and hypertension. For all CAD risk factors, with the exception of family CAD history, all of the associated odds ratios increased in proportion to the magnitude of
TABLE 3. Age and other conventional CAD risk factors in men and women with increasing coronary artery calcium scores Coronary artery calcium score
Men n (%) Age 55 years (%) Cigarette use (ever/current) (%) Family history of CAD (%) Hypercholesterolemia (%) Diabetes (%) Hypertension (%) Women n (%) Age 55 years (%) Cigarette use (ever/current) (%) Family history of CAD (%) Hypercholesterolemia (%) Diabetes (%) Hypertension (%)
100 to 399.9
400
p
5,011 (23) 31 54 52 40 3.8 23
3,163 (14) 49 61 52 42 5.2 28
2,043 (9) 70 65 52 42 8.6 37
0.001 0.001 0.001 0.001 0.001 0.001
1,338 (15) 60 51 54 49 4.9 31
691 (8) 78 53 57 51 8.0 35
292 (3) 84 59 54 52 9.6 46
0.001 0.001 0.003 0.001 0.001 0.001
0
1.0 to 9.9
10.0 to 99.9
5,020 (23) 11 43 47 31 1.5 12
6,952 (31) 14 47 55 33 2.1 18
4,042 (46) 28 46 55 33 2.1 16
2,355 (27) 40 42 59 38 4.1 25
167
AEP Vol. 13, No. 3 March 2003: 163–169
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
CAC. It was noted, however, that a history of hypercholesterolemia in both genders and diabetes in men were not associated with minimal CAC scores (1.0 to 9.9). Also notable is the fact that trends in the associations between CAD risk factors and the extent of CAC were similar when CAC was expressed as gender specific quartiles. Several differences between men and women were noted in the odds ratios from the previous logistic regression models. Among women, histories of hypercholesterolemia and diabetes posed greater risk for any detectable CAC whereas age, cigarette use, family history of CAD and hypertension posed greater risk in men.
assessment for the prediction of angiographic CAD, reported age, the ratio of total cholesterol to high-density lipoprotein cholesterol, and the CAC score to be significantly associated with the presence of CAD. However, a family history of heart disease and a personal history of hypertension were not reported to be significantly associated with CAD. All risk factors examined in this study were associated with the presence of CAC for both men and women. The odds ratios were significant, but apart from age, only histories of diabetes and hypertension in both men and women were found to confer twice the risk for CAC. Past studies have shown that men and women share many of the known CAD risk factors including age, cigarette use, a family history of CAD, and histories of hypercholesterolemia, diabetes and hypertension (26–30). Despite these similarities, gender differences exist. It is possible that these differences are due to estrogen although the cardioprotective effects of estrogen have not yet been confirmed in prospective clinical trials (31–33). The magnitude of risk associated with specific CAD risk factors also varies between men and women. Data from the Framingham study demonstrated that the relative risks for CAD associated with cigarette use, hypercholesterolemia and hypertension are similar in men and women, but the relative risks associated with age and diabetes vary by gender (31). The results of the current study concur with these findings. Age 55 years posed a significantly greater risk for CAC in men compared with women. In contrast, diabetes posed a greater risk for CAC among women. Certain limitations must be considered before forming any conclusions. First, despite the fact that the study population is relatively large, it remains fairly homogeneous.
