nejmoa1012592

8/3/2019 NEJMoa1012592

1/9

T he n e w e n g l a n d j o u r n a l o f medicine

n engl j med 365;3 nejm.org july 21, 2011 213

original article

Carotid-Wall IntimaMedia Thickness

and Cardiovascular EventsJoseph F. Polak, M.D., M.P.H., Michael J. Pencina, Ph.D.,

Karol M. Pencina, Ph.D., Christopher J. ODonnell, M.D., M.P.H.,Philip A. Wolf, M.D., and Ralph B. DAgostino, Sr., Ph.D.

From the Department of Radiology, TuftsMedical Center (J.F.P.); and the Depart-

ments of Biostatistics (M.J.P.), Mathe-matics and Statistics (K.M.P., R.B.D.), andMedicine and Neurology (P.A.W.), BostonUniversity both in Boston; and the Na-tional Heart, Lung, and Blood Institute,Division of Intramural Research, Framing-ham, MA (C.J.O.). Address reprint requeststo Dr. Polak at the Department of Radiol-ogy, Tufts Medical Center, 800 Washing-ton St., Box 299, Boston, MA 02111, or [email protected].

N Engl J Med 2011;365:213-21.Copyright 2011 Massachusetts Medical Society.

A B S T RA C T

Background

Intimamedia thickness of the walls of the common carotid artery and internal ca-rotid artery may add to the Framingham risk score for predicting cardiovascular events.

Methods

We measured the mean intimamedia thickness of the common carotid artery andthe maximum intimamedia thickness of the internal carotid artery in 2965 mem-bers of the Framingham Offspring Study cohort. Cardiovascular-disease outcomeswere evaluated for an average follow-up of 7.2 years. Multivariable Cox proportional-hazards models were generated for intimamedia thickness and risk factors. We eval-uated the reclassification of cardiovascular disease on the basis of the 8-year Framing-ham risk score category (low, intermediate, or high) after adding intimamediathickness values.

Results

A total of 296 participants had a cardiovascular event. The risk factors of the Fram-ingham risk score predicted these events, with a C statistic of 0.748 (95% confidenceinterval [CI], 0.719 to 0.776). The adjusted hazard ratio for cardiovascular diseasewith a 1-SD increase in the mean intimamedia thickness of the common carotidartery was 1.13 (95% CI, 1.02 to 1.24), with a nonsignif icant change in the C statisticof 0.003 (95% CI, 0.000 to 0.007); the corresponding hazard ratio for the maximumintimamedia thickness of the internal carotid artery was 1.21 (95% CI, 1.13 to1.29), with a modest increase in the C statistic of 0.009 (95% CI, 0.003 to 0.016).The net reclassification index increased signif icantly after addition of intimamediathickness of the internal carotid artery (7.6%, P

8/3/2019 NEJMoa1012592

2/9

Th e n e w e n g l a n d j o u r n a l o f medicine

n engl j med 365;3 nejm.org july 21, 2011214

Carotid-wall intimamedia thicknessis a surrogate measure of atherosclerosis1associated with cardiovascular risk factors2

and with cardiovascular outcomes.3-7 The intimamedia thickness is the distance from the lumenintima interface to the mediaadventitia interfaceof the artery wall, as measured on noninvasively

acquired ultrasonographic images of the carotidarteries. Increased intimamedia thickness of thecommon carotid artery represents a form of athero-sclerosis that is manifested as dif fuse arterial-wall thickening, 8 whereas increased intimamediathickness of the proximal internal carotid arteryis a surrogate for focal atherosclerotic plaque.9The mean intimamedia thickness of the commoncarotid artery is a more reproducible measurethan the intimamedia thickness of the internalcarotid artery and is believed to be better suitedfor cardiovascular risk assessment10 and inter-

vention studies.11 Although measurement of in-timamedia thickness is promoted as a tool forcardiovascular risk assessment12,13 in primaryprevention, the incremental predictive value ofthe intimamedia thickness of either the com-mon carotid artery or the internal carotid artery,over and above the value of traditional cardio-vascular risk factors, is questionable.14,15

We hypothesized that the intimamedia thick-nesses of the common carotid artery and inter-nal carotid artery would add to the predictive valueof Framingham risk factors regarding new-onsetcardiovascular events in a population-based co-hort; we tested this hypothesis in the Framing-ham Offspring Study cohort. We also hypoth-esized that consideration of the intimamediathickness would improve the classification of pa-tients into Framingham risk score categories.

