J A C C : C A R D I O V A S C U L A R I M A G I N G V O L . 9 , N O . 2 , 2 0 1 6

ª 2 0 1 6 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N I S S N 1 9 3 6 - 8 7 8 X / $ 3 6 . 0 0

P U B L I S H E D B Y E L S E V I E R h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j c m g . 2 0 1 5 . 0 6 . 0 3 0

Relationship of Coronary Calcium onStandard Chest CT Scans With Mortality

Jan M. Hughes-Austin, PT, MPT, PHD,a,b Arturo Dominguez III, MD,c Matthew A. Allison, MD, MPH,b,c,d

Christina L. Wassel, PHD,e Dena E. Rifkin, MD, MS,b,c,d Cindy G. Morgan, MS,b Michael R. Daniels, BS,b

Umaira Ikram, MD,b Jessica B. Knox, MD,b C. Michael Wright, MD,f Michael H. Criqui, MD, MPH,b

Joachim H. Ix, MD, MASb,c,d

ABSTRACT

Fro

Fa

Sch

Ca

an

Lu

He

rel

Ma

OBJECTIVES The aim of this study was to determine the correlation between coronary artery calcium (CAC) scores on

3 mm electrocardiography (ECG)-gated computed tomography (CT) scans and standard 6 mm chest CT scans, and to

compare relative strength of associations of CAC on each scan type with mortality risk.

BACKGROUND Coronary artery calcification predicts cardiovascular disease (CVD) and all-cause mortality, and

is typically measured on ECG-gated 3 mm CT scans. Patients undergo standard 6 mm chest CTs for various clinical

indications much more frequently, but CAC is not usually quantified. To better understand the usefulness of

standard chest CTs to quantify CAC, we conducted a case-control study among persons who had both scan types.

METHODS Between 2000 and 2003, 4,544 community-living individuals self- or physician-referred for “whole-body”

CT scans, had 3 mm ECG-gated CTs and standard 6 mm chest CTs, and were followed for mortality through 2009. In this

nested case-control study, we identified 157 deaths and 494 controls frequency matched (1:3) on age and sex. The

Agatston method quantified CAC on both scan types. Unconditional logistic regression determined associations with

mortality, accounting for CVD risk factors.

RESULTS Participants were 68 � 11 years of age and 63% male. The Spearman correlation of CAC scores between the

2 scan types was 0.93 (p < 0.001); median CAC scores were lower on 6 mm CTs compared to 3 mm CTs (22 vs.104

Agatston units, p < 0.001). Adjusted for traditional CVD risk factors, each standard deviation higher CAC score on 6 mm

CTs was associated with 50% higher odds of death (odds ratio: 1.5; 95% confidence interval: 1.2 to 1.9), similar to 50%

higher odds on the 3 mm ECG-gated CTs (odds ratio: 1.5; 95% confidence interval: 1.1 to 1.9).

CONCLUSIONS CAC scores on standard 6 mm chest CTs are strongly correlated with 3 mm ECG-gated CTs and similarly

predict mortality in community-living individuals. Chest CTs performed for other clinical indications may provide

an untapped resource to garner CVD risk information without additional radiation exposure or expense.

(J Am Coll Cardiol Img 2016;9:152–9) © 2016 by the American College of Cardiology Foundation.

C oronary artery disease (CAD) is the leadingcause of mortality in the United States (1).Detection of coronary artery calcification

(CAC) is a strong predictor of CAD, cardiovascularevents, and all-cause mortality (2,3), above andbeyond the Framingham risk score (2,4). CAC is usu-ally quantified on dedicated 3 mm sliced computed

m the aDepartment of Orthopaedic Surgery, School of Medicine, Universit

mily Medicine and Public Health, School of Medicine, University of Califor

ool of Medicine, University of California, San Diego, California; dVetera

lifornia; eDepartment of Epidemiology, Graduate School of Public Health

d fScripps Health, La Jolla, California. Funding for this project comes fro

ng, and Blood Institute R01HL116395, American Heart Association Estab

art Association Fellow to Faculty Award 0475029N, and K23DK09152

ationships relevant to the contents of this paper to disclose.

nuscript received April 16, 2015; revised manuscript received June 15, 20

tomography (CT) scans that are electrocardiography(ECG) gated, so as to minimize motion artifact fromthe beating heart and provide relatively fine cutsthrough the coronary arteries. These scans are donefrequently in research settings, but uncommonly inclinical practice because most insurance providers donot cover the cost of the scan for preventive medicine

y of California, San Diego, California; bDepartment of

nia, San Diego, California; cDepartment of Medicine,

ns Affairs San Diego Healthcare System, San Diego,

, University of Pittsburgh, Pittsburgh, Pennsylvania;

m the National Institutes of Health, National Heart,

lished Investigator Award EIA18560026, American

1. All authors have reported that they have no

15, accepted June 18, 2015.