DISCUSSION The results of this study demonstrate that for both men and women, all conventional risk factors for CAD were significantly associated with the presence of any detectable CAC, and the mean CAC score increased with increasing numbers of CAD risk factors. Overall, the odds ratios for CAD risk factors and the CAC score (Table 4) are similar in magnitude to those reported by large epidemiologic studies examining CAD risk factors in relation to the development of symptomatic CAD (19–23). These similarities in risk estimates further demonstrate the link between subclinical CAD (as measured by CAC) and clinical CAD. The findings from this study support the presence of an association between CAD risk factors and the presence of CAC among asymptomatic individuals as reported by a previous study (24). Another study (25) comparing the effect of EBT-detected CAC and conventional CAD risk factor
TABLE 4. Association of age and other conventional CAD risk factors with coronary artery calcium scores Coronary artery calcium score 0
Men Age 55 years Cigarette use (ever/current) Family history of CAD Hypercholesterolemia Diabetes Hypertension Women Age 55 years Cigarette use (ever/current) Family history of CAD Hypercholesterolemia Diabetes Hypertension
1.0 to 9.9
10.0 to 99.9
400
100 to 399.9
Odds Ratio
95% CI
Odds Ratio
95% CI
Odds Ratio
95% CI
Odds Ratio
3.62* 1.38* 1.36* 1.24* 1.83* 1.95*
3.28–3.99 1.29–1.47 1.28–1.46 1.16–1.33 1.42–2.34 1.77–2.14
1.35* 1.14* 1.35* 1.05 1.18 1.59*
1.20–1.52 1.05–1.23 1.25–1.45 0.97–1.13 0.88–1.57 1.43–1.78
3.47* 1.40* 1.32* 1.39* 1.87* 2.01*
3.10–3.89 1.29–1.52 1.21–1.44 1.28–1.52 1.40–2.49 1.79–2.25
7.34* 1.78* 1.43* 1.44* 2.17* 2.28*
2.84* 1.13† 1.24* 1.31* 2.17* 1.79*
2.59–3.12 1.03–1.23 1.13–1.35 1.19–1.44 1.66–2.82 1.60–2.00
1.67* 0.89† 1.27* 1.08 1.80* 1.57*
1.49–1.87 0.80–0.99 1.14–1.41 0.97–1.21 1.32–2.44 1.38–1.79
3.44* 1.34* 1.09 1.53* 2.01* 1.85*
3.01–3.95 1.18–1.53 0.95–1.25 1.34–1.75 1.41–2.87 1.59–2.16
8.78* 1.51* 1.41* 1.49* 3.42* 1.87*
95% CI
Odds Ratio
95% CI
6.50–8.28 1.60–1.97 1.29–1.58 1.29–1.60 1.58–2.97 2.00–2.60
17.24* 2.02* 1.59* 1.45* 3.24* 2.83*
14.96–19.85 1.78–2.31 1.39–1.83 1.26–1.66 2.28–4.60 2.41–3.32
7.15–10.76 1.26–1.82 1.17–1.69 1.24–1.79 2.26–5.17 1.53–2.29
11.55* 1.79* 1.27 1.40 † 3.66* 2.93*
8.26–16.15 1.37–2.34 0.97–1.66 1.07–1.83 2.13–6.29 2.23–3.85
The multivariable logistic regression models were created for each coronary artery calcium score category using block entry of all CAD risk factor variables. For each model, the reference group was comprised of negative scores. The analyses were performed separately for men and women. † 0.01 p 0.05, * p 0.001.
168
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
AEP Vol. 13, No. 3 March 2003: 163–169
This is primarily a result of two factors: (i) the study’s reliance on self-referred individuals who tend to be highly motivated to assess their CAD risk and (ii) the individual’s ability to afford the expense associated with the EBT CAC scan. As mentioned previously, the population is predominantly white (92%) with higher than average socioeconomic status compared with the general population. According to the 1990 census information, 75% of the US adult population completed at least 12 years of education and the median income was $30,056 (US Department of Commerce, Economics and Statistics Administration, 1993). In this study, almost all of the men and women completed at least 12 years of education and the majority reported an annual income $50,000. These findings raise the possibility that this study population differs from the general population in their use of preventive health care, screening procedures, and other preventive measures. The lack of substantial representation from other races and socioeconomic classes also makes it difficult to generalize the findings from this study to the overall population. Furthermore, women were underrepresented in this study (72% male versus 28% female). However, despite the high socioeconomic status of the study sample and the under-representation of women and minorities, the prevalence of CAD risk factors in our study sample are similar to estimates for the general US population using data from the National Health and Nutrition Survey (NHANES) and the Atherosclerosis Risk in Communities (ARIC) Study (34, 35). The findings of this study have important clinical and research implications. CAC, a specific marker for atherosclerosis, can be detected and quantified with EBT, and this study demonstrates that the extent of CAC increases with increasing numbers of CAD risk factors in both men and women. It is important to note, however, that in this sample of healthy asymptomatic individuals, a small portion of men (8%) and women (5%) had CAC in the absence of conventional CAD risk factors. Finally, the association of conventional CAD risk factors with subclinical CAD (as measured by CAC) reinforces the importance of early CAD risk factor identification followed by risk factor modification in unassuming asymptomatic men and women.