Methods

Study Participants

The study population consisted of members of

the Framingham Offspring Study cohort, com-posed of non-Hispanic whites, who were under-going the sixth examination cycle, from February1995 through September 1998. Of the 3532 per-sons seen during the clinic visit, 2965 who didnot have current disease underwent ultrasonog-raphy, of whom 2946 had interpretable images ofthe internal carotid artery. Missing data were dueto scheduling issues or unavailability of the ultra-sonographic device. Details of the Framingham

Offspring Study design have been published pre-viously.16 All participants in our study providedwritten informed consent, and the institutionalreview board at Boston Medical Center approvedour study protocol.

Assessment of Risk Factors

During the clinic visit, a medical history was ob-tained, and a physical examination was performed.The measured cardiovascular risk factors relevantto our study were systolic blood pressure, cigarette-smoking status, total cholesterol level, high-densi-ty lipoprotein (HDL) cholesterol level, and presenceor absence of hypertension treatment, diabetes,and a history of cardiovascular disease, includingstroke.

The systolic blood pressure was determinedfrom the average of two measurements of restingsystolic blood pressure, performed by a physician

using a 14-cmwide blood-pressure cuff on theright arm. Smoking status was ascertained on thebasis of the self-reported history of cigarette smok-ing. The presence of diabetes was based on a his-tory of diabetes, defined as either a current orprevious fasting glucose level of 126 mg per decili-ter (7 mmol per liter) or higher or current or previ-ous use of antihyperglycemic medication. All lipidanalyses were performed at the Framingham HeartStudy laboratory according to the StandardizationProgram of the Centers for Disease Control andPrevention and the Lipid Research Clinics of theNational Heart, Lung, and Blood Institute.17,18

Assessment of IntimaMedia Thickness

Ultrasonographic images were acquired at end di-astole (defined as the R wave of an electrocardio-gram) by a sonographer certified by the Registry ofDiagnostic Medical Sonographers. Intimamediainterface lines were manually traced as continuouslines by a certif ied reader, and intimamedia thick-ness values were calculated.19 The mean intimamedia thickness of the common carotid artery was

measured over a segment of the common carotidartery that was 1 cm long, located approximately0.5 cm below the carotid-artery bulb, and consid-ered not to contain any plaque (i.e., not to have anyperceivable protrusion of the artery wall into thelumen).20 The maximum intimamedia thicknessof the internal carotid artery was defined as thegreatest intimamedia thickness in either the rightor left internal carotid artery extending from thebulb to 1 cm above the carotid sinus, ascertained

The New England Journal of Medicine

Downloaded from nejm.org by jefri efendi on September 24, 2011. For personal use only. No other uses without permission.

Copyright 2011 Massachusetts Medical Society. All rights reserved.

8/3/2019 NEJMoa1012592

3/9

Carotid IntimaMedia Thickness and Cardiovascular Events


from a total of four views on each side. Repro-ducibility was assessed by replicating measure-ments for 37 participants.19 Pearsons correlationcoefficient for replicate readings was 0.94 for themean intimamedia thickness of the commoncarotid artery and 0.76 for the maximum intimamedia thickness of the internal carotid artery.

Incident Cardiovascular Disease

All cardiovascular events in the Framingham Off-spring Study cohort were adjudicated by a panelof three physicians, on the basis of a review of datacollected from Framingham clinic visits, inpatienthospitalizations, and office records. For this study,we used the Framingham Heart Study definitionof cardiovascular disease: coronary heart disease(i.e., a fatal coronary event, myocardial infarction,coronary insufficiency, or angina), a cerebrovascu-lar event (i.e., ischemic stroke, hemorrhagic stroke,

or transient ischemic attack), peripheral arterialdisease (i.e., intermittent claudication), or heartfailure.21

Statistical Analysis

Multivariable Cox proportional-hazards modelswere generated, with the following candidate vari-ables: age, sex, systolic blood pressure, total cho-lesterol level, HDL cholesterol level, presence orabsence of diabetes, and smoking status theoriginal components of the 10-year Framinghamrisk score for general cardiovascular disease. Pres-ence or absence of hypertensive treatment was add-ed to the models.21

Two separate models were then created by add-ing the mean intimamedia thickness of thecommon carotid artery and the maximum intimamedia thickness of the internal carotid artery asseparate predictor variables. Hazard ratios weregenerated for each independent variable in each ofthe two models. We evaluated differences betweenthe two models, after confirming the calibrationof the original Cox models, by comparing the re-

spective C statistics before and after the additionof the variables for intimamedia thickness.22

The incremental effect of adding intimamediathickness to the Framingham risk score for pre-dicting cardiovascular outcomes was evaluatedwith the use of the net reclassification index.23First, we stratified participants into one of threerisk categories on the basis of their Framinghamrisk score for general cardiovascular disease (cal-culated for this study as an 8-year score): low risk

(0 to 20%). Then, we used the intimamediathickness to reclassify the risk category, specifi-cally to ascertain whether there would be improve-ment in the net reclassification index that is,whether reclassification would assign persons inwhom cardiovascular disease developed to a high-

er risk category and those in whom cardiovascu-lar disease did not develop to a lower risk catego-ry.23 The net reclassification index also accountsfor movement between categories in the oppositedirection than predicted and applies differentweights to events and nonevents.23 Separate net-reclassification-index values were calculated formen and women as well as for participants whowere 60 years of age or younger and participantswho were older than 60 years at baseline, on thebasis of the probability of a cardiovascular eventand the probability of no event, as predicted from

pooled models (combining women and men andall ages) with and without inclusion of intimamedia thickness of the internal carotid artery.

We also investigated models in which the pres-ence of plaque, defined as an intimamedia thick-ness of more than 1.5 mm, was used as a predictorvariable and examined how the presence of plaqueaffected the prediction of events within the Fram-ingham risk score categories, using log-rank statis-tics across all strata. All analyses were performedwith the use of SAS software (version 9.1), and atwo-sided P value of less than 0.05 was consid-ered to indicate statistical significance.

Results

Characteristics of the Participants

A total of 2965 participants with a mean (SD) ageof 5810 years and no history of cardiovasculardisease, 1629 (55.3%) of whom were women, werefollowed for an average of 7.2 years. There were296 first-time cardiovascular events. The baselinecharacteristics of the participants are shown in

Table 1, and a breakdown of types of cardiovas-cular events is given in the Supplementary Appen-dix (available with the full text of this article atNEJM.org).

Predictive Value of IntimaMedia Thickness

Multivariable Cox proportional-hazards modelswere constructed to include Framingham risk fac-tors only, risk factors plus the mean intimamediathickness of the common carotid artery (Table 2),




8/3/2019 NEJMoa1012592

4/9



and risk factors plus the maximum intimamediathickness of the internal carotid artery (Table 3).The results showed that the Framingham risk fac-tors were all significant predictors of cardiovascu-lar disease. The addition of the mean intimamediathickness for the common carotid artery was sig-nificantly associated with the risk of cardiovascu-lar disease (hazard ratio per 1-SD increase in thick-ness, 1.13; 95% confidence interval [CI], 1.02 to1.24; P = 0.02). However, the C statistic increasedonly nonsignif icantly, by 0.003 (95% CI, 0.000 to0.007; P = 0.07), from 0.748 (95% CI, 0.719 to 0.776)to 0.751 (95% CI, 0.723 to 0.779). The maximum

intimamedia thickness of the internal carotid ar-tery was also significantly associated with the riskof cardiovascular disease (hazard ratio per 1-SDincrease in thickness, 1.21; 95% CI, 1.13 to 1.29;P

8/3/2019 NEJMoa1012592

5/9



carotid artery also modestly but signif icantly add-ed predictive value to the Framingham risk scorewhen dichotomized at a threshold used to defineplaque (>1.5 mm, vs. 1.5 mm for no plaque).

Ultrasonographic measurements of intimamedia thickness can be limited to the commoncarotid artery,5,7 averaged across multiple carotid-artery segments,4,6 or combined as a score.3 Areview of eight epidemiologic studies showed thatthe intimamedia thickness of the common ca-rotid artery by itself (in all eight studies) or com-bined with the intimamedia thickness of the in-ternal carotid artery and presented as a score (inone of the eight studies) had independent predic-tive power with respect to cardiovascular events.24Three studies with separate measurements for thecommon and internal carotid arteries showed sig-

nificant associations of cardiovascular events withintimamedia thickness.4,25,26 Our study confirmsthat intimamedia thickness of the common ca-rotid artery and that of the internal carotid arteryare independent predictors of cardiovascular out-comes.

It is not clear whether the intimamedia thick-ness incrementally adds value to the Framinghamrisk factors for cardiovascular-risk prediction. Theaddition of intimamedia thickness measurements

slightly increased the predictive power with respectto cardiovascular risk assessment in one study14and with respect to stroke in another study.27 Thepresence of plaque (defined as an internal-carotid-artery intimamedia thickness 1.9 mm) has beenshown to be associated with increased eventrates.28 Our data clearly show that addition of theintimamedia thickness of the internal carotid ar-tery increases the net reclassification index for riskcategories based on the Framingham risk factors.

Reclassification is a practical approach to gaug-ing the effects of adding new risk factors to thetraditional Framingham risk factors when differ-ences in the C statistic are marginal.23 A recentmeta-analysis reviewed studies suggesting that anew risk factor added predictive value to the Fram-ingham risk score.29 We performed our study ac-

cording to the criteria proposed in the meta-anal-ysis: verification of regression calibration, predictivevalue of the new risk factor in a multivariablemodel with the Framingham risk factors, positivechange in the C statistic, and an increased netreclassification index. The intimamedia thick-ness of the internal carotid artery satisfied allthese metrics, whereas the intimamedia thick-ness of the common carotid artery did not. TheAtherosclerosis Risk in Communities study (ARIC;

Table 2. Hazard Ratios for Cardiovascular Disease, According to Models with and without Common Carotid Artery(CCA) IntimaMedia Thickness.*

Risk Factor Model with Risk Factors OnlyModel with Risk Factors and CCA

IntimaMedia Thickness

Hazard Ratioor C Statistic

(95% CI) P Value


(95% CI) P Value

Sex, female vs. male 0.74 (0.580.96) 0.02 0.77 (0.590.99) 0.04

Age, per increase of 1 yr 1.05 (1.041.07)

8/3/2019 NEJMoa1012592

6/9



ClinicalTrials.gov number, NCT00005131)4 showedan increase in the area under the curve from 0.742to 0.755 and an increase in the net reclassificationindex of 9.9%15 for predicting incident coronaryheart disease with the use of the intimamediathickness of the common carotid artery in com-bination with the presence or absence of plaque,whereas the results with the use of the intimamedia thickness of the common carotid arteryalone were modest.

Our study differs from the ARIC study in sev-eral respects. The mean follow-up period was 7.2years in our study, as compared with 10 years inthe ARIC study; our study was smaller (2946 par-

ticipants, as compared with 13,145); and we ex-cluded plaques from our measurement of intimamedia thickness in the common carotid artery,whereas in the ARIC study, plaques were includ-ed.15,30 Plaque in the common carotid artery mayaccount for the weak positive associations betweenintimamedia thickness of the common carotidartery and outcomes in the ARIC study. Furtherclarification of the predictive power of risk fac-

Table 3. Hazard Ratios for Cardiovascular Disease, According to Models with and without Internal Carotid Artery (ICA) IntimaMediaThicknesses.*

Risk FactorModel with Risk Factors

Only

Model with Risk Factorsand ICA IntimaMedia

Thickness

Model with Risk Factorsand ICA IntimaMedia

Thickness >1.5 mm


(95% CI) P Value


(95% CI) P Value


(95% CI) P Value

Sex, female vs. male 0.74 (0.570.95) 0.02 0.78 (0.611.01) 0.06 0.79 (0.611.02) 0.07

Age, per increase of 1 yr 1.05 (1.041.07)

8/3/2019 NEJMoa1012592

7/9



tors and intimamedia thickness will most likelyrequire verification in other epidemiologic cohortsand attention to differences in the protocol for

measurement of the intimamedia thickness.We chose to assess the net reclassification in-dex for three clinically relevant risk categories.21,23These categories were intended to ref lect clinicalpractice, in which high risk may suggest the needfor treatment, low risk indicates the absence ofclinical symptoms of any form of cardiovasculardisease, and intermediate risk reflects the state inbetween. Similar categories are used in widely ap-

plied national guidelines for lipid-lowering inter-ventions.31 Splitting the middle category into twowould introduce the potential for additional move-

ment between categories that may not have clinicalimplications unless attention is given to specif icrisk factors.31 Even within the three Framinghamrisk categories we used, the presence of plaque,defined as an intimamedia thickness of morethan 1.5 mm, was a signif icant predictor of car-diovascular events (Fig. 1), suggesting that fur-ther analyses are needed to evaluate the effect ofplaque on risk stratification.

Fractionwithou

tCardiovascular

Dis

ease

1.0

0.8

0.9

0.7

0.6

0.5

0.00 1 2 3 4 5 6 7 8

Years

A AnyCVDRisk

Plaque

No plaque

Fractionwithou

tCardiovascular

Dis

ease

1.0

0.8

0.9

0.7

0.6

0.5

0.00 1 2 3 4 5 6 7 8

Years

B LowCVDRisk(0to20%)

Plaque

No plaque

Figure 1. KaplanