Coronary Calcium and the Prediction of Coronary Heart Disease

Coronary Calcium and the Coronary Calcium and the Prediction of Coronary Heart Prediction of Coronary Heart

DiseaseDisease

Kiang Liu

OverviewOverview

• Why Should We Study Why Should We Study Subclinical Cardiovascular Disease (CVD)??

• Coronary Calcium and the Prediction of Coronary Coronary Calcium and the Prediction of Coronary EventsEvents

• MESA Findings on Ethnic Differences of Coronary MESA Findings on Ethnic Differences of Coronary Calcium and on Coronary Calcium and the Prediction Calcium and on Coronary Calcium and the Prediction of Coronary Eventsof Coronary Events

• Improvement of Using CAC to Identify People at High Improvement of Using CAC to Identify People at High Risk for Coronary EventsRisk for Coronary Events

• ConclusionsConclusions

Why Should We Study Why Should We Study Subclinical CVD? Subclinical CVD?

From NHLBI Chartbook, 2009From NHLBI Chartbook, 2009

Crude Death Rates for Selected Crude Death Rates for Selected Causes, U.S., 1950-2006Causes, U.S., 1950-2006

Cause of Death Number

Total 2,426,264

1 Heart Disease * 631,636

2 Cancer 559,888

3 Cerebrovascular Disease (stroke) 137,119

4 COPD and allied conditions ** 124,583

5 Accidents 121,599

6 Diabetes 72,449

7 Alzheimer’s disease 72,432

8 Influenza and pneumonia 56,326

9 Nephritis 45,344

10 Septicemia 34,234

All other causes of death 570,654

* Includes 425,425 deaths from coronary heart disease

** Chronic lower respiratory diseases.

From NHLBI Chartbook, 2009From NHLBI Chartbook, 2009

Leading Causes of DeathLeading Causes of DeathU.S., 2006U.S., 2006

No CHD

CHD

Total Cholesterol DistributionTotal Cholesterol Distribution CHD vs Non-CHD PopulationCHD vs Non-CHD Population

Adapted from Castelli WP. Adapted from Castelli WP. AtherosclerosisAtherosclerosis. 1996;124(suppl):S1-S9. 1996;124(suppl):S1-S9..

35% of CHD 35% of CHD occurs in people occurs in people with TC<200 mg/dLwith TC<200 mg/dL

150 200

Total Cholesterol (mg/dL)

250 300

Framingham Heart Study—26-Year Follow-up

Percentage of MRFIT MenPercentage of MRFIT Men with 1+, 2+, or 3+ Risk Factors by age groupwith 1+, 2+, or 3+ Risk Factors by age group

RISK FACTOR

STATUS

AGES 35-39 AGES 40-57

CHD Death All Other CHD Death All Other

Number 1,112 67,572 12,689 237,767

1+ RISK FACTORS

98.6%

(97.9-99.3)

89.8%

(89.6-90.0)

98.8%

(98.6-99.0)

93.6%

(93.5-93.7)

2+ RISK FACTORS

85.8%

(83.7-87.9)

57.8%

(57.4-58.2)

86.1%

(85.5-86.7)

65.6%

(65.4-65.8)

3+ RISK FACTORS

46.9%

(44.0-49.8)

17.0%

(16.7-17.3)

38.9%

(38.1-39.7)

18.4%

(18.2-18.6)

19-year follow-up for CHD mortality

Adapted from Greenland et al. Adapted from Greenland et al. JAMAJAMA. 2003; 290: 891-7.. 2003; 290: 891-7.

Subclinical disease: No manifest disease, but Subclinical disease: No manifest disease, but pathogenic changes have occurredpathogenic changes have occurred

Identify people at high risk if subclinical CVD Identify people at high risk if subclinical CVD can be detected non-invasively before it can be detected non-invasively before it produces clinical signs and symptomsproduces clinical signs and symptoms

Subclinical Cardiovascular Subclinical Cardiovascular Disease (CVD)Disease (CVD)

Measures of Subclinical CVDMeasures of Subclinical CVD

Examples:Examples:

• Coronary Calcium ScoreCoronary Calcium Score

• Carotid IMTCarotid IMT

• Flow Mediated VasodilationFlow Mediated Vasodilation

• Ankle Brachial IndexAnkle Brachial Index

• Cardiac MRI measuresCardiac MRI measures

• BiomarkersBiomarkers

Coronary Calcium and the Prediction Coronary Calcium and the Prediction of Coronary Eventsof Coronary Events

No CoronaryArtery Calcium

Extensive CoronaryArtery Calcium

EBT EBT CoronaryCoronary Calcium Screen Calcium Screen

Graph showing square-root sum of coronary calcium areas (mm) by electron-beam computed tomography vs. square-root

sum of atherosclerotic plaque areas (mm) for each of the individual coronary arteries studied (n=38)

Rumberger, J. A. et al. Rumberger, J. A. et al. CirculationCirculation. 1995;92:2157-62. 1995;92:2157-62

Studies on Coronary Calcium as aStudies on Coronary Calcium as aPredictor of Coronary EventsPredictor of Coronary Events

Study SizeAverage

Follow-up (Months)

Events Risk Ratio

Detrano et al.JACC 1996

4221

(55 11 yrs) 30

MI + cardiac deaths

Q4 vs. Q1RR=10

Arad et al.