guidelines for use in asymptomatic persons. Mayo Clin Proc. 1999;74: 243–252. O’Rourke RA, Brundage BH, Froelicher VF, Greenland P, Grundy SM, Hachamovitch R, et al. ACC/AHA expert consensus document on electron-beam computed tomography for the diagnosis and prognosis of coronary artery disease. Circulation. 2000;102: 126–140. Wexler L, Brundage B, Crouse J, Detrano R, Fuster V, Maddahi J, et al. Coronary artery calcium: pathophysiology, epidemiology, image methods and clinical implications—A scientific statement from the American Heart Association. Circulation. 1996;94:1175–1192. Stanford W, Thompson BH. Imaging of coronary artery calcification: its importance in assessing atherosclerotic disease. Radiol Clin North Am. 1999;37:257–272. Mautner SL, Mautner GC, Froehlich J, Feuerstein IM, Proschan MA, Roberts WC, et al. Coronary artery disease: prediction with in vitro electron beam CT. Radiology. 1994;192:625–630. Rumberger JA, Simons B, Fitzpatrick LA, Sheedy PF, Schwartz RS. Coronary artery calcium area by electron-beam computed tomography and coronary atherosclerotic plaque area. Circulation. 1995;92:2157– 2162. Sangiorgi G, Rumberger JA, Severson A, Edwards WD, Gregoire J, Fitzpatrick LA, et al. Arterial calcification and not lumen stenosis is highly correlated with atherosclerotic plaque burden in humans: a histologic study of 723 coronary artery segments using nondecalcifying methodology. J Am Coll Cardiol. 1998;31:126–133. Kajinami K, Takekoshi N. Noninvasive prediction of coronary atherosclerosis by quantification of coronary artery calcification using electron beam computed tomography: comparison with electrocardiographic and thallium exercise stress test results. J Am Coll Cardiol. 1995;25:626–632. Guerci AD, Spadaro LA, Popma JJ, Goodman KJ, Brundage BH, Lerner G, et al. Relation of coronary calcium score by electron beam computed tomography to arteriographic findings in asymptomatic and symptomatic adults. Am J Cardiol. 1997;79:128–133. Agatston AS, Janowitz WR, Hildner FJ, Zustmer NR, Viamonte M, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol. 1990;15:827–832. Arad Y, Spadaro LA, Goodman K, Newstein D, Guerci A. Prediction of coronary events with electron beam tomography. J Am Coll Cardiol. 2000;36:1253–1260. Raggi P, Callister TQ, Cooil B, He ZX, Lippolis NJ, Russo DJ, et al. Identification of patients at increased risk of first unheralded acute myocardial infarction by electron-beam computed tomography. Circulation. 2000;101:850–855. Wong ND, Hsu JC, Detrano RC, Diamond G, Eisenberg H, Gardin JM. Coronary artery calcium evaluation by electron beam computed tomography and its relation to new cardiovascular events. Am J Cardiol. 2000;86:495–498. Secci A, Wong N, Tang W, 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. Hoff JA. Coronary artery calcium screening using electron beam tomography. Thesis. University of Illinois at Chicago, 1999. Devries S, Wolfkiel C, Shah V, Chomka E, Rich S. Reproducibility of the measurement of coronary artery calcium with ultrafast computed tomography. Am J Cardiol. 1995;75:973–975. Myers RH, Kiely DK, Cupples A, Kannel WB. Parental history is an independent risk factor for coronary artery disease: The Framingham Study. Am Heart J. 1990;120:963–969. Stamler J, Wentworth D, Neaton JD (MRFIT Research Group). Is the relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? JAMA. 1986;256: 2823–2828.
