Despite a significant reduction in coronary artery disease

Transcription

1 Clinical Investigation and Reports Identification of Patients at Increased Risk of First Unheralded Acute Myocardial Infarction by Electron-Beam Computed Tomography Paolo Raggi, MD; Tracy Q. Callister, MD; Bruce Cooil, PhD; Zuo-Xiang He, MD; Nicholas J. Lippolis, MD; Donald J. Russo, MD; Alan Zelinger, MD; John J. Mahmarian, MD Background There is a clear relationship between absolute calcium scores (CS) and severity of coronary artery disease. However, hard coronary events have been shown to occur across all ranges of CS. Methods and Results We conducted 2 analyses: in group A, 172 patients underwent electron-beam CT (EBCT) imaging within 60 days of suffering an unheralded myocardial infarction. In group B, 632 patients screened by EBCT were followed up for a mean of 32 7 months for the development of acute myocardial infarction or cardiac death. The mean patient age and prevalence of coronary calcification were similar in the 2 groups (53 8 versus 52 9 years and 96% each). In group B, the annualized event rate was 0.11% for subjects with CS of 0, 2.1% for CS 1 to 99, 4.1% for CS 100 to 400, and 4.8% for CS 400, and only 7% of the patients had CS 400. However, mild, moderate, and extensive absolute CSs were distributed similarly between patients with events in both groups (34%, 35%, and 27%, respectively, in group A and 44%, 30%, and 22% in group B). In contrast, the majority of events in both groups occurred in patients with CS 75th percentile (70% in each group). Conclusions Coronary calcium is present in most patients who suffer acute coronary events. Although the event rate is greater for patients with high absolute CSs, few patients have this degree of calcification on a screening EBCT. Conversely, the majority of events occur in individuals with high CS percentiles. Hence, CS percentiles constitute a more effective screening method to stratify individuals at risk. (Circulation. 2000;101: ) Key Words: calcium tomography myocardial infarction coronary disease Despite a significant reduction in coronary artery disease (CAD) mortality, during the past decade the incidence of acute myocardial infarction (AMI) has continued to rise in the United States. 1 Several studies have shown the potential for aggressive lipid-lowering therapy to influence CAD outcome, 2,3 but our ability to predict and prevent the majority of events remains limited. Kannel 4 recently suggested that to maximize the cost-effectiveness of primary prevention, it is necessary to target high-risk candidates by use of multifactorial risk-stratification approaches and more sophisticated diagnostic technologies than those currently in use. At present, the possibility that electron-beam CT (EBCT) may provide such an opportunity is being investigated. Coronary artery calcium has long been identified as a marker of underlying CAD, 5 9 and EBCT is extremely sensitive for the detection and quantification of the extent of coronary artery calcification (CAC). Measures of CAC expressed as calcium scores (CSs) show a close correlation with the atherosclerotic plaque burden, and the simple presence of CAC on a screening EBCT test has been reported to be associated with high odds ratios for developing a variety of cardiovascular events. 14 However, although a clear correlation exists between increasing absolute CS, severity of coronary luminal stenosis, 15 and subsequent revascularization procedures, 14,16 previous investigations consistently showed that patients suffered coronary events across all ranges of absolute CSs Indeed, it is likely that age- and sex-adjusted CS values may be more appropriate than absolute CS to assess the individual risk, because they may better reflect the underlying individual atherogenic process. In this study, we compared the value of absolute and relative CSs, based on expected values for age- and sex- Received May 14, 1999; revision received September 14, 1999; accepted September 29, From EBT Research Foundation (T.Q.C., P.R., N.J.L., D.J.R.) and Vanderbilt University (B.C.), Nashville, Tenn; Baylor College of Medicine, Houston, Tex (Z.-X.H., J.J.M.); and Christ Hospital and Medical Center and Rush Medical College, Oak Lawn, Ill (A.Z.). Presented in part at the 48th Meeting of the American College of Cardiology, New Orleans, La, March 7 11, Tracy Q. Callister, Paolo Raggi, Nicholas J. Lippolis, Donald J. Russo. How should we use coronary artery calcium scores to predict events? J Am Coll Cardiol. 1999;33:414A. Abstract. Guest Editor for this article was Bruce Brundage, MD, Bend Memorial Clinic, Bend, Ore. Drs Callister and Russo own publicly traded shares of Imatron Inc, South San Francisco, Calif, the maker of electron-beam computed tomography scanners. Correspondence to Paolo Raggi, MD, FACP, FACC, Director, EBT Research Foundation, 64 Valleybrook Dr, Hendersonville, TN praggi@attglobal.net 2000 American Heart Association, Inc. Circulation is available at 850

2 Raggi et al Prediction of Hard Coronary Events 851 TABLE 1. Calcium Score Nomogram for 9728 Consecutive Subjects Age, y Men (5433) (479) (859) (1066) (1085) (853) (613) (478) 25th percentile th percentile th percentile th percentile Women (4297) (288) (589) (822) (903) (693) (515) (485) 25th percentile th percentile th percentile th percentile The number of patients in each group is in parentheses. matched control subjects, for predicting hard cardiac events. We performed EBCT scanning in 2 distinct patient groups. (1) Group A consisted of 172 patients who had suffered an AMI as their first clinical manifestation of CAD and who underwent EBCT imaging shortly after their acute events. In this group, the imaging was conducted to estimate the extent of plaque burden that could have been identified had EBCT scanning been performed before the patients suffered an AMI. (2) A telephone survey for the occurrence of hard events was conducted in 632 patients who had originally been referred for a screening EBCT by their primary care physicians because of the presence of risk factors for CAD (group B). The average follow-up was 32 7 months. In both groups, absolute CSs and percentiles were calculated. Methods Patient Selection One hundred seventy-two patients who had suffered an AMI with no previous clinical manifestation of CAD were enrolled in group A. The diagnosis of AMI was based on standard clinical and ECG criteria as well as a characteristic rise and fall in plasma creatine kinase MB fraction. 21,22 Patients who had Q-wave and non Q-wave AMIs were included. The mean and median times between AMI and EBCT imaging were 31 and 35 days, respectively (range, 3 to 60 days). Because of potential interference with the CS calculations, patients who had undergone coronary angioplasty and stent placement in the peri-infarction period were excluded, whereas plain angioplasty was allowed. This part of the study was performed to assess the presence and extent of CAC that could have been identified had these patients undergone an EBCT screening test before suffering a nonfatal AMI. Six hundred thirty-two asymptomatic patients, referred for EBCT screening by primary care physicians because of the presence of risk factors for CAD, constituted the cohort for the second portion of our study (group B). None of the subjects were self-referred. Clinical follow-up for cardiac events was conducted via telephone interview after a mean period of 32 7 months (range, 24 to 48 months) from the initial screening. Follow-up was completed in all patients. Diagnosis of AMI was confirmed by means of medical record review and the same clinical, laboratory, and ECG criteria as used for the post-ami patients described above. Sudden cardiac death or death due to AMI was verified through review of the death certificates or hospital records and was performed by an investigator blinded to the results of the EBCT screening scan. Data on risk factors for CAD were collected by a cardiac nurse before the initial EBCT imaging procedure. Arterial hypertension was defined by current use of antihypertensive medications or known but untreated hypertension. Current smoking was necessary for the definition of positive smoking status. Hypercholesterolemia was defined as currently receiving cholesterol-lowering medications or the presence of known but untreated hypercholesterolemia. Patients currently receiving insulin or oral hypoglycemic agents were classified as diabetic. A family history of premature CAD was considered present if CAD had occurred in a first- or second-degree relative at 55 years of age. All patients gave informed consent to participate in the study, and the internal review boards of our institutions approved the research protocol. Imaging Protocol All patients underwent EBCT imaging with an Imatron C-100 or C-150 scanner, those in group A within 6 weeks of AMI. Images were obtained with 100-ms scan time and 3-mm single-slice thickness, with a total of 40 slices starting at the level of the carina and proceeding to the level of the diaphragm. Tomographic imaging was electrocardiographically triggered to 80% of the R-R interval. Coronary calcification was defined as a plaque of 4 consecutive pixels (area 1.37 mm 2 ) with a density of 130 HU. Quantitative CSs were calculated according to the method described by Agatston et al. 23 All EBCT scans were reviewed in random order by 3 separate investigators, and because calcium scoring was performed only once on each patient, interinvestigator score variability was not calculated. Statistical Analysis Four absolute CS categories were considered: zero, mild score (CS 1 to 99), moderate score (CS 100 to 400), and severe score ( 400). CSs were also expressed as age- and sex-adjusted CS percentiles derived from a group of 9728 subjects who underwent EBCT imaging at 1 center (EBT Research Foundation, Nashville, Tenn; Table 1). These patients had a distribution of risk factors for CAD similar to those of the patients in the study groups: systemic hypertension, 47%; diabetes mellitus, 11%; smoking, 43%; hypercholesterolemia, 56%; and family history of CAD, 69% (see Table 2 for comparison). For patients in the prospective cohort, the absolute event rates, risk ratios, and relative risk of events were calculated in each CS and risk factor quartile. The quartiles of CS percentile were created by subdividing the population into 4 numerically equal groups. The event rate was derived by assigning to each numerically balanced quartile group the observed number of events in each quartile. Risk factor quartile groups were created by including the following risk factors: men 45 years old, women 55 years old, current smoking, systemic hypertension, diabetes mellitus, hypercholesterolemia, and family history of premature CAD.