Circulation 199611732

(55 11 yrs)19

MI, CABG, PTCA, cardiac death

Calcium score >160 vs. 160

OR = 35.4

Arad et al.

JACC 1998 Abstract43 OR=23

Secci et al.Circulation 1997

3263

(66 8 yrs) 32MI + cardiac deathrevasc

Log calcium score5

OR=1.24(NS)OR=2.87

Agatston et al.Circulation 1996

3672

(52 yrs) 36-72Angina, MI,angiography,CABG, PTCA

Calcium score ≥ 50 vs. 0OR = 6.87

Detrano et al.Circulation 1999

11964

(66 8 yrs)44 MI, cardiac

death, revasc

No significant improvement of ROC curve area6

11 Patients underwent angiography Patients underwent angiography22 Asymptomatic subjects Asymptomatic subjects3 High-risk subjects High-risk subjects 4 4 High-risk asymptomatic subjectsHigh-risk asymptomatic subjects55 Adjusting for age and CVD risk factors Adjusting for age and CVD risk factors66 No significant improvement of ROC curve area in addition to CVD Risk model No significant improvement of ROC curve area in addition to CVD Risk model

Association Between CAC Score Association Between CAC Score Categories and Hard Events in MenCategories and Hard Events in Men

CAC Score CAC Score RangeRange

Relative RiskRelative Risk 95% CI95% CI

0.00.0 1.01.0† ReferentReferent

Quartile 1Quartile 1 1.0-3.81.0-3.8 1.761.76 0.39-7.880.39-7.88

Quartile 2Quartile 2 4.0-30.54.0-30.5 2.842.84 0.73-11.110.73-11.11

Quartile 3Quartile 3 31-16931-169 5.615.61† 1.57-20.061.57-20.06

Quartile 4Quartile 4 170-7,000170-7,000 7.247.24† 2.01-26.152.01-26.15

The multivariable Cox proportional-hazards regression analysis was performed with adjustment for age and other CAD risk factors.†0.001<P<0.01

Kondos, et al. Circulation. 2003;107:2571-6.

Multivariable Association Between CAC Multivariable Association Between CAC and Cardiac Events (Hazard Ratio)and Cardiac Events (Hazard Ratio)

MenMen WomenWomen

Hard Hard eventsevents 3.86* (1.17-12.70)3.86* (1.17-12.70) 1.53 (0.23-10.09)1.53 (0.23-10.09)

Soft eventsSoft events 26.826.8†† (3.72-193.11) (3.72-193.11) 3.08* (1.11-8.58)3.08* (1.11-8.58)

All eventsAll events 10.4610.46† † (3.85-28.40)(3.85-28.40) 2.57* (1.06-6.23)2.57* (1.06-6.23)

*0.01 < P < 0.05†0.001 < P < 0.01

Kondos, et al. Circulation. 2003;107:2571-6.

MESA Findings on Coronary MESA Findings on Coronary Calcium and the Prediction of Calcium and the Prediction of

Coronary EventsCoronary Events

MESAMulti-Ethnic Study Multi-Ethnic Study of Atherosclerosisof Atherosclerosis

• An NHLBI-sponsored multi-center study investigating prevalence, correlates, and progression of subclinical CVD (i.e., disease detected noninvasively before it produces clinical signs and symptoms) in a population-based sample of 6,814 men and women aged 45-84.

• 6 U.S. field centers recruited healthy men and women (~50% each sex) from 4 racial/ethnic groups (42% white, 13% Chinese, 24% African-American, and 21% Hispanic).

MESAObjectives

To determine characteristics related to progression To determine characteristics related to progression of subclinical to clinical CVD.of subclinical to clinical CVD.

To determine characteristics related to progression To determine characteristics related to progression of subclinical CVD.of subclinical CVD.

To assess ethnic, age, and gender differences in To assess ethnic, age, and gender differences in subclinical disease prevalence and risk of subclinical disease prevalence and risk of progression to clinical cardiovascular diseaseprogression to clinical cardiovascular disease

Multi-Ethnic Study Multi-Ethnic Study of Atherosclerosisof Atherosclerosis

Collaborating Centers Collaborating Centers in MESAin MESA

• Northwestern UnivNorthwestern Univ

• Univ of MNUniv of MN

• Wake ForestWake Forest

•Johns HopkinsJohns Hopkins

•ColumbiaColumbia

• U of VermontU of Vermont

• UCLA (2)UCLA (2)

• Univ of WashingtonUniv of Washington

• New Engl Med CntrNew Engl Med Cntr

WhiteWhite

(n=1218)(n=1218)

African AAfrican A

(n=815)(n=815)

HispanicHispanic

(n=700)(n=700)

Chinese Chinese

(n=379)(n=379)

P Value†P Value†

Age, yrsAge, yrs 63.363.3 62.862.8 61.7***61.7*** 63.163.1 .009.009

% <HS% <HS 3.83.8 12.4***12.4*** 42.1***42.1*** 16.7***16.7*** < .0001< .0001

BMIBMI 27.927.9 28.8***28.8*** 28.7***28.7*** 24.0***24.0*** < .0001< .0001

Waist, cmWaist, cm 100.8100.8 100.7100.7 100.8100.8 87.7***87.7*** < .0001< .0001

% current smoker% current smoker 11.311.3 19.5***19.5*** 15.7**15.7** 9.99.9 < .0001< .0001

% former smoker% former smoker 48.848.8 42.8**42.8** 45.345.3 36.9***36.9*** .0003.0003

Alcohol, dk/wkAlcohol, dk/wk 6.86.8 4.4***4.4*** 5.2***5.2*** 2.8***2.8*** < .0001< .0001

Cohort Characteristics - MenCohort Characteristics - Men

Age adjusted; *p< 0.05; **p<0.01; ***p<0.001 compared to White; Age adjusted; *p< 0.05; **p<0.01; ***p<0.001 compared to White; ††PP values for overall group comparisons values for overall group comparisons

WhiteWhite African AAfrican A HispanicHispanic ChineseChinese P ValueP Value§§

SBP, mmHgSBP, mmHg 123.6123.6 130.1***130.1*** 126.2**126.2** 123.5123.5 <.0001<.0001

DBP, mmHgDBP, mmHg 73.773.7 77.1***77.1*** 74.8*74.8* 74.874.8 <.0001<.0001

LDL-C, mg/dlLDL-C, mg/dl 117.3117.3 113.5**113.5** 118.9118.9 116.7116.7 .006.006

HDL-C, mg/dlHDL-C, mg/dl 45.345.3 46.8**46.8** 42.8***42.8*** 45.945.9 <.0001<.0001

TG, mg/dLTG, mg/dL 127.9127.9 105.9***105.9*** 153.4***153.4*** 136.2*136.2* <.0001<.0001

Glucose mg/dlGlucose mg/dl 101.2101.2 108.9***108.9*** 113.4***113.4*** 109.0***109.0*** <.0001<.0001

% Hypertension% Hypertension†† 36.836.8 57.1***57.1*** 38.638.6 34.634.6 <.0001<.0001

% Diabetes% Diabetes 8.18.1 20.4***20.4*** 20.6***20.6*** 14.1***14.1*** <.0001<.0001

% Chol Rx% Chol Rx 18.418.4 15.0*15.0* 13.2**13.2** 13.6*13.6* .008.008Age adjusted;*p< 0.05; **p<0.01; ***p<0.001 compared to White; Age adjusted;*p< 0.05; **p<0.01; ***p<0.001 compared to White; ††BP >140/90 or reported to be on anti-BP >140/90 or reported to be on anti-hypertensive medication; hypertensive medication; ‡‡Fasting glucose >126 mg/dl or on hypoglycemic medication;Fasting glucose >126 mg/dl or on hypoglycemic medication;§§PP values for overall group values for overall group comparisonscomparisons

Cohort Characteristics - Men, continuedCohort Characteristics - Men, continued

WhiteWhite

(n=1313)(n=1313)

African AAfrican A

(n=1019)(n=1019)

HispanicHispanic

(n=750)(n=750)

Chinese Chinese

(n=404)(n=404)

P ValueP Value††

Age, yrsAge, yrs 63.063.0 62.662.6 62.162.1 62.862.8 .26.26

% <HS% <HS 5.55.5 11.7***11.7*** 48.1***48.1*** 31.3***31.3*** < .0001< .0001

BMIBMI 27.527.5 31.3***31.3*** 30.0***30.0*** 23.9***23.9*** < .0001< .0001

Waist, cmWaist, cm 94.994.9 101.6***101.6*** 100.5***100.5*** 86.4***86.4*** < .0001< .0001

% current smoker% current smoker 11.711.7 15.9**15.9** 9.89.8 1.3***1.3*** < .0001< .0001

% former smoker% former smoker 39.939.9 31.9***31.9*** 21.3***21.3*** 2.2***2.2*** < .0001< .0001

Alcohol, dk/wkAlcohol, dk/wk 3.43.4 1.9***1.9*** 1.2***1.2*** 0.8***0.8*** < .0001< .0001

Age adjusted; *p< 0.05; **p<0.01; ***p<0.001 compared to White; Age adjusted; *p< 0.05; **p<0.01; ***p<0.001 compared to White; ††PP values for overall group comparisons values for overall group comparisons

Cohort Characteristics - WomenCohort Characteristics - Women

WhiteWhite African AAfrican A HispanicHispanic ChineseChinese P ValueP Value§§

SBP, mmHgSBP, mmHg 122.4122.4 132.9***132.9*** 128.3***128.3*** 125.1*125.1* <.0001<.0001

DBP, mmHgDBP, mmHg 66.966.9 72.5***72.5*** 68.4**68.4** 69.2***69.2*** <.0001<.0001

LDL-C, mg/dlLDL-C, mg/dl 117.1117.1 118.7118.7 119.8119.8 113.9113.9 .01.01

HDL-C, mg/dlHDL-C, mg/dl 59.059.0 57.0**57.0** 52.7***52.7*** 53.3***53.3*** <.0001<.0001

TG, mg/dlTG, mg/dl 126.8126.8 99.2***99.2*** 143.1***143.1*** 137.0**137.0** <.0001<.0001

Glucose mg/dlGlucose mg/dl 94.894.8 105.2***105.2*** 107.9***107.9*** 102.5***102.5*** <.0001<.0001

% Hypertension% Hypertension†† 36.436.4 61.1***61.1*** 46.5***46.5*** 40.340.3 <.0001<.0001

% Diabetes% Diabetes 4.94.9 17.1***17.1*** 17.4***17.4*** 13.3***13.3*** <.0001<.0001

% Chol Rx% Chol Rx 16.216.2 18.418.4 15.515.5 15.515.5 .26.26

Cohort Characteristics - Women, continuedCohort Characteristics - Women, continued

Age adjusted;*p< 0.05; **p<0.01; ***p<0.001 compared to White; Age adjusted;*p< 0.05; **p<0.01; ***p<0.001 compared to White; ††BP >140/90 or reported to be on anti-BP >140/90 or reported to be on anti-hypertensive medication; hypertensive medication; ‡‡Fasting glucose >126 mg/dl or on hypoglycemic medication;Fasting glucose >126 mg/dl or on hypoglycemic medication;§§PP values for overall group values for overall group comparisonscomparisons

55

3138

25

43

22

44

30

0

20

40

60

80

100

Men Women

White

African American

Hispanic

Chinese

***

*p< 0.05, **p<0.01, ***p<0.001 compared to White.*p< 0.05, **p<0.01, ***p<0.001 compared to White.

******

******

Age-Adjusted Prevalence of Coronary Age-Adjusted Prevalence of Coronary Calcium (CAC) Calcium (CAC) >> 20 20

Unadjusted Kaplan–Meier Cumulative-Event Curves for Unadjusted Kaplan–Meier Cumulative-Event Curves for Major Incident CHD by CAC Score, MESAMajor Incident CHD by CAC Score, MESA

Rates for major coronary events (myocardial infarction and Rates for major coronary events (myocardial infarction and death from coronary heart disease)death from coronary heart disease)

Differences among all curves are statistically significant (Differences among all curves are statistically significant (PP<0.001).<0.001).

Detrano, et al. Detrano, et al. N Engl J MedN Engl J Med. 2008;358:13.. 2008;358:13.

Rates for any coronary event

Differences among all curves are statistically significant (Differences among all curves are statistically significant (PP<0.001).<0.001).


Unadjusted Kaplan–Meier Cumulative-Event Curves for Unadjusted Kaplan–Meier Cumulative-Event Curves for Any Incident CHD by CAC Score, MESAAny Incident CHD by CAC Score, MESA

Risk Factor Adjusted Risk of Incident Risk Factor Adjusted Risk of Incident CHD by CAC Score, MESACHD by CAC Score, MESA

Coronary-Coronary-Artery Calcium Artery Calcium ScoreScore

Major Coronary EventMajor Coronary Event†† Any Coronary EventAny Coronary Event

No./No. No./No. at Riskat Risk

Hazard RatioHazard Ratio

(95% CI)(95% CI)

PP ValueValue

No./No. at No./No. at RiskRisk

Hazard RatioHazard Ratio

(95% CI)(95% CI)

PP ValueValue

00 8/34098/3409 1.001.00 15/340915/3409 1.001.00

1-1001-100 25/172825/1728 3.89 (1.72-8.79)3.89 (1.72-8.79) <0.001<0.001 39/172839/1728 3.61 (1.96-6.65)3.61 (1.96-6.65) <0.001<0.001

101-300101-300 24/75224/752 7.08 (3.05-16.47)7.08 (3.05-16.47) <0.001<0.001 41/75241/752 7.73 (4.13-14.47)7.73 (4.13-14.47) <0.001<0.001

>300>300 32/83332/833 6.84 (2.93-15.99)6.84 (2.93-15.99) <0.001<0.001 67/83367/833 9.67 (5.20-17.98)9.67 (5.20-17.98) <0.001<0.001

LogLog22 (CAC+1) (CAC+1) §§ 1.20 (1.12-1.29)1.20 (1.12-1.29) <0.001<0.001 1.26 (1.19-1.33)1.26 (1.19-1.33) <0.001<0.001

†Major coronary events are myocardial infarction and death from coronary heart disease§Each unit increase in log2 (CAC+1) represents a doubling of the coronary-artery calcium score

Detrano, et al. N Engl J Med. 2008;358:13.

Risk of Incident CHD Associated with CAC Risk of Incident CHD Associated with CAC Score in Four Racial or Ethnic Groups*Score in Four Racial or Ethnic Groups*

Racial or Racial or Ethnic Ethnic GroupGroup


No.No. Hazard RatioHazard Ratio

(95% CI)(95% CI)§§

PP Value Value No.No. Hazard RatioHazard Ratio

(95% CI)(95% CI)

PP Value Value

WhiteWhite 4141 1.17 (1.06-1.30)1.17 (1.06-1.30) <0.005<0.005 7474 1.22 (1.13-1.22 (1.13-1.32)1.32)

<0.001<0.001

ChineseChinese 66 1.25 (0.95-1.63)1.25 (0.95-1.63) 0.110.11 1414 1.36 (1.12-1.36 (1.12-1.66)1.66)

<0.005<0.005

BlackBlack 1818 1.35 (1.16-1.57)1.35 (1.16-1.57) <0.001<0.001 3838 1.39 (1.25-1.39 (1.25-1.56)1.56)

<0.001<0.001

HispanicHispanic 2424 1.15 (1.02-1.29)1.15 (1.02-1.29) <0.025<0.025 3636 1.18 (1.07-1.18 (1.07-1.30)1.30)

<0.001<0.001*CAC denotes coronary-artery calcium score and CI confidence interval†Major coronary events are myocardial infarction and death from coronary heart disease§Hazard ratios were calculated with the use of Cox regression for coronary heart disease (major event and any event) for baseline levels of log2 (CAC+1) after adjustment for risk factors and interactions between racial or ethnic group and coronary calcium score and between racial or ethnic group and diabetes (the only significant interaction). Hazard ratios are calculated on the basis of a doubling of CAC+1

Detrano, et al. N Engl J Med. 2008;358:13.

Area under the Curve (AUC), Risk Factors vs. Risk Area under the Curve (AUC), Risk Factors vs. Risk Factors plus CAC Score to Predict Incident CHDFactors plus CAC Score to Predict Incident CHD

by Race or Ethnicity by Race or Ethnicity**, MESA Study, MESA Study

Racial or Racial or Ethnic Ethnic GroupGroup


AUC for AUC for Risk Risk

Factors Factors AloneAlone

AUC for Risk AUC for Risk Factors plus Factors plus

Coronary Artery Coronary Artery Calcium ScoreCalcium Score

PP Value Value AUC for Risk AUC for Risk Factors AloneFactors Alone

AUC for Risk AUC for Risk Factors plus Factors plus

Coronary Artery Coronary Artery Calcium ScoreCalcium Score

PP Value Value

WhiteWhite 0.760.76 0.790.79 0.100.10 0.750.75 0.790.79 0.020.02

ChineseChinese 0.830.83 0.880.88 0.050.05 0.740.74 0.850.85 <0.001<0.001

BlackBlack 0.790.79 0.870.87 0.040.04 0.810.81 0.870.87 0.0050.005

HispanicHispanic 0.840.84 0.860.86 0.110.11 0.800.80 0.840.84 0.100.10

TotalTotal 0.790.79 0.830.83 0.0060.006 0.770.77 0.820.82 <0.001<0.001

*Separate models are fitted for each racial or ethnic group. AUC denotes area under the receiver-operating characteristic *Separate models are fitted for each racial or ethnic group. AUC denotes area under the receiver-operating characteristic curve. P values are for the comparison between AUC without and AUC with the coronary-artery calcium score.curve. P values are for the comparison between AUC without and AUC with the coronary-artery calcium score.††Major coronary events are myocardial infarction and death from coronary heart diseaseMajor coronary events are myocardial infarction and death from coronary heart disease


Hazard Ratios (HRs) for an Incident CVD, CHD, or Stroke Event in Hazard Ratios (HRs) for an Incident CVD, CHD, or Stroke Event in Relation to a 1-SD Increment of Maximal Carotid IMT or CAC Score Relation to a 1-SD Increment of Maximal Carotid IMT or CAC Score

(MESA 2000-04)(MESA 2000-04)

Measure*Measure* HR Per 1-SD IncrementHR Per 1-SD Increment

(95% CI)(95% CI)

Z Z StatisticStatistic PP Value Value

CVD (n = 222)CVD (n = 222)Age-, race-, and sex-adjustedAge-, race-, and sex-adjusted

zz Score max IMT Score max IMT 1.3 (1.1-1.4)1.3 (1.1-1.4) 4.14.1 <.001<.001In(CAC score + 1)In(CAC score + 1) 2.1 (1.8-2.5)2.1 (1.8-2.5) 8.68.6 <.001<.001

Multivariable-AdjustedMultivariable-Adjustedbb

zz Score max IMT Score max IMT 1.2 (1.0-1.3)1.2 (1.0-1.3) 2.72.7 .007.007In(CAC score + 1)In(CAC score + 1) 1.9 (1.6-2.2)1.9 (1.6-2.2) 7.57.5 <.001<.001

CHD (n = 159)CHD (n = 159)Age-, race-, and sex-adjustedAge-, race-, and sex-adjusted




Stroke (n = 59)Stroke (n = 59)Age-, race-, and sex-adjustedAge-, race-, and sex-adjusted

zz Score max IMT Score max IMT 1.4 (1.2-1.8)1.4 (1.2-1.8) 3.53.5 .001.001In(CAC score + 1)In(CAC score + 1) 1.1 (0.8-1.5)1.1 (0.8-1.5) 0.80.8 .41.41


zz Score max IMT Score max IMT 1.3 (1.1-1.7)1.3 (1.1-1.7) 2.52.5 .01.01In(CAC score + 1)In(CAC score + 1) 1.1 (0.8-1.4)1.1 (0.8-1.4) 0.40.4 .71.71

Folsom, et al. AFolsom, et al. Arch Intern Medrch Intern Med. 2008;168:12.. 2008;168:12.Abbreviations: CAC, coronary artery calcium; CHD, coronary heart disease; CVD, cardiovascular disease; IMT, intima-Abbreviations: CAC, coronary artery calcium; CHD, coronary heart disease; CVD, cardiovascular disease; IMT, intima-media thickness; max, maximum; MESA, Multi-Ethnic Study of Atherosclerosismedia thickness; max, maximum; MESA, Multi-Ethnic Study of Atherosclerosis*Coronary artery calcium score and IMT were in the same model *Coronary artery calcium score and IMT were in the same model bbAdjusted as described in the “Methods” sectionAdjusted as described in the “Methods” section

Hazard Ratios (HR) for Incident CVD, CHD, or Stroke Event in Relation Hazard Ratios (HR) for Incident CVD, CHD, or Stroke Event in Relation to Quartiles of Maximal Carotid IMT or CAC Score (MESA 2000-04)to Quartiles of Maximal Carotid IMT or CAC Score (MESA 2000-04)

HR (95% Cl)

Measure* <50th Percentile Quartile 3 Quartile 4 X2 Statistic

P Value

CVD (n = 222)Age-, race-, and sex-adjusted

z Score max IMT 1 [Reference] 1.4 (0.9-2.0) 2.2 (1.5-3.2) 20.1 <.001

CAC Score 1 [Reference] 2.6 (1.6-4.1) 5.3 (3.4-8.2) 58.4 <.001Multivariable-Adjustedb

z Score max IMT 1 [Reference] 1.3 (0.9-2.0) 1.7 (1.2-2.5) 8.7 .01CAC Score 1 [Reference] 2.3 (1.5-3.7) 4.4 (2.8-6.8) 44.7 <.001

CHD (n = 159)Age-, race-, and sex-adjusted

z Score max IMT 1 [Reference] 1.5 (1.0-2.4) 2.1 (1.4-3.3) 11.5 <.01CAC Score 1 [Reference] 4.1 (2.2-2.7) 10.3 (5.6-18.9) 63.8 <.001

Multivariable-Adjustedb

z Score max IMT 1 [Reference] 1.5 (0.9-2.3) 1.7 (1.1-2.7) 5.4 .07CAC score 1 [Reference] 3.5 (1.9-6.6) 8.2 (4.5-15.1) 51.5 <.001

Stroke (n = 59)Age-, race-, and sex-adjusted

z Score max IMT 1 [Reference] 0.9 (0.4-2.0) 2.4 (1.2-4.7) 9.9 <.01

CAC Score 1 [Reference] 1.4 (0.8-2.7) 1.2 (0.6-2.4) 0.7 .70Multivariable-Adjustedb

z Score max IMT 1 [Reference] 0.9 (0.4-2.0) 1.8 (0.9-3.6) 4.7 .10CAC score 1 [Reference] 1.3 (0.6-2.6) 1.0 (0.5-2.1) 0.6 .73

Folsom, et al. AFolsom, et al. Arch Intern Medrch Intern Med. 2008;168:12.. 2008;168:12.Abbreviations: CAC, coronary artery calcium; CHD, coronary heart disease; CVD, cardiovascular disease; IMT, Abbreviations: CAC, coronary artery calcium; CHD, coronary heart disease; CVD, cardiovascular disease; IMT, intima-media thickness; max, maximum; MESA, Multi-Ethnic Study of Atherosclerosisintima-media thickness; max, maximum; MESA, Multi-Ethnic Study of Atherosclerosis*Coronary artery calcium score and IMT were in the same model *Coronary artery calcium score and IMT were in the same model bbAdjusted as described in the “Methods” Adjusted as described in the “Methods” sectionsection

Prevalence of High CAC Score in People Prevalence of High CAC Score in People with Different Risk Factor Status, MESAwith Different Risk Factor Status, MESA

0

10

20

30

40

50

0 1 2 3+ 0 1 2 3+

Number of Risk Factors

% o

f C

AC

≥ 1

00

Men Women

Risk Factors: LDL ≥ 130 mg/dl or Rx; BP 140/90 mmHg or Rx; fasting glucose ≥ 126 mg/dl or Rx; current smoker

% 100 ≤ CAC< 300% 300 ≤ CAC

Improvement of Using CAC to Improvement of Using CAC to Identify People at High Risk for Identify People at High Risk for

Coronary EventsCoronary Events

Current Paradigm for Risk Estimation Current Paradigm for Risk Estimation and Treatment: ATP-IIIand Treatment: ATP-III

Estimate 10-year risk (FRS)

<10% >20% or DM10-20%

Further testing

Lifestyleand drug therapy

Lifestylemodification

“Intensity of prevention efforts should match the absolute risk of the patient”

Calculation of the NRICalculation of the NRI

Proportion of events and nonevents correctly reclassified

Pencina et al. Statist. Med. 2008;27:157-72

Events reclassified higher – events reclassified lower Total number of events

PLUS

Nonevents reclassified lower – nonevents reclassified higher Total number of nonevents

5-Year Risk from Model with CAC

5-Year Risk from Model without CAC

0% to <3%

3% to 10%

>10%

0% to <3%, n=3,746

Events, nProportion with

events

3,31034

0.9

3% to 10%, n=1847Events, nProportion with

events

843525.4

>10%, n=285Events, nProportion with

events

18328

14.4

Participants not reclassified Participants not reclassified



0% to <3%

3% to 10%

>10%

0% to <3%, n=3,746


events

430224.8

61

20.0

3% to 10%, n=1847Events, nProportion with

events

29248

14.8


events

Participants reclassified to higher risk Participants reclassified to higher risk



0% to <3%

3% to 10% >10%

0% to <3%, n=3,746


events3% to 10%, n=1847


events

712151.9


events

322

6.2

707

8.9

Participants reclassified to lower riskParticipants reclassified to lower risk

Calculation of NRICalculation of NRI

NRI for events: 0.23

PLUS = 0.25 (95% CI 0.16-0.34

NRI for non-events: 0.02 P<0.001)

Reclassification among intermediate Reclassification among intermediate risk participantsrisk participants

• Among 1847 intermediate risk participants, 39% were reclassified to low risk and 16% were reclassified to high risk

– NRI = 0.55 (95% CI 0.41-0.69, P<0.001)

– NRI = 0.29 for events and 0.26 for nonevents

• 115 events occurred among intermediate risk participants; 48 were among people reclassified to high risk and 15 were among those reclassified to low risk

CAC Score Group Framingham Risk Score Categories (n = 5660)

0-2.5% (1730)

2.6-5% (1045)

5.1-7.5% (442)

7.6-10% (779)

10.1-15% (617)

15.1-20% (793)

> 20% (254)

p value

Median CAC score > 0 (IQR)

28.6 (7.4, 91.6)

39.7 (11.9, 140.6)

62.5 (15.9, 211.2)

71.5 (19.3, 257)

111.6 (27.7, 284.1)

134.6 (33.5, 427.6)

198.6 (56.5, 483.7)

CAC > 0: n, (%)N = 2626

386 (22.3)

411 (39.3)

198 (44.8)

449 (57.6)

394 (63.9)

579 (73) 209 (82.3) < 0.01

NNS (CAC > 0) 4.5 2.5 2.2 1.7 1.6 1.4 1.2

< 0.01CAC ≥ 100: n, (%)N = 1163

89 (5.1)

132 (12.6)

81 (18.3)

193 (24.8)

205 (33.2)

324 (40.9)

139 (54.7)

NNS (CAC ≥ 100) 19.4 7.9 5.5 4.0 3.0 2.5 1.8

CAC ≥ 300: n, (%)N = 574

29 (1.7)

46 (4.4) 33 (7.5)102 (13.1)

96 (15.6)191 (24.1)

77 (30.3) < 0.01

NNS (CAC ≥ 300) 59.7 22.7 13.4 7.6 6.4 4.2 3.3

CAC Prevalence, Amount and ‘Number Needed to Screen’ Compared with Framingham Risk Score Categories

IQR: Interquartile Range CAC (25th percentile, 75th percentile)NNS: Number needed to screen to identify one individual with CAC value above a specified CAC cutoff point, within each specified FRS stratum

Okwuosa TM, et al. Distribution of Coronary Artery Calcium Scores by Framingham 10-Year Risk Strata in the Multi-Ethnic Study of Atherosclerosis (MESA): Potential Implications for Coronary Risk Assessment. J Am Coll Cardiol 2011.

ConclusionsConclusions

Coronary calcium score is strongly associated with Coronary calcium score is strongly associated with incident coronary heart diseaseincident coronary heart disease and provides and provides additional predictive information beyond that provided additional predictive information beyond that provided by standard risk factors in four major racial and ethnic by standard risk factors in four major racial and ethnic groups in the U.S.A.groups in the U.S.A.

However the CT scan to detect coronary calcium However the CT scan to detect coronary calcium produces radiation exposure that is equivalent to one produces radiation exposure that is equivalent to one year’s exposure from natural sources. year’s exposure from natural sources.

Coronary calcium scores may be used to screen Coronary calcium scores may be used to screen asymptomatic patients who have multiple CVD risk asymptomatic patients who have multiple CVD risk factors or whose Framingham Risk Scores are in the factors or whose Framingham Risk Scores are in the intermediate range.intermediate range.

0

10

20

30

40

50

60

70

Age-Adjusted CHD Mortality Rate per 10,000 Age-Adjusted CHD Mortality Rate per 10,000 Person Years by SBP, Cholesterol, and SmokingPerson Years by SBP, Cholesterol, and Smoking

MRFIT Screenee StudyMRFIT Screenee Study

Neaton JD, Neaton JD, Wentworth D. Wentworth D. Arch Intern MedArch Intern Med. . 1992;152:561992;152:56

Smokers

Non-Smokers

SBP (mmHg) Cholesterol

(ml/dl)

Additional FindingsAdditional Findings

Lumbar Bone Density and ABI, CAC, and AACLumbar Bone Density and ABI, CAC, and AAC Baseline CAC and coronary artery stenoses during Baseline CAC and coronary artery stenoses during

follow-up follow-up ECG abnormalities and CAC and carotid IMTECG abnormalities and CAC and carotid IMT Incidence and progression of aortic valve calcium; Incidence and progression of aortic valve calcium;

relationship between aortic valve calcium and severity relationship between aortic valve calcium and severity of CACof CAC

Arterial age as a function of CACArterial age as a function of CAC

Incident CHD and CVD by CAC Score in Low Incident CHD and CVD by CAC Score in Low RiskRisk11 Women Based on FRS, MESA Women Based on FRS, MESA

CAC ScoreCAC Score No. of Events/No. of Events/

Total No. (%)Total No. (%)

Unadjusted HRUnadjusted HR PP Value Value Adjusted HR Adjusted HR ‡‡ PP Value Value

CHDCHD00 6/1814 (0.3)6/1814 (0.3) 1 [Reference]1 [Reference] 1 [Reference]1 [Reference]

1-991-99 8/589 (1.4)8/589 (1.4) 4.2 (1.5-12.0)4.2 (1.5-12.0) .008.008 2.4 (0.8-7.3)2.4 (0.8-7.3) .12.12100-299100-299 3/176 (1.7)3/176 (1.7) 5.7 (1.4-22.9)5.7 (1.4-22.9) .01.01 1.5 (0.3-8.3)1.5 (0.3-8.3) .63.63≥≥300300 7/105 (6.7)7/105 (6.7) 22.3 (7.5-66.5)22.3 (7.5-66.5) <.001<.001 8.3 (2.3-30.0)8.3 (2.3-30.0) .001.001

CVDCVD00 10/1814 (0.6)10/1814 (0.6) 1 [Reference]1 [Reference] 1 [Reference]1 [Reference]

1-991-99 11/589 (1.9)11/589 (1.9) 3.4 (1.5-8.1)3.4 (1.5-8.1) .005.005 2.0 (0.8-4.9)2.0 (0.8-4.9) .13.13100-299100-299 4/176 (2.3)4/176 (2.3) 4.5 (1.4-14.3)4.5 (1.4-14.3) .01.01 1.4 (0.4-5.6)1.4 (0.4-5.6) .62.62≥≥300300 9/105 (8.6)9/105 (8.6) 17.3 (7.0-42.5)17.3 (7.0-42.5) <.001<.001 6.0 (2.1-17.2)6.0 (2.1-17.2) .001.001

11Low Risk: FRS<10%Low Risk: FRS<10%‡‡Adjusted for age, ethnicity, body mass index, low-density lipoprotein cholesterol level, hypertension, Adjusted for age, ethnicity, body mass index, low-density lipoprotein cholesterol level, hypertension, smoking, family history of CHD, estrogen use and statin usesmoking, family history of CHD, estrogen use and statin use

Lakoski, et al. Lakoski, et al. Arch Intern MedArch Intern Med. 2007;167:22.. 2007;167:22.

ResultsResults

• Final cohort of 5878 participants

• 209 CHD events, overall event rate 3.5%

– 122 major events (96 MI, 14 death from CHD, 12 resuscitated cardiac arrest)

– 87 angina events (73 followed by revascularization, 14 not followed by revascularization)

• AUROC using traditional risk factors was 0.76, increased to 0.81 with the addition of CAC (P<0.001)

• Both models well calibrated

– HL χ2 statistic model 1: 6.72, P=0.46;

model 2: 9.15 P=0.24

Polonsky et al JAMA 2010

Coronary Calcium and the Prediction of Coronary Heart Disease

Documents

Transcript of Coronary Calcium and the Prediction of Coronary Heart Disease