REFERENCES 1. Eggen DA, Strong JP, McGill HC. Coronary calcification: relationship to clinically significant coronary lesions and race, sex, and topographic distribution. Circulation. 1965;32:948–955. 2. Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insull W Jr, et al. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis: a report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation. 1995;92:1355–1374. 3. Rumberger JA, Brundage BH, Rader DJ, Kondos GT. Electron beam computed tomographic coronary calcium scanning: a review and
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17. 18.
19.
20.
169
AEP Vol. 13, No. 3 March 2003: 163–169
Hoff et al. CAD RISK FACTORS AND CORONARY ARTERY CALCIUM
21. Dawber TR. The Framingham Study: The Epidemiology of Atherosclerotic Disease. Cambridge: Harvard University Press; 1980. 22. Kannel WB, Castelli WP, Gordon T, McNamara PM. Serum cholesterol, lipoproteins, and the risk of coronary heart disease: The Framingham Study. Ann Intern Med. 1971;74:1–12. 23. The Pooling Project Research Group. Relationship of blood pressure, serum cholesterol, smoking habit, relative weight and ECG abnormalities to incidence of major coronary events: Final report of the Pooling Project. J Chron Dis. 1978;31:201–306. 24. Wong ND, Kouwabunpat D, Vo AN, Detrano RC, Eisenberg H, Goel M, et al. Coronary calcium and atherosclerosis by ultrafast computed tomography in asymptomatic men and women: relation to age and risk factors. Am Heart J. 1994;127:422–430. 25. Guerci AD, Spadaro LA, Goodman KJ, Lledo-Perez A, Newstein D, Lerner G, et al. Comparison of electron beam computed tomography scanning and conventional risk factor assessment for the prediction of angiographic coronary artery disease. J Am Coll Cardiol. 1998;32: 673–679. 26. Eaker ED, Packard B, Wenger NK, Clarkson TB, Tyroler HA. Coronary artery disease in women. Am J Cardiol. 1988;61:641–644. 27. Wenger NK. Coronary heart disease: an older woman’s major health risk. BMJ. 1997;315:1085–1090.
28. Kannel WB. Metabolic risk factors for coronary heart disease in women: perspective from the Framingham Study. Am Heart J. 1987; 114:413–419. 29. Kuhn FE, Rackley CE. Coronary artery disease in women: Risk factors, evaluation, treatment and prevention. Arch Intern Med. 1993; 153:2626– 2636. 30. Murdaugh CL, O’Rourke RA. Coronary heart disease in women: special considerations. Curr Probl Cardiol. 1988;13:73–156. 31. Anderson KM, Wilson PW, Odell PM, Kannel WB. An updated coronary risk profile: a statement for health professionals. Circulation. 1991;83: 356–362. 32. Kannel WB. Nutrition and the occurrence and prevention of cardiovascular disease in the elderly. Nutr Rev. 1988;46:68–78. 33. Dobson AJ, Alexander HM, Heller RF, Lloyd DM. Passive cigarette use and the risk of heart attack or coronary death. Med J Aust. 1991; 154:793–797. 34. Garg R, Madans JH, Kleinman JC. Regional variation in ischemic heart disease incidence. J Clin Epidemiol. 1992;45:149–156. 35. Howard G, Manolio TA, Burke GL, Wolfson SK, O’Leary DH. Does the association of risk factors and atherosclerosis change with age? An analysis of the combined ARIC and CHS cohorts. The Atherosclerosis Risk in Communities (ARIC) and Cardiovascular Health Study (CHS) investigators. Stroke. 1997;28:1693–1701.