3 852 Circulation February 29, 2000 ANOVA and 2-sample t tests were used to compare the means of continuous variables. 2 analysis and the 2-sample test on proportions were used to compare categorical variables. All cited probability values are 2-tailed, and a value of P 0.05 was considered statistically significant. All data are expressed as mean SD. Results Post-AMI Cohort The clinical characteristics of group A patients are shown in Table 2. The overall mean CS was , and this was similar in the 126 men ( ) and 46 women ( ). The median CS for the entire group was 160. EBCT imaging showed CAC in 165 patients (96%). Four women and 3 men did not have CAC, and although their mean age was lower than that of patients with CAC, this difference did not reach statistical significance (47 8 versus 53 8; P 0.058). Of note, all patients without CAC at the time of AMI were smokers. The individual CSs were mild (1 to 99) in 34%, moderate (100 to 400) in 35%, and severe ( 400) in 27% of the subjects (Table 3; P NS for all comparisons by 2 analysis). Table 3 shows the distribution of CS percentiles found in study patients in relation to the age- and sex-adjusted values presented in Table 1. The majority of group A patients (87%) hadacs 50th percentile for age and sex, 70% showed a CS 75th percentile, and 42% showed a CS 90th percentile. All comparisons of observed and expected prevalence of CSs were highly statistically significant (87% versus 50%, 70% versus 25%, and 42% versus 10%; P 0.001). Prospective Cohort The clinical characteristics of group B patients are shown in Table 2. In this cohort, 50% of patients were men, and 54% had CAC on the screening EBCT. The mean CS for the entire cohort was ; median, 3.1. During 32 7 months, 27 patients (4%) had either a fatal cardiac event (n 8) or nonfatal AMI (n 19). Of these 27 patients, 26 (96%) had CAC on the initial EBCT, and their mean and median CSs were statistically greater than in the 605 individuals without events ( versus , P 0.001, and 141 versus 2.4, P 0.001, respectively). TABLE 2. Clinical and EBCT Characteristics Group A (n 172) Group B, Hard Event (n 27) Group B, No Event (n 605) Men 126 (73) 15 (56) 301 (50) Age, y (mean SD) Patients with CAC 165 (96)* 26 (96) 314 (52)* CS (mean SD) Current smokers 112 (65) 20 (74) 247 (40) Systemic hypertension 89 (52) 17 (63) 289 (48) Diabetes mellitus 26 (15) 7 (26) 64 (11) Hypercholesterolemia 106 (62) 19 (70) 359 (59) Family history of CAD 81 (47) 20 (74) 434 (72) Values are n (%) where appropriate. *P ; P TABLE 3. Distribution of Absolute CS Values and CS Percentiles Group A (n 172) Group B, Hard Event (n 27) Group B, No Hard Event (n 605) Absolute CS 0 7 (4) 1 (4) 291 (48) (34) 12 (44) 207 (34) (35) 8 (30) 66 (11) (27) 6 (22) 41 (6) CS percentiles 50th percentile 150 (87) 23 (85) 258 (43) 75th percentile 121 (70) 19 (70) 162 (27) 90th percentile 72 (42) 11 (41) 82 (14) Values are n (%). Prevalence of mild, moderate, and severe CS and CS percentiles for the patients in group B are shown in Table 3. As in the retrospective analysis, 2/3 of the events occurred in patients with mild to moderate absolute CS values (74% in group B versus 69% in group A, P NS), and 1/4 occurred in patients with severe CS (22% in group B versus 27% in group A; P NS). These results are compatible with angiographic and pathological findings demonstrating that the majority of patients who suffer an AMI or sudden cardiac death have nonobstructive coronary artery lesions The annualized event rate was 0.11% for the 292 subjects with a CS of 0, compared with 2.1% for those with a CS of 1 to 99, 4.1% for those with a CS of 100 to 400, and 4.8% for those with a CS 400. Although 22% of the events occurred in patients with a CS 400, only 7% of the entire cohort showed a CS in this range, and the majority of events occurred in patients with mild or moderate CSs. Hence, although a larger absolute CS identified a population at high risk of events, these values reflected only a small segment of the population at risk and would therefore not constitute an optimal screening method. Conversely, as shown in the retrospective group, the majority of patients who suffered an acute AMI in group B had a CS percentile value in the upper range of normal (85% 50th percentile, 70% 75th percentile, and 41% 90th percentile). Again, the comparisons of observed and expected prevalence of CS percentiles were all highly statistically significant (85% versus 50%, 70% versus 25%, and 41% versus 10%; P 0.001). Limiting the analysis to patients who showed a CS 75th percentile demonstrated that a much greater proportion of events occurred in patients in this quartile group than in all other lower quartile groups considered together (19 events in 181 patients versus 8 events in 451 patients; P 0.001). Risk Analysis of Prospective Patient Cohort Table 4 shows a comparison of total number of events, annualized absolute event rates, and odds ratios for an event in the 27 prospective patients identified according to quartiles of CS percentiles and risk factors. The risk ratio for events in the highest risk factor quartile was 6.5-fold greater than in the

4 Raggi et al Prediction of Hard Coronary Events 853 TABLE 4. Total Number of Events, Annualized Event Rates, and Odds Ratios (Relative to the First Quartile) for Quartiles of CS and Risk Factors in the Prospective Patient Group (632 Patients) No. of Events (Annualized Event Rate) Odds Ratio (95% CI) Quartile CS% Risk Factors CS% Risk Factors 1st 1 (0.2%) 2 (0.5%) nd 1 (0.2%) 6 (1.4%) 1.0 (0.1, 16.1) 3.1 (0.6, 15.5) 3rd 6 (1.4%) 6 (1.4%) 6.2 (0.7, 52.1) 3.1 (0.6, 15.5) 4th 19 (4.5%) 13 (3.1%) 21.5 (2.8, 162.4) 7.0 (1.6, 31.5) CS% indicates calcium score percentile. lowest quartile (13/2 6.5). In contrast, the risk ratio for patients in the highest CS quartile (ie, CS 75th percentile) was 19 times that of patients in the lowest quartile (19/1 19). Furthermore, the odds ratios for events in the upper 2 quartiles of CS percentiles were substantially greater than in the respective risk factor quartiles (6.2 and 21.5 versus 3.1 and 7.0, respectively), although there was some overlapping of the CIs. The annualized relative risk of events (Table 5) ranged from 0.09 to 1.05 (12-fold difference) between the lowest and the highest quartiles in patients identified according to risk factors and from to 2.7 (59-fold difference) when the grouping was done according to CS quartiles. This analysis suggests that although traditional risk factors for CAD are very valuable predictors, percentiles of CS are a substantially better means of identifying patients at risk of hard coronary events. Discussion Calcification of the coronary arteries, identified and measured on EBCT imaging, is highly prevalent in patients suffering an acute AMI and strongly predictive of hard events. This study demonstrates that CS percentiles, based on patient age and sex, are better predictors than absolute CS values and categorical risk factors for CAD. In this regard, Wilson et al 27 recently showed that the presence of categorical risk factors for CAD is as predictive of cardiovascular events as the measurement of such continuous variables as systolic hypertension and LDL cholesterol levels. Therefore, the assumptions made in this study about the relative ability of risk factor categories to predict events in comparison to CS categories are pertinent. As previously shown, CAC is present with a very high prevalence in patients who suffer a coronary event, 16,25,26,28,29 indicating that this is a true marker of CAD. However, previous studies showed an apparent paradox: patients with high CSs were shown to be at high risk of hard events, but only a small portion of the population would show high CS, TABLE 5. Annualized Relative Risk of Hard Events in 1st and 4th Quartile of CS and Risk Factors in the Prospective Patient Group (632 Patients) Quartile CS Percentiles Risk Factors 1st th and events were evenly distributed across all ranges of absolute CS. 16,19 Our study suggests that high CS percentiles are highly prevalent in patients suffering a hard event independent of the absolute CS and that these measurements may allow a better segregation of the portion of an asymptomatic population that may benefit from therapeutic interventions. The process of coronary calcification is currently believed to be an active process similar in many aspects to true bone formation Some investigators view this process as a repair mechanism to the damage caused by various noxious stimuli. 17 A larger-than-expected amount of coronary calcium might indicate the presence of an aggressive damage of the arterial wall requiring either a faster or a more extensive repair. In this context, the amount of coronary calcium relative to the patient s age and sex could be seen as an indicator of plaque activity rather than plaque quiescence. Rosamond et al 1 showed that in recent years, the incidence of AMI has continued to increase even though the mortality rate from AMI has declined. Hence, primary prevention of CAD must remain the focus of public health policies of industrialized nations. The Framingham equations 36,37 provide a population-based method to assess the median risk of developing CAD but do not allow discernment of the individual s risk. The National Cholesterol Education Program (NCEP)-II guidelines 38 provide a very helpful aid for the prevention of CAD in people at risk, but published data suggest a maximum 3- to 5-fold increase in estimated risk between patients with the lowest and the highest risk profiles Sophisticated computer simulations have been attempted to improve on the predictive ability of these models, but with poor applicability to the everyday clinical practice. 39 Furthermore, according to the NCEP-II guidelines, 21.1 million Americans ( 17% of the adult population) need treatment for the presence of risk factors for CAD. 42 However, even when more liberal treatment criteria are applied, as suggested in the recent Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS) trial, 3 only 37% of the AMIs are prevented. Conversely, our EBCT analysis suggests that a larger number of hard events could be addressed by implementation of this screening strategy. If patients with CS 75th percentile were treated, 70% of the events could potentially be prevented, and if patients with CS 90th percentile were treated, 40% of the events could be prevented. Hence, EBCT screening for coronary calcification

5 854 Circulation February 29, 2000 should improve the practicing physician s ability to stratify individuals at high risk of events, to whom aggressive treatment of traditional risk factors for CAD can be more appropriately directed. Of note, the data presented in this study confirm previous observations pertaining to the absence of CAC on EBCT imaging in a percentage of young smokers who suffer an acute coronary event. 16,26,28,29,43 Similar findings have been reported in pathological studies conducted in young individuals who died of cardiac arrest or AMI This indicates that prudence should be used in characterizing the absence of CAC or the presence of small amounts of CAC in young smokers as a nonthreatening marker for acute coronary events. Study Limitations Our postevent imaging analysis in group A was limited to nonfatal AMI cases. However, nonfatal AMI is the most prevalent clinical manifestation of coronary heart disease in men and the second most prevalent in women, and its incidence continues to rise in recent years. 1,47,48 We can only conjecture that the amount of CAC found in patients who die of an acute coronary event is at least as extensive as, if not greater than, the amount found in patients who suffer a nonfatal event. Interestingly, in the prospective group, the patients who died had a larger amount of coronary calcification than those who suffered a nonfatal AMI (median CS, 391 versus 75, P 0.04). Finally, the presence of risk factors for CAD was not directly assessed but rather inferred from current medical regimens or patient self-reporting. Conclusions The incidence of cardiovascular events is greatly increased in the presence of CAC. Although high CSs portend a high risk of events, only a relatively small portion of the AMIs occur in this fraction of the population, and the majority of events occur in patients with mild to moderate amounts of CAC. Conversely, high age- and sex-adjusted CS percentiles appear to be closely related to the occurrence of subsequent hard events, and we suggest that this measurement should be used in the assessment of the risk of a hard event in asymptomatic individuals undergoing EBCT screening. Acknowledgment This study was supported in part by the Dean s Fund for Faculty Research, Owen Graduate School of Management, Vanderbilt University. References 1. Rosamond WD, Chambless LE, Folsom AR, Cooper LS, Conwill DE, Clegg L, Wang CH, Heiss G. Trends in the incidence of myocardial infarction and in mortality due to coronary heart disease, 1987 to N Engl J Med. 1998;339: Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR, MacFarlane PW, McKillop JH, Packard CJ, West of Scotland Coronary Prevention Study Group. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. N Engl J Med. 1995;333: Downs JR, Clearfield M, Weis S, Whitney E, Shapiro DR, Beere PA, Langendorfer A, Stein EA, Kruyer W, Gotto AM Jr. Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: results of AFCAPS/TexCAPS research. JAMA. 1998; 279: Kannel WB. The worth of controlling plasma lipids. Am J Cardiol. 1998;81: Blankenhorn DH, Stern D. Calcification of the coronary arteries. Am J Roentgenol. 1959;81: Beadenkopf WG, Daoud AS, Love BM. Calcification of the coronary arteries and its relation to arteriosclerosis and myocardial infarction. Am J Roentgenol. 1964;92: Eggen DA, Strong JP, McGill HC. Coronary calcification: relationship to clinically significant coronary lesions and race, sex, and topographic distribution. Circulation. 1965;92: Tanenbaum SR, Kondos GT, Veselik KE, Prendergast MR, Brundage BH, Chomka EV. Detection of calcific deposits in coronary arteries by ultrafast computed tomography and correlation with angiography. Am J Cardiol. 1989;63: Breen JF, Sheedy PF II, Schwartz RS, Stanson AW, Kaufmann RB, Moll PP, Rumberger JA. Coronary artery calcification detected with ultrafast CT as an indication of coronary artery disease. Radiology. 1992;185: Simons DB, Schwartz RS, Edwards WD, Sheedy PF, Breen JF, Rumberger JA. Noninvasive definition of anatomic coronary artery disease by ultrafast computed tomographic scanning: a quantitative pathologic comparison study. J Am Coll Cardiol. 1992;20: Mautner SL, Mautner GC, Froehlich J, Feuerstein IM, Proschan MA, Roberts WC, Doppman JL. Coronary artery calcification: assessment with electron beam CT and histomorphometric correlation. Radiology. 1994;192: Rumberger JA, Simons DB, Fitzpatrick LA, Sheedy PF, Schwartz RS. Coronary artery calcium area by electron-beam computed tomography and coronary atherosclerotic plaque area: a histopathologic correlative study. Circulation. 1995;15: Sangiorgi G, Rumberger JA, Severson A, Edwards WD, Gregoire J, Fitzpatrick LA, Schwartz RS. Arterial calcification and not lumen stenosis is highly correlated with atherosclerotic plaque burden in humans: a histologic study of 723 coronary artery segments using non-decalcifying methodology: electron beam computed tomography and coronary artery disease: scanning for coronary artery calcification. J Am Coll Cardiol. 1998;31: Arad Y, Spadaro LA, Goodman K, Lledo-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: Kennedy J, Shavelle R, Wang S, Budoff M, Detrano RC. Coronary calcium and standard risk factors in symptomatic patients referred for coronary angiography. Am Heart J. 1998;135: 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: Wexler L, Brundage B, Crouse J, Detrano R, Fuster V, Maddahi J, Rumberger J, Stanford W, White R, Taubert K. Coronary artery calcification: pathophysiology, epidemiology, imaging methods, and clinical implications: a statement for health professionals from the American Heart Association Writing Group. Circulation. 1996;94: Schwartz RS. Coronary artery calcification and prognosis: same song, second verse? J Am Coll Cardiol. 1994;24: Wong ND, Detrano RC, Abrahamson D, Tobis JM, Gardin JM. Coronary artery screening by electron beam computed tomography: facts, controversy, and future. Circulation. 1995;92: Detrano R, Hsiai T, Wang S, Puentes G, Fallavollita J, Shields P, Stanford W, Wolfkiel C, Georgiou D, Budoff M, Reed J. Prognostic value of coronary calcification and angiographic stenoses in patients undergoing coronary angiography. J Am Coll Cardiol. 1996;27: Jaffe AS. Biochemical detection of acute myocardial infarction. In: Gersh BJ, Rahimtoola SH, eds. Acute Myocardial Infarction. New York, NY: Elsevier; 1991: Dissmann R, Lindener T, Schroder R. Estimation of enzymatic infarct size: direct comparison of the marker enzymes creatine kinase and alphahydroxybutyrate dehydrogenase. Am Heart J. 1998;135: Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol. 1990;15: Davies MJ, Thomas AC. Plaque fissuring: the cause of acute myocardial infarction, sudden ischaemic death, and crescendo angina. Br Heart J. 1985;53: Ambrose JA, Tannenbaum MA, Alexopoulos D, Hjemdahl-Monsen CE, Leavy J, Weiss M, Borrico S, Gorlin R, Fuster V. Angiographic pro-