AB BR E V I A T I O N S

AND ACRONYM S

BMI = body mass index

CAC = coronary artery calcium

CAD = coronary artery disease

CT = computed tomography

CVD = cardiovascular disease

ECG = electrocardiogram /

electrocardiography

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 9 , N O . 2 , 2 0 1 6 Hughes-Austin et al.F E B R U A R Y 2 0 1 6 : 1 5 2 – 9 Coronary Artery Calcium on Chest CT Scans and Mortality

153

and because the U.S. Preventive Task Force currentlydoes not recommend preventive CAC screening in in-dividuals without a history of CAD (5,6).

Standard chest CT scans are used for numerousclinical indications including lung cancer screenings,evaluation for pulmonary embolism, adenopathy,pleural diseases, and pneumonia, among others. Cal-cium within the coronary arteries can be easily recog-nized on these scans (7), and prior studies haveevaluated CAC on lung CTs for CAD screening insmokers at high risk for lung cancer (8,9). However,CAC screening may be most useful in persons at in-termediate risk for CAD (2), where presence andseverity of CAC may modify the approach to preven-tive strategies such as use of statins and other in-terventions. In comparison with the approximate600,000 3 mm ECG-gated CT scans done in the UnitedStates annually, it is estimated that over 7.1 million6mm lung CT scans are done annually for other clinicalindications (10). While several studies have demon-strated agreement between 3 mm ECG-gated CTs andstandard chest CTs in their measurement of CAC(7,9,11), whether standard chest CTs can predict out-comes in the general population, and whether resultsare similar to 3mmelectron beamCT scan data, despitethe wider cuts and absence of ECG gating, is unknown.

SEE PAGE 160

To better understand the usefulness of standardchest CTs for this purpose,we conducted a case-controlstudy among persons who had both scan types whenthey were seen in 2000 to 2003 and who were followedfor mortality for approximately 8 years thereafter.

METHODS

STUDY POPULATION AND STUDY DESIGN. Studyparticipants were recruited through a San Diego car-diovascular imaging clinic between 2000 and 2003,where 4,544 community-living individuals who weremostly asymptomatic were self-referred or referred byprimary care physicians for “whole-body” CT scans.Participants were followed for mortality through2009, during which 173 participants died. Using anested case-control study design, each death (case)was frequency matched on sex and age within 1 yearwith 3 surviving controls, resulting in 518 controls, andtotaling 691 participants (12). Following the selectionof cases and matching controls, we further excludedany participant who had undergone any angioplasty,stent, or bypass revascularization procedure (n ¼ 41),which resulted in a total of 651 participants—157 deathsand 494 controls. Five of the oldest cases did not haveage matches within 1 year. Thus, controls for these

5 cases were within 3 to 5 years of the ages ofthese 5 cases. A 3 mm ECG-gated CT was ob-tained and scored for CAC at the time of theinitial visit. A 6mm standard chest CTwas alsoobtained and read for general lung pathology,but was not scored for CAC at the time of thebaseline visit. Our case-control design, nestedwithin a prospective cohort study, allowed usa targeted approach to retrieve chest CTs andreread them for CACwithout significant loss of

statistical power. All participants provided informedconsent and the study was approved by the Universityof California, San Diego, Human Research ProtectionsProgram.

CORONARY ARTERY CALCIFICATION IMAGING AND

SCORING. CT scans were performed using an ImatronC-150 scanner (San Francisco, California), which is anelectron beam CT scanner with a high-resolution de-tector system. We used the standard single-sectionmode, which involves an image acquisition time of100 ms and 3 mm section thickness. The 3 mm ECG-gated CT scans were electrocardiographically trig-gered at 40% or 65% of the R-R interval, depending onthe participant’s heart rate, and resulted in 1.0 and 1.3mSv of radiation forwomen andmen, respectively (13).For the 6 mm chest CT scans, subjects were scannedfrom the sternal notch to base of the diaphragmwithout ECG gating. Radiation exposure for this typeof scan, as presented in lung cancer screening litera-ture for low-dose CT scan screening is approximately1.5 mSv (14). Methods for scoring CAC follow thosedescribed by Agatston et al. (15) on both scan types.The 3 mm ECG-gated scans were read at the time ofscan acquisition, while the 6 mm chest CT scans wereread for CAC in 2012. Readers were blinded to partici-pant clinical data and to their 3 mm CT CAC score.

MORTALITY. In 2009, we linked the data with theSocial Security Death Index to identify individualswho had died in the intervening period. Potentialdeaths identified by the Social Security Death Indexwere cross referenced with their patient clinical re-cords to confirm identity.

COVARIATES. Height and weight were measured atclinic visits, and bodymass index (BMI) was calculated(in kg/m2). Age and sex were obtained through self-report, and a questionnaire was used to obtain aparticipant’s medical history including smokingstatus (never, former, or current), prevalent hyper-tension, and cholesterol medication use. Nonfastingserum lipid and glucose levels were obtainedvia finger-stick using the Cholestech LDX system(Hayward, California). Diabetes was defined as serumglucose >200 mg/dl, or the use of glucose-lowering

TABLE 1 Participant Characteristics by Case-Control Status

Cases(n ¼ 157)

Controls(n ¼ 494) p Value

Age, yrs 67.7 � 11.7 67.8 � 10.7 0.92

Body mass index, kg/m2 26.6 � 4.6 26.8 � 4.2 0.70

Total cholesterol 203.5 � 42.5 201.9 � 40.0 0.67

HDL cholesterol 52.4 � 17.9 51.7 � 16.9 0.70

Female 67 (43) 186 (38) 0.26

Diabetes 13 (8) 16 (3) 0.01

Hypertension 64 (41) 188 (38) 0.53

Lipid medication use 9 (6) 38 (8) 0.41

Smoking status

Never smoked 70 (45) 269 (54) 0.03

Former smoker 67 (43) 195 (39) 0.47

Current smoker 20 (13) 32 (6) 0.01

CVD family history 43 (27) 87 (18) 0.01

3 mm EBCT CAC score* 210 (14, 608) 84 (0, 484) 0.01

6 mm chest CT CAC score* 55 (0, 230) 15 (0, 146) <0.01

Values are mean � SD, n (%), or median (25th, 75th). *Used Wilcoxon rank sum test to determinedifferences between cases and controls.

CAC ¼ coronary artery calcium; CT ¼ computed tomography; CVD ¼ cardiovascular disease;EBCT ¼ electron beam computed tomography; HDL ¼ high-density lipoprotein.

FIGURE 1 Scatter

and 6 mm Chest CT

Spearman correlation

Controls (n ¼ 494).

ECG ¼ electrocardio

Hughes-Austin et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 9 , N O . 2 , 2 0 1 6

Coronary Artery Calcium on Chest CT Scans and Mortality F E B R U A R Y 2 0 1 6 : 1 5 2 – 9

154

medications. Dyslipidemia was defined as total tohigh-density lipoprotein cholesterol ratio >5, or theuse of cholesterol-lowering medication. Systolic anddiastolic blood pressures were obtained from a trainedtechnician after the participant rested for 5 min.Hypertension was defined as systolic blood pressure>140 mm Hg, diastolic blood pressure >90 mm Hg, orthe use of antihypertensive medications.

Plot of Agatston CAC Scores on 3 mm Cardiac ECG-Gated CT Scans

scans

r ¼ 0.93, p < 0.001. (Blue circles) Cases (n ¼ 157). (Pink circles)

CAC ¼ coronary artery calcium; CT ¼ computed tomography;

graphy.

STATISTICAL ANALYSIS. We evaluated differencesin demographics and traditional cardiovascular dis-ease (CVD) risk factors in cases and controls usingStudent t tests or Wilcoxon rank sum tests forcontinuous variables and chi-square tests or Fisherexact tests for categorical variables. We evaluated thecorrelation of CAC scores on the 2 scan types usingSpearman correlations for continuous CAC scores,given skewed distributions, and Kappa statistics forcategorical CAC scores. We also performed a Bland-Altman analysis to determine the bias and limits ofagreement between the 3 mm ECG-gated CT and the6 mm standard chest CT scans (16). Next, we usedunconditional logistic regression to examine associ-ations of each scan type with mortality. Initial modelswere unadjusted. Subsequent models adjusted fordemographics and traditional CVD risk factors(age, sex, diabetes, hypertension, dyslipidemia, BMI,smoking, and family history of CVD). We evaluatedeach CAC score as both a continuous and categoricalvariable. For analysis using a continuous CAC score,we added 1 to each CAC score (such that zero scoreswere not excluded from the analysis) and naturallog transformed the distribution (Ln[CAC þ1]) toapproximate a normal distribution of CAC scores. Weused standardized coefficients to compare strength ofassociation per standard deviation of LnCACþ1 acrossthe 2 scan types. For analysis using a categorical CACscore, we categorized patients into 4 groups accord-ing to standard 3 mm Agatston CAC score cutpoints,which facilitated comparisons across scan types (17).

In evaluating the 6 mm chest CT scan Agatston CACscores, 34% had Agatston scores of 0, whereas 24% hadscores of 0 using the 3 mm ECG-gated CT scans. Usingstandard cutpoints proposed by Detrano et al. (17), wedefined 4 categories of CAC for both 3 mm ECG-gatedand 6 mm chest CT scans: CAC ¼ 0, CAC ¼ 1 to 100,CAC ¼ 101 to 300, and CAC >300. All analyses wereconducted using Stata version 11.0SE (Stata Corp.,College Station, Texas) and SAS version 9.4 (SASInstitute, Cary, North Carolina), and p values <0.05were considered statistically significant.

RESULTS

The 651 participants in this study had a mean ageof 68 years, and 63% were male. Cases and con-trols had similar BMI, total cholesterol, high-densitylipoprotein cholesterol, and use of lipid-loweringmedications. However, a greater proportion of caseshad diabetes and hypertension, were former or cur-rent smokers, and had a family history of CVD. Casesalso had higher median CAC scores on both the 3 and6 mm CT scans compared to controls at baseline

FIGURE 2 Bland-Altman Plot for Agreement Between 3 mm ECG-Gated CT Scans and

6 mm Standard Chest CT Scans Using Log Transformed Agatston CAC Scores

*Log transformed mean bias (center dashed line) was 1.17 (�0.855), which

back-transforms to 3.23, with limits of agreement ranging from 0.6 to 17.3

(top and bottom dashed lines). Abbreviations as in Figure 1.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 9 , N O . 2 , 2 0 1 6 Hughes-Austin et al.F E B R U A R Y 2 0 1 6 : 1 5 2 – 9 Coronary Artery Calcium on Chest CT Scans and Mortality

155

(Table 1). Race/ethnicity was not required to obtain aCT scan, thus it was provided by 199 individuals inthis study. Of these 199 individuals with reportedethnicity, 100% (43) of deaths occurred in Caucasians.This subsample of 199 included 0.6% Asian, 12%Hispanic, and 2% other—the remaining 85% wereCaucasian.

The correlation between CAC scores on the 3 mmECG-gated CT and the 6 mm standard chest CT was0.93 (p < 0.001) (Figure 1). However, the median CACscores were significantly lower on the 6 mm CT scanthan on the 3 mm ECG-gated scan (22 vs. 104 Agatstonunits, respectively, p < 0.001). The Kappa statistic foragreement between CAC score categories on the 3 mmECG-gated CT compared to the 6 mm standard chestCT was 0.41, and the weighted Kappa statistic was0.62, indicating moderate to substantial agreementbetween the 2 scans using standard clinical cutpoints(CAC ¼ 0, 1 to 100, 101 to 300, and >300). Bland-Altman analysis showed a mean bias of 3.23, withlimits of agreement between 0.6 and 17.3, as shown inFigure 2, suggesting that Agatston scores on 3 mmECG-gated CT scans were, on average, approximately323% higher than Agatston scores on the 6 mm stan-dard chest CT scan (16). As illustrative examples of thedifference in sensitivity, Figure 3 presents CACmeasured on 3 mm ECG-gated CT (CAC ¼ 372) and 6mm chest CT (CAC ¼ 117) at approximately the samelocation within an individual. In a participant withmore advanced CAC, Figure 4 presents 2 consecutiveslices within the same individual where CACmeasured on 3 mm ECG-gated CT totaled to 3,213 andCAC measured on 6 mm chest CT totaled to 1,044.These examples demonstrate that, while CAC is

FIGURE 3 Illustrative Example of the Difference in Sensitivity BetwSlice Within an Individual

(A) 3 mm ECG-gated CT (Agatston CAC score ¼ 372); (B) 6 mm chest C

present and quantifiable on both the 3 and 6 mmscans, the clarity and amount of CAC is higher onaverage using a 3 mm ECG-gated CT scan.

Table 2 shows the associations of CAC scores onthe 2 scan types with mortality. When evaluatingCAC on the 6 mm chest CT scans, compared to theCAC ¼ 0 reference category, there was a graded

een 3 mm ECG-Gated CT Scans and 6 mm Chest CT Scans in a Single

T (Agatston CAC score ¼ 117). Abbreviations as in Figure 1.

FIGURE 4 Illustrative Example of the Difference in Sensitivity Between 3 mm ECG-Gated CT Scans and 6 mm Chest CT Scans in

2 Consecutive Slices Within an Individual

(A1, A2) Two consecutive slices from 3 mm ECG-gated CT within an individual (Agatston CAC score ¼ 3,213). (B1, B2) Two consecutive slices

from 6 mm chest CT within the same individual (Agatston CAC score ¼ 1,044). Abbreviations as in Figure 1.

Hughes-Austin et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 9 , N O . 2 , 2 0 1 6

Coronary Artery Calcium on Chest CT Scans and Mortality F E B R U A R Y 2 0 1 6 : 1 5 2 – 9

156

relationship of higher CAC score with higher odds ofmortality such that those with CAC scores between1 and 100 had 1.9-fold higher odds of mortality, be-tween 101 and 300 had 2.3-fold higher odds of mor-tality, and those with CAC greater than 300 had2.6-fold higher odds for mortality in models adjustedfor demographics and traditional CVD risk factors. Incomparison, with the 3 mm ECG-gated scans,compared to the CAC ¼ 0 reference category, par-ticipants with CAC scores between 1 and 100 had2.1-fold higher odds of mortality, between 101 and300 had 2.9-fold higher odds of mortality, and thosegreater than 300 had 3.2-fold higher odds of mor-tality in adjusted models. These associations havebeen plotted in Figure 5. When CAC scores wereevaluated as a continuous variable, each SD higherCAC score on the 6 mm chest CTs was associatedwith 1.5-fold higher odds of mortality. Similarly, eachSD higher score on the 3 mm ECG-gated scans wasassociated with 1.5-fold higher odds of mortality. Allof these results were statistically significant.

DISCUSSION

In this population of community-living individualswho self-referred or were referred by their primarycare physicians for “whole-body” CT scans, CACscores on standard 6 mm chest CTs were stronglycorrelated to those on the 3 mm ECG-gated CT scansspecifically designed for CAC measurement. Scores oneither scan type were strongly associated with all-cause mortality, and the relative strength of associa-tion was similar irrespective of the scan type.

CAC screening has been considered useful forprevention of CVD because it can further riskstratify persons who are considered intermediaterisk for incident CVD events by the Framinghamrisk score. Several investigators have reported thatstandard chest CT scans used for lung cancerscreening can also detect and quantify CAC(3,4,9,11,18,19). Our study confirms these results,and extends them to a community-living populationfor the first time. Moreover, to our knowledge, this

TABLE 2 Association of 6 mm Standard Chest CT Versus 3 mm ECG-Gated EBCT CAC Scores With All-Cause Mortality

6 mm nongated chest CT

Agatston CAC score range 0 1–100 101–300 >300 Continuous modelper SD Ln [CAC þ1]

Deaths/controls 41/192 53/152 29/70 34/80

Unadjusted OR (95% CI) 1.00 (Ref) 1.6 (1.0–2.6) 1.9 (1.1–3.4) 2.0 (1.2–3.4) 1.3 (1.1–1.6)

p value 0.037 0.018 0.01 0.002

Adjusted* OR (95% CI) 1.00 (Ref) 1.9 (1.1–3.1) 2.3 (1.2–4.3) 2.6 (1.4–4.9) 1.5 (1.2–1.9)

p value 0.017 0.009 0.004 0.001

3 mm ECG-gated EBCT

Agatston CAC score range 0 1–100 101–300 >300 Continuous modelper SD Ln [CAC þ1]

Deaths/controls 26/139 35/120 31/77 65/158

Unadjusted OR (95% CI) 1.00 (Ref) 1.6 (0.9–2.7) 2.2 (1.2–3.9) 2.2 (1.3–3.7) 1.3 (1.1–1.6)

p value 0.122 0.011 0.002 0.005

Adjusted* OR (95% CI) 1.00 (Ref) 2.1 (1.1–3.8) 2.9 (1.5–5.7) 3.2 (1.7–6.0) 1.5 (1.1–1.9)

p value 0.022 0.002 0.000 0.002

*Adjusted for age, sex, diabetes, hypertension, hyperlipidemia, body mass index, smoking, and family history of CVD.

CI ¼ confidence interval; ECG ¼ electrocardiography; OR ¼ odds ratio; other abbreviations as in Table 1.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 9 , N O . 2 , 2 0 1 6 Hughes-Austin et al.F E B R U A R Y 2 0 1 6 : 1 5 2 – 9 Coronary Artery Calcium on Chest CT Scans and Mortality

157

is the first study to compare the correlation andrelative strengths of association of CAC on thestandard chest CT and the 3 mm ECG-gated CT withmortality.

Our results may have important implications forpreventive cardiology clinical practice. In 2007,approximately 7.1 million chest CT scans were per-formed in the United States, compared to 600,000 CTscans for calcium scoring, presumably using 3 mmECG-gated CTs (10). There is concern that radiationexposure from CT scans may increase cancer risk

FIGURE 5 Odds Ratios and 95% Confidence Intervals for the Associ

Chest CT Scans and Mortality

*Adjusted for age, sex, diabetes, hypertension, hyperlipidemia, body ma

Abbreviations as in Figure 1.

(20,21). In addition, many health insurance plans donot cover the expense of CT scanning for CAC mea-surement for preventive care purposes. Given thatchest CTs are done frequently for numerous otherclinical indications, many individuals may alreadyhave scans that can be scored for CAC,whichmay guidepreventive cardiology care. Thus, if our results areconfirmed, health care providers may consider utiliz-ing previously obtained CT scans to assess CAC whileavoiding the potential risks and expense of repeat CTscans designed specifically for CAC measurement.

ations Between 3 mm ECG-Gated CT Scans and Mortality and 6 mm

ss index, smoking, and family history of cardiovascular disease.

PERSPECTIVES

COMPETENCY IN MEDICAL KNOWLEDGE:

Standard chest CT scans are often not read for CAC

severity by radiologists when reading scans obtained

for other clinical indications. We show that detection

of CAC on standard chest CT scans is closely corre-

lated with CAC on the ECG-gated CT scans, and

similarly predicts mortality. Therefore, chest CT scans

obtained for various clinical indications can also

inform CVD risk through measurement of CAC.

TRANSLATIONAL OUTLOOK: While similarly pre-

dictive of events, a CAC score on a standard chest CT

was lower than on an ECG-gated CT on the same in-

dividual. Future studies are required to determine

applicable cutpoints for CAC scores on standard CT

scans relative to the ECG-gated CT.

Hughes-Austin et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 9 , N O . 2 , 2 0 1 6

Coronary Artery Calcium on Chest CT Scans and Mortality F E B R U A R Y 2 0 1 6 : 1 5 2 – 9

158

Currently, CAC observed on standard chest CTscans is not routinely reported by radiologists, anduse of the Agatston method is even less common. Forexample, Williams et al. (22) reported that CAC onchest CTs was recorded in the final radiology report inonly 44% of patients with known CAC. Jacobs et al.(23) demonstrated that, by using simple visualgrading to measure CAC on standard chest CT, CACwas strongly associated with future CVD events.Thus, by simple visual grading system evaluated byothers (23,24), or by the Agatston method used here,the detection of CAC on standard chest CT has prog-nostic implications. As it has important implicationsfor preventive care, we believe it should be system-atically reported.

The median CAC scores on the 6 mm chest CT scanswere substantially lower on average than on the 3 mmECG-gated CT scans. This may be because there arefewer slices to evaluate and score on the 6 mm scan.With fewer slices, there may also be volume aver-aging in plaque based on interpolation algorithmsfrom slice to slice. Also, small plaques may be missedbetween 6 mm cuts. Therefore, 6 mm chest CT scansmay be less sensitive for low levels of CAC, and thegood prognosis provided by a 3 mm ECG-gated scanshowing zero CAC may not be as robust for a 6 mmchest CT with zero CAC. Absence of any CAC on anECG-gated 3 mm scan has been associated with lowrisk for incident CVD, and relatively standard CACcutpoints (e.g., 0 to 100, 101 to 300, >300) have beenused frequently in prior studies. In this study, wehave extended these cutpoints to 6 mm standardchest CT scans, which were similarly associated withmortality risk, but the implication of a specific CACscore may differ across scan types. For example themedian 6 mm standard chest CT scan CAC scores were3, 46, and 286 in individuals in the 3 mm CT scan CACcategories of 1 to 100, 101 to 300, and >300, respec-tively. Establishing applicable cutpoints for AgatstonCAC scores on 6 mm standard chest CT scans warrantfurther development. However, our data suggest thatCAC scores on 6 mm chest CT scans may underesti-mate the CAC burden compared to 3 mm ECG-gatedCT scores.STUDY LIMITATIONS. Our study population wasmostly older non-Hispanic white adults, many ofwhom self-referred for “whole-body” CT scans forpreventive care. Such individuals may be particularlymotivated to prevent chronic diseases. Future studiesin other settings are required to confirm thesefindings.As discussed previously, although the associations ofeither scan type with mortality were similar, there areabsolute differences in the CAC scores across scantypes, whichmay impact sensitivity to detect low level

CAC using 6 mm chest CT scans and will influencespecific cutpoints. Thus, specific CAC values should beinterpreted with knowledge about the scan type.

CONCLUSIONS

Despite the absence of ECG gating and wider slicethickness, CAC scores on standard 6 mm chest CTs arehighly correlated with those on 3 mm ECG-gated CTscans, and are similarly associated with mortality riskin community-living individuals. Most insuranceproviders do not routinely cover the expense of 3 mmECG-gated scans for CAC scoring, and converselyapproximately 7 million chest CT scans are doneannually in the United States for other clinical in-dications (10). Persons who have chest CTs for otherclinical indications may benefit from systematicreading of CAC to garner additional information onCVD risk without the added expense and radiationexposure required for dedicated 3 mm ECG-gatedscans.

ACKNOWLEDGMENTS The authors thank the partic-ipants of this study, the clinic staff who received andexamined these participants, and Julie Denenberg forher assistance and organization.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Joachim H. Ix, Division of Nephrology and Hyper-tension, Department of Medicine, University of Cali-fornia San Diego, and San Diego Veterans AffairsHealthcare System, 3350 La Jolla Village Drive, Mailcode 111-H. San Diego, California 92161. E-mail: joix@ucsd.edu.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 9 , N O . 2 , 2 0 1 6 Hughes-Austin et al.F E B R U A R Y 2 0 1 6 : 1 5 2 – 9 Coronary Artery Calcium on Chest CT Scans and Mortality

159

RE F E RENCE S

1. Thomas KL, Honeycutt E, Shaw LK, Peterson ED.Racial differences in long-term survival amongpatients with coronary artery disease. Am Heart J2010;160:744–51.

2. Polonsky TS, McClelland RL, Jorgensen NW,et al. Coronary artery calcium score and risk clas-sification for coronary heart disease prediction.JAMA 2010;303:1610–6.

3. Shemesh J, Henschke CI, Shaham D, et al.Ordinal scoring of coronary artery calcifications onlow-dose CT scans of the chest is predictive ofdeath from cardiovascular disease. Radiology2010;257:541–8.

4. Shaw LJ, Raggi P, Callister TQ, Berman DS.Prognostic value of coronary artery calciumscreening in asymptomatic smokers and non-smokers. Eur Heart J 2006;27:968–75.

5. O’Malley PG, Greenberg BA, Taylor AJ. Cost-effectiveness of using electron beam computedtomography to identify patients at risk for clinicalcoronary artery disease. Am Heart J 2004;148:106–13.

6. U.S. Preventive Services Task Force. FinalRecommendation Statement: Coronary HeartDisease: Screening Using Non-Traditional RiskFactors. 2009. Available at: http://www.uspreventiveservicestaskforce.org/Page/Document/RecommendationStatementFinal/coronary-heart-disease-screening-using-non-traditional-risk-factors.Accessed October 23, 2014.

7. Kirsch J, Buitrago I, Mohammed TL, Gao T,Asher CR, Novaro GM. Detection of coronary cal-cium during standard chest computed tomographycorrelates with multi-detector computed tomog-raphy coronary artery calcium score. Int J Car-diovasc Imaging 2012;28:1249–56.

8. Jacobs PC, Gondrie MJ, van der Graaf Y, et al.Coronary artery calcium can predict all-cause

mortality and cardiovascular events on low-doseCT screening for lung cancer. AJR Am J Roent-genol 2012;198:505–11.

9. Takx RA, de Jong PA, Leiner T, et al. Automatedcoronary artery calcification scoring in non-gatedchest CT: agreement and reliability. PLoS One2014;9:e91239.

10. Berrington de Gonzalez A, Mahesh M, Kim KP,et al. Projected cancer risks from computedtomographic scans performed in the United Statesin 2007. Arch Intern Med 2009;169:2071–7.

11. Kim YK, Sung YM, Cho SH, Park YN, Choi HY.Reliability analysis of visual ranking of coronaryartery calcification on low-dose CT of the thoraxfor lung cancer screening: comparison with ECG-gated calcium scoring CT. Int J Cardiovasc Imag-ing 2014;30:81–7.

12. Rothman K, Greenland S. Matching. ModernEpidemiology. 2nd edition. Philadelphia, PA: Lip-pincott-Raven, 1998.

13. Hunold P, Vogt FM, Schmermund A, et al.Radiation exposure during cardiac CT: effectivedoses at multi-detector row CT and electron-beamCT. Radiology 2003;226:145–52.

14. Aberle DR, Berg CD, Black WC, et al. The Na-tional Lung Screening Trial: overview and studydesign. Radiology 2011;258:243–53.

15. Agatston AS, Janowitz WR, Hildner FJ,Zusmer NR, Viamonte M Jr., Detrano R. Quantifi-cation of coronary artery calcium using ultrafastcomputed tomography. J Am Coll Cardiol 1990;15:827–32.

16. Bland JM, Altman DG. Measuring agreement inmethod comparison studies. Stat Methods MedRes 1999;8:135–60.

17. Detrano R, Guerci AD, Carr JJ, et al. Coronarycalcium as a predictor of coronary events in four

racial or ethnic groups. N Engl J Med 2008;358:1336–45.

18. Wu MT, Yang P, Huang YL, et al. Coronaryarterial calcification on low-dose ungated MDCTfor lung cancer screening: concordance study withdedicated cardiac CT. AJR Am J Roentgenol 2008;190:923–8.

19. Arcadi T, Maffei E, Sverzellati N, et al. Coro-nary artery calcium score on low-dose computedtomography for lung cancer screening. World JRadiol 2014;6:381–7.

20. Brenner DJ. Radiation risks potentially asso-ciated with low-dose CT screening of adultsmokers for lung cancer. Radiology 2004;231:440–5.

21. Linet MS, Slovis TL, Miller DL, et al. Cancerrisks associated with external radiation fromdiagnostic imaging procedures. CA Cancer J Clin2012;62:75–100.

22. Williams KA Sr., Kim JT, Holohan KM. Fre-quency of unrecognized, unreported, or under-reported coronary artery and cardiovascularcalcification on noncardiac chest CT. J CardiovascComput Tomogr 2013;7:167–72.

23. Jacobs PC, Gondrie MJ, Mali WP, et al. Unre-quested information from routine diagnostic chestCT predicts future cardiovascular events. EurRadiol 2011;21:1577–85.

24. Blair KJ, Allison MA, Morgan C, et al. Com-parison of ordinal versus Agatston coronarycalcification scoring for cardiovascular diseasemortality in community-living individuals. Int JCardiovasc Imaging 2014;30:813–8.

KEY WORDS chest computed tomography,coronary artery calcium, epidemiology,mortality