Comparative Value of Coronary Artery Calcium and Multiple Blood Biomarkers for Prognostication of Cardiovascular Events

Comparative Value of Coronary Artery Calcium and Multiple Blood Biomarkers for Prognostication of Cardiovascular Events Jamal S. Rana, MD a,b, Heidi Gransar, MS a, Nathan D. Wong, PhD c, Leslee Shaw, PhD d, Michael Pencina, PhD e, Khurram Nasir, MD f, Alan Rozanski, MD g, Sean W. Hayes, MD a, Louise E. Thomson, MB, ChB a, John D. Friedman, MD a,b, James K. Min, MD a,b, and Daniel S. Berman, MD a,b, * The value of coronary artery calcium (CAC) scoring versus multiple biomarkers in increasing risk prediction for cardiovascular disease (CVD) remains unknown. The study group consisted of 1,286 asymptomatic participants (mean SD 59 8 years old) with no known coronary heart disease. Mean follow-up time was 4.1 0.4 years with the primary outcome of combined CVD (cardiac death, myocardial infarction, stroke, and late target vessel revascularization). CAC was calculated by the method of Agatston. Biomarkers measured were C-reactive protein, interleukin-6, myeloperoxidase, B-type natriuretic peptide, and plasminogen activator-1. During follow-up 35 participants developed CVD events including cardiac deaths (6%), myocardial infarction (23%), strokes (17%), and late revascularizations (54%). In Cox proportional-hazards models adjusted for Framingham Risk Score (FRS), presence of log CAC beyond FRS was associated with increased hazards for CVD events (hazard ratio 1.7, 95% confidence interval [CI] 1.4 to 2.0, p <0.001). Multiple biomarkers score was also associated with increased risk beyond FRS (hazard ratio 2.1, p 0.02) per 1-U increase in score; however, the c-statistic did not increase significantly (0.75, 95% CI 0.68 to 0.84, p 0.32). The c-statistic increased when log CAC was incorporated into FRS without or with multiple biomarkers score (c-statistic 0.84, p 0.003 and p 0.008 respectively). Addition of CAC to risk factors showed significant reclassification (net reclassification improvement 0.35 (95% CI 0.11 to 0.58, p 0.007; integrated discrimination index 0.076, p 0.0001), whereas addition of multiple biomarkers score did not show significant reclassification. In conclusion, in this study of asymptomatic subjects without known CVD, addition of CAC but not biomarkers substantially improved risk reclassification for future CVD events beyond traditional risk factors. 2012 Published by Elsevier Inc. (Am J Cardiol 2012;109:1449 1453) a Department of Imaging, Cedars-Sinai Medical Center, Los Angeles, California; b Department of Medicine, Cedars-Sinai Heart Institute, Los Angeles, California; c Division of Cardiology, University of California, Irvine, California; d Emory University School of Medicine, Atlanta, Georgia; e Department of Biostatistics, School of Public Health, Boston University, Harvard Clinical Research Institute, Boston, Massachusetts; f Johns Hopkins Ciccarone Preventive Cardiology Center, Baltimore, Maryland; g Department of Cardiology, St. Luke s-roosevelt Hospital Center, New York, New York. Manuscript received October 21, 2011; revised manuscript received and accepted January 2, 2012. This study was supported by a grant from the Eisner Foundation, Los Angeles, California. *Corresponding author: Tel: 310-423-4224; fax: 310-423-0823. E-mail address: Bermand@cshs.org (D.S. Berman). In population-based studies, coronary artery calcium (CAC) scoring has emerged as a robust predictor of incident cardiovascular disease (CVD) events and provides incremental predictive information beyond that yielded by standard risk factors. 1 3 In this regard, the most recent American College of Cardiology/American Heart Association guideline for assessment of cardiovascular risk in asymptomatic adults recommends use of CAC scanning as a class IIa indication for risk stratification of asymptomatic patients with an intermediate risk of events. 4 In several recent studies, use of multiple blood biomarkers has also shown promise for the prediction of future CVD events. 5 7 Given the potential utility of other blood biomarkers for risk assessment and the absence of available data directly comparing CAC to multiple biomarkers, we evaluated asymptomatic subjects within the Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research (EISNER) prospective randomized trial who underwent evaluation by CAC and measure of multiple biomarkers. 8 Methods The study population represented a subset of 1,286 adults (mean 58.6 8.5 years old, 47% women) without known CVD or symptoms within the EISNER research study, as we have previously reported. 8 The present study participants represent those for whom consecutive CAC and biomarker data were available. The present study was approved by the Cedars-Sinai Medical Center institutional review board. A dedicated medical history was taken to assess previous cardiac disease, diabetes, and medication use. Collection of fasting lipid profile and measurements of blood pressure and body mass index has been detailed previously. 8 Subjects were scanned using electron beam computed tomography 0002-9149/12/$ see front matter 2012 Published by Elsevier Inc. www.ajconline.org doi:10.1016/j.amjcard.2012.01.358

1450 The American Journal of Cardiology (www.ajconline.org) Table 1 Characteristics of study population Characteristics Total Population (n 1,286) No (n 1,251) Events Yes (n 35) Age (years) 58.6 8.5 58.5 8.5 61.5 8.0 Men 679 (52.8%) 655 (52.4%) 24 (68.6%) Systolic blood pressure (mm Hg) 133.1 17.5 132.8 17.2 144.5 22.2 Diastolic blood pressure (mm Hg) 82.0 10.7 81.9 10.7 85.1 12.0 Hypertension 744 (57.9%) 715 (57.2%) 29 (82.9%) Body mass index (kg/m 2 ) 27.5 5.2 27.4 5.2 28.7 5.8 Total cholesterol (mg/dl) 215.8 41.4 215.6 41.5 224.6 37.1 Low-density lipoprotein cholesterol (mg/dl) 136.1 37.6 135.8 37.6 145.7 37.3 High-density lipoprotein cholesterol (mg/dl) 54.2 16.5 54.3 16.6 50.6 13.8 Diabetes mellitus 104 (8.1%) 97 (7.8%) 7 (20.0%) Current smoking 70 (5.4%) 67 (5.4%) 3 (8.6%) C-reactive protein ( g/ml) 5.0 7.6 4.9 7.5 7.1 10.7 Log-transformed 0.8 1.3 0.8 1.3 1.1 1.3 Interleukin-6 (pg/ml) 92.7 785.0 93.2 795.3 74.4 188.6 Log-transformed 3.4 1.0 3.4 1.0 3.4 1.0 Myeloperoxidase (ng/ml) 15.2 34.8 15.3 35.3 13.8 9.2 Log-transformed 2.5 0.5 2.5 0.5 2.5 0.5 B-type natriuretic peptide (pg/ml) 49.3 104.7 48.9 105.4 63.6 75.9 Log-transformed 3.4 0.9 3.4 0.9 3.6 1.0 Plasminogen activator-1 (ng/ml) 15.7 10.2 15.6 10.2 19.0 9.7 Log-transformed 2.5 0.8 2.5 0.8 2.8 0.5 Multiple biomarker score* 0.00 0.6 0.01 0.6 0.34 0.6 Coronary artery calcium 116.3 347.3 98.6 279.4 747.7 1,125.2 Values are presented as mean SD or number of subjects (percentage). * Score 0.119 C-reactive protein 0.049 interleukin-6 0.107 myeloperoxidase 0.348 B-type natriuretic peptide 0.542 plasminogen activator-1 where log-transformed biomarkers are standardized to mean SD of 0 1. Table 2 Biomarkers, coronary artery calcium, and estimation of risk HR Multivariable Adjusted c-statistic HR (95% CI) p Value p Value* Model 1: Framingham Risk Score 1.1 (1.06 1.14) 0.001 0.73 (0.66 0.82) Model 1 plus C-reactive protein 1.2 (0.9 1.7) 0.23 0.73 (0.65 0.82) 0.95 Model 1 plus interleukin-6 1.1 (0.8 1.5) 0.64 0.74 (0.66 0.82) 0.54 Model 1 plus myeloperoxidase 0.8 (0.6 1.2) 0.34 0.73 (0.65 0.82) 1.00 Model 1 plus B-type natriuretic peptide 1.2 (0.8 1.6) 0.37 0.74 (0.66 0.82) 0.68 Model 1 plus plasminogen activator-1 1.4 (0.9 2.0) 0.13 0.74 (0.67 0.82) 0.48 Model 1 plus multiple biomarkers score 2.1 (1.1 3.8) 0.02 0.75 (0.68 0.84) 0.32 Model 1 plus log coronary artery calcium 1.7 (1.4 2.0) 0.001 0.84 (0.78 0.91) 0.003 Model 1 plus log coronary artery calcium plus multiple biomarkers score 1.7 (1.4 2.0) 0.001 0.84 (0.77 0.92) 0.008 Hazard ratio expressed per 1-SD increment in log biomarker concentration and 1-U increase in combined biomarkers risk score. Coronary artery calcium is log-transformed (log [coronary artery calcium 1]). * The c-statistic p values compared to model 1. (General Electric/Imatron, South San Francisco, California) or a multidetector scanner (Siemens, Munich, Germany). Foci of CAC were identified and scored by an experienced technician using semiautomatic commercial software (ScImage, Inc., Los Altos, California) and verified by a board-certified cardiologist. 8,9 The following biomarkers were studied: C-reactive protein (CRP), interleukin-6, myeloperoxidase, B-type natriuretic peptide, and plasminogen activator-1, representing distinct biological pathways for inflammation, left ventricular dysfunction, and fibrinolysis. Serum and ethylenediaminetetraacetic acid plasma samples for all study participants were stored at 70 C for up to 4 years until all baseline samples were accumulated. All biomarker assays were performed by Alere, Inc. (San Diego, California). CRP (range 0.065 to 35 g/ml) was measured by the Luminex Competitive method (Alere, Inc.). Interleukin-6 (range 16 to 60,000 pg/ml) was measured using a microtiter immunoassay. Myeloperoxidase (range 2 to 1,200 ng/ml) was measured by the Luminex Sandwich immunoassay (Alere, Inc., protocol CLN0002, internal Alere myeloperoxidase project documentation). Plasma B-type natriuretic peptide (range

Preventive Cardiology/Coronary Artery Calcium Versus Blood Biomarkers 1451 Table 3 Reclassification based on addition of multiple biomarkers score and log coronary artery calcium to Framingham risk factors for four-year cardiovascular disease risk 4-Year Risk in Model With Multiple Biomarkers Score 2.4% 2.4 8% 8% Overall Reclassified as Higher Risk Reclassified as Lower Risk 4-year risk in model without multiple biomarkers score 2.4% Number of participants 726 71 0 797 Number with no events 720 68 0 788 68 N/A Number of events 6 3 0 9 3 N/A 2.4 8% Number of participants 48 343 20 411 Number with no events 93 276 26 395 26 93 Number of events 2 12 2 16 2 2 8% Number of participants 0 17 54 71 Number with no events 0 15 46 61 N/A 15 Number of events 0 2 8 10 N/A 2 Net reclassification improvement (95% 0.04, 0.13 to 0.21 (0.65) CI, p value) Integrated discrimination improvement (p value) 0.005 (0.26) 4-Year Risk in Model With Log Coronary Artery Calcium 2.4% 2.4 8% 8% Overall Reclassified as Higher Risk Reclassified as Lower Risk 4-year risk in model without log coronary artery calcium 2.4% Number of participants 675 107 15 797 Number with no events 672 104 12 788 116 N/A Number of events 3 3 3 9 6 N/A 2.4 8% Number of participants 216 135 60 411 Number with no events 213 131 51 395 51 213 Number of events 3 4 9 16 9 3 8% Number of participants 14 17 40 71 Number with no events 14 15 32 61 N/A 29 Number of events 0 2 8 10 N/A 2 Net reclassification improvement (95% 0.35, 0.11 0.58, (0.007) CI, p value) Integrated discrimination improvement (p value) 0.076 (0.0001) 6.6 to 2,600 pg/ml) was analyzed in the blood sample anticoagulated with ethylenediaminetetraacetic acid using the quantitative immunofluorescence assay. 10 Plasminogen activator-1 (range 0.1 to 55 ng/ml) was measured using the microtiter immunoassay. Follow-up for CVD events consisted of administering patient questionnaires, patient interviews, and/or using hospital records to obtain outcome data on consented patients. 8 All deaths were verified by the National Death Index and/or independent review of death reports by 2 physicians. Primary end point of the present study was time to first CVD event defined by cardiac death, myocardial infarction, stroke, and late revascularizations ( 90 days after CAC scanning). Mean follow-up time of study for CVD events was 4.1 0.4 years. For all analyses, continuous biomarker variables were standardized after being log-transformed toward normality, resulting in mean SD of 0 1. CAC was logtransformed (log [CAC 1]) toward normality in all the models. We performed multivariable Cox proportional hazards models to examine the association between biomarkers and incident events after verifying that the assumption of proportional hazards was met using Schoenfeld residuals. All models were adjusted for Framingham Risk Score (FRS). 11 Hazard ratios (HRs) were expressed per 1 SD in the log increment in the respective biomarker. Each biomarker was individually tested in models for CVD events with adjustment for conventional risk factors. These analyses included only biomarkers known to have a clinical interest (5 biomarkers for

1452 The American Journal of Cardiology (www.ajconline.org) cardiovascular events). In addition, a risk score was developed 7 using the linear prediction from a fitted Cox model consisting solely of these 5 biologically relevant biomarkers, taking the form of a risk score ( 1 CRP) ( 2 interleukin-6) ( 3 myeloperoxidase) ( 4 B-type natriuretic peptide) ( 5 plasminogen activator-1), where 1 to 5 denote estimates of beta coefficients associated with each standardized log-transformed biomarker. We constructed a Harrell c-statistic and area under the receiver operating characteristics curves for predicting CVD events at 4 years and compared them for CAC and biomarkers. We pursued further discrimination based on combined biomarkers risk score achieving a p value 0.05. We also evaluated the ability of multiple biomarkers score and CAC to reclassify risk, as we previously described. 12,13 Using a Cox model consisting of FRS components, 11 participants were initially classified as low, intermediate, or high risk if their predicted 4-year risk of CVD was 2.4%, 2.4% to 8%, or 8%, respectively. These correspond to FRS cutpoints 6%, 6% to 20%, or 20% for 10-year risk adjusted to 4 years. Cross tabulations of risk categories based on models with and without multiple biomarkers risk score or CAC were performed to describe the number and percentage of participants who were reclassified appropriately (i.e., to a lower risk group for nonevents or to a higher risk group for events) and inappropriately (i.e., to a lower risk group for events or to a higher risk group for nonevents). We calculated the net reclassification improvement and integrated discrimination index. 12 A p value 0.05 was considered statistically significant. All statistical analyses were performed with STATA 11 (STATA Corp., College Station, Texas). Results Baseline characteristics of the 1,286 asymptomatic subjects comprising the study group are presented in Table 1. Thirty-five (2.7%) CVD events occurred, consisting of 2 cardiac deaths (6%), 8 myocardial infarctions (23%), 6 strokes (17%), and 19 late revascularizations (54%). Age, systolic blood pressure, hypertension, log CAC, diabetes, and multiple biomarkers score (p values not shown) were significantly different between the event and nonevent groups. Table 2 presents the relation between CAC and estimation of risk by c-statistic and multivariable adjusted HRs. The c-statistic for FRS alone was 0.73 (95% confidence interval [CI] 0.66 to 0.82). After consideration of FRS, among the 5 biomarkers assessed, none was individually associated with significant improvement in c-statistic or higher risk of CVD events. Beyond FRS, the multiple biomarkers score showed an increased risk of CVD events (HR 2.1, 1.1 to 3.8, p 0.02); however, the c-statistic for multiple biomarkers score was not significant (0.75, 0.68 to 0.84, p 0.32). Beyond FRS, log CAC was associated with a significantly higher risk of CVD events (HR 1.7, 95% CI 1.4 to 2.0, p 0.001) and improved the c-statistic to 0.84 (0.78 to 0.91, p 0.003). After considering FRS and log CAC, adding the multiple biomarkers score did not change the c-statistic for discrimination of CVD events. We evaluated risk reclassification provided by the multiple biomarkers score added to cardiac risk factors. There was no improvement in risk reclassification (Table 3)bynet reclassification improvement (0.04, 95% CI 0.13 to 0.21, p 0.65) or integrated discrimination index (0.005, p 0.26). However, addition of CAC to risk factors successfully reclassified subjects (net reclassification improvement 0.35, 95% CI 0.11 to 0.58, p 0.007; integrated discrimination index 0.076, p 0.0001; Table 3). Discussion In the present study in a population of asymptomatic subjects without known CVD, addition of CAC to risk factors was successful in risk prediction, discrimination, and reclassification for future CVD events. Although a combination of biomarkers improved risk prediction, the combination did not discriminate or reclassify subjects at risk of future CVD events. To our knowledge, this study represents the first of its kind to directly compare the prognostic value of CAC to multiple biomarkers potentially associated with CVD. Limited studies have compared the value of CAC versus a single biomarker. In the St. Francis Heart Study, CAC was predictive of future events of CAC, whereas CRP was not. 14 In a population-based study, Möhlenkamp et al 15 demonstrated improved net reclassification for CAC and CRP but a larger value for CAC (23.8% vs 10.5%). In the Rotterdam study, 16 CRP did not improve the c-statistic when added to clinical risk factors, whereas CAC improved discrimination and reclassification when added to clinical risk factors. Our study findings are consistent with most of these studies and identified no additive value of CRP beyond traditional coronary artery disease risk factors. Further, a multiple biomarkers approach was unsuccessful at discrimination or reclassification of subjects. CAC is a direct measurement of atherosclerosis and in this regard serves as a risk marker rather than a risk factor, as is the case for biomarkers. 17 Our results are in direct accordance findings of the Multi-Ethnic Study of Atherosclerosis (MESA) in which a prognostic utility of CAC was noted over high-sensitivity CRP. 18 In this regard, use of CAC may be a useful strategy to identify at-risk patients who may benefit from targeted risk-modification therapies. 19 We recently reported the long-term outcomes of 2,137 subjects randomized to CAC scan versus no scan in which CAC scans were associated with improvements in 4-year FRS. 8 Costs in the scan group were comparable to those in the no-scan group, balanced by lower and higher resource use for subjects with normal CAC scans versus very high CAC scores, respectively. For prognostic assessment, any new potential marker of disease should add significant predictive information beyond traditionally acquired information or previous techniques that are already in common clinical use. 20 In our study we compared CAC to multiple biomarkers in systematic fashion by assessment of discrimination and reclassification of risk. Despite our study cohort being a low-risk healthy population, CAC demonstrated improved risk stratification, whereas biomarkers did not. Our results agree with a recent commentary by Wilkins and Lloyd-Jones 21 that,

Preventive Cardiology/Coronary Artery Calcium Versus Blood Biomarkers 1453 when used in the context of 10-year risk-estimation equations, current blood biomarkers are unlikely to result in improved discrimination or substantial reclassification for population screening. This study is not without limitations. Although data from the present study represent those from patients prospectively enrolled in a randomized controlled trial, they nevertheless represent subjects from a single center. We measured CRP instead of high-sensitivity CRP; however, we evaluated risk per 1-SD increments for biomarkers to minimize any bias owing to variability of different cutoffs in different assays of a biomarker. Also, given the low event rate, we examined the composite end point of CVD events including late revascularizations and acknowledge a lack of an adequate number of hard events to examine these end points separately. Acknowledgment: The authors acknowledge and appreciate the contributions of Alere, Inc. (San Diego, California) in performing the biomarker assays. 1. Detrano R, Guerci AD, Carr JJ, Bild DE, Burke G, Folsom AR, Liu K, Shea S, Szklo M, Bluemke DA, O Leary DH, Tracy R, Watson K, Wong ND, Kronmal RA. Coronary calcium as a predictor of coronary events in four racial or ethnic groups. N Engl J Med 2008;358:1336 1345. 2. Polonsky TS, McClelland RL, Jorgensen NW, Bild DE, Burke GL, Guerci AD, Greenland P. Coronary artery calcium score and risk classification for coronary heart disease prediction. JAMA 2010;303: 1610 1616. 3. Erbel R, Möhlenkamp S, Moebus S, Schmermund A, Lehmann N, Stang A, Dragano N, Grönemeyer D, Seibel R, Kälsch H, Bröcker- Preuss M, Mann K, Siegrist J, Jöckel KH; Heinz Nixdorf Recall Study Investigative Group. Coronary risk stratification, discrimination, and reclassification improvement based on quantification of subclinical coronary atherosclerosis: the Heinz Nixdorf Recall study. J Am Coll Cardiol 2010;56:1397 1406. 4. Greenland P, Alpert JS, Beller GA, Benjamin EJ, Budoff MJ, Fayad ZA, Foster E, Hlatky MA, Hodgson JM, Kushner FG, Lauer MS, Shaw LJ, Smith SC, Jr., Taylor AJ, Weintraub WS, Wenger NK, Jacobs AK; American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. 2010 ACCF/AHA guideline for assessment of cardiovascular risk in asymptomatic adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation 2010; 122(suppl):e584 e636. 5. Zethelius B, Berglund L, Sundström J, Ingelsson E, Basu S, Larsson A, Venge P, Arnlöv J. Use of multiple biomarkers to improve the prediction of death from cardiovascular causes. N Engl J Med 2008;358: 2107 2116. 6. Rana JS, Cote M, Després JP, Sandhu MS, Talmud PJ, Ninio E, Wareham NJ, Kastelein JJ, Zwinderman AH, Khaw KT, Boekholdt SM. Inflammatory biomarkers and the prediction of coronary events among people at intermediate risk: the EPIC-Norfolk prospective population study. Heart 2009;95:1682 1687. 7. Wang TJ, Gona P, Larson MG, Tofler GH, Levy D, Newton-Cheh C, Jacques PF, Rifai N, Selhub J, Robins SJ, Benjamin EJ, D Agostino RB, Vasan RS. Multiple biomarkers for the prediction of first major cardiovascular events and death. N Engl J Med 2006;355:2631 2639. 8. Rozanski A, Gransar H, Shaw LJ, Kim J, Miranda-Peats L, Wong ND, Rana JS, Orakzai R, Hayes SW, Friedman JD, Thomson LE, Polk D, Min J, Budoff MJ, Berman DS. Impact of coronary artery calcium scanning on coronary risk factors and downstream testing the EISNER (Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research) prospective randomized trial. J Am Coll Cardiol 2011;57:1622 3162. 9. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M Jr, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990;15:827 832. 10. Palazzuoli A, Maisel A, Caputo M, Fineschi M, Quatrini I, Calabrò A, Campagna MS, Franci B, Grothgar S, Pierli C, Nuti R. B-type natriuretic peptide levels predict extent and severity of coronary disease in non-st elevation coronary syndromes and normal left ventricular systolic function. Regul Pept 2011;167:129 133. 11. D Agostino RB, Sr., Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, Kannel WB. General cardiovascular risk profile for use in primary care: the Framingham heart study. Circulation 2008;117: 743 753. 12. Pencina MJ, D Agostino RB Sr, D Agostino RB Jr, Vasan RS. Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond. Stat Med 2008;27:157 172. 13. Pencina MJ, D Agostino RB Sr, Steyerberg EW. Extensions of net reclassification improvement calculations to measure usefulness of new biomarkers. Stat Med 2011;30:11 21. 14. Arad Y, Goodman KJ, Roth M, Newstein D, Guerci AD. Coronary calcification, coronary disease risk factors, C-reactive protein, and atherosclerotic cardiovascular disease events: the St. Francis Heart Study. J Am Coll Cardiol 2005;46:158 165. 15. Möhlenkamp S, Lehmann N, Moebus S, Schmermund A, Dragano N, Stang A, Siegrist J, Mann K, Jöckel KH, Erbel R; Heinz Nixdorf Recall Study Investigators. Quantification of coronary atherosclerosis and inflammation to predict coronary events and all-cause mortality. J Am Coll Cardiol 2011;57:1455 1464. 16. Elias-Smale SE, Proença RV, Koller MT, Kavousi M, van Rooij FJ, Hunink MG, Steyerberg EW, Hofman A, Oudkerk M, Witteman JC. Coronary calcium score improves classification of coronary heart disease risk in the elderly: the Rotterdam study. J Am Coll Cardiol 2010;56:1407 1414. 17. Shah PK. Screening asymptomatic subjects for subclinical atherosclerosis: can we, does it matter, and should we? J Am Coll Cardiol 2010;56:98 105. 18. Blaha MJ, Budoff MJ, DeFilippis AP, Blankstein R, Rivera JJ, Agatston A, O Leary DH, Lima J, Blumenthal RS, Nasir K. Associations between C-reactive protein, coronary artery calcium, and cardiovascular events: implications for the JUPITER population from MESA, a population-based cohort study. Lancet 2011;378:684 692. 19. Shaw LJ, Min JK, Budoff M, Gransar H, Rozanski A, Hayes SW, Friedman JD, Miranda R, Wong ND, Berman DS. Induced cardiovascular procedural costs and resource consumption patterns after coronary artery calcium screening: results from the EISNER (Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research) study. J Am Coll Cardiol 2009;54:1258 1267. 20. Wilson PW. Challenges to improve coronary heart disease risk assessment. JAMA 2009;302:2369 2370. 21. Wilkins JT, Lloyd-Jones DM. Biomarkers for coronary heart disease clinical risk prediction: a critical appraisal. Counterpoint. Prev Cardiol 2010;13:160 165.