6 Raggi et al Prediction of Hard Coronary Events 855 gression of coronary artery disease and the development of myocardial infarction. J Am Coll Cardiol. 1988;1: Wilson PWF, D Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation. 1998;97: Schmermund A, Baumgart D, Gorge G, Seibel R, Gronemeyer D, Ge J, Haude M, Rumberger J, Erbel R. Coronary artery calcium in acute coronary syndromes: a comparative study of electron-beam computed tomography, coronary angiography, and intracoronary ultrasound in survivors of acute myocardial infarction and unstable angina. Circulation. 1997;96: Schmermund A, Baumgart D, Adamzik M, Ge J, Gronemeyer D, Seibel R, Sehnert C, Gorge G, Haude M, Erbel R. Comparison of electron-beam computed and intracoronary ultrasound in detecting calcified and noncalcified plaques in patients with acute coronary syndromes and no or minimal to moderate angiographic coronary artery disease. Am J Cardiol. 1998;81: Shemesh J, Stroh CI, Tenenbaum A, Hod H, Boyko V, Fisman EZ, Motro M. Comparison of coronary calcium in stable angina pectoris and in first acute myocardial infarction utilizing double helical computerized tomography. Am J Cardiol. 1998;81: Bostrom K, Watson KE, Horn S, Wortham C, Herman IM, Demer LL. Bone morphogenetic protein expression in human atherosclerotic lesions. J Clin Invest. 1993;91: Fitzpatrick LA, Severson A, Edwards WD, Ingram RT. Diffuse calcification in human coronary arteries: association of osteopontin with atherosclerosis. J Clin Invest. 1994;94: Ikeda T, Shirasawa T, Esaki Y, Yoshiki S, Hirokawa K. Osteopontin mrna is expressed by smooth muscle-derived foam cells in human atherosclerotic lesions of the aorta. J Clin Invest. 1993;92: Shanahan CM, Cary NR, Metcalfe JC, Weissberg PL. High expression of genes for calcification-regulating proteins in human atherosclerotic plaques. J Clin Invest. 1994;93: Kragel AH, Reddy SG, Wittes JT, Roberts WC. Morphometric analysis of the composition of the atherosclerotic plaques in the four major epicardial coronary arteries in acute myocardial infarction and in sudden coronary death. Circulation. 1989;80: Deleted in proof. 36. Anderson KM, Wilson PWF, Odell PM, Kannel WB. An updated coronary risk profile: a statement for health professionals. Circulation. 1991; 83: Abbott RD, McGee D. The probability of developing certain cardiovascular disease in eight years at specified values of some characteristics. In: Kannel WB, Wolf PA, Garrison RJ, eds. The Framingham Study: An Epidemiological Investigation of Cardiovascular Disease. Washington, DC: US Department of Health, Education, and Welfare; 1987: section 37. Publication NIH National Cholesterol Education Program. Second report of the expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel II). Circulation. 1994;89: Grover SA, Coupal L, Hu XP. Identifying adults at increased risk of coronary disease. JAMA. 1995;274: Kinosian B, Glick H, Garland G. Cholesterol and coronary heart disease: predicting risk by levels and ratios. Ann Intern Med. 1994;121: Silagy C, Mant D, Carpenter L, Muir J, Neil A. Modelling different strategies to prevent coronary heart disease in primary care. J Clin Epidemiol. 1994;47: Gotto AM. Lipid-lowering therapy for the primary prevention of coronary heart disease. J Am Coll Cardiol. 1999;33: Callister TQ, Raggi P, Lippolis NJ, Russo DJ. Assessing the effects of smoking on coronary arteriosclerosis by electron beam computed tomography. Chest. 1998;114:310S. Abstract. 44. Virmani R, Burke A, Farb A. Coronary risk factors and plaque morphology in men with coronary disease who died suddenly. Eur Heart J. 1998;19: Burke AP, Farb A, Malcom GT, Liang Y, Smialek J, Virmani R. Effect of risk factors on the mechanism of acute thrombosis and sudden coronary death in women. Circulation. 1998;97: Virmani R, Burke AP, Farb A. Plaque morphology in sudden coronary death. Cardiologia. 1998;43: Lerner D, Kannel WB. Patterns of coronary heart disease morbidity and mortality in the sexes: a 26-year follow-up of the Framingham population. Am Heart J. 1986;111: Sytkowski PA, D Agostino RB, Belanger A, Kannel WB. Sex and time trends in cardiovascular disease incidence and mortality: the Framingham Heart Study, Am J Epidemiol. 1996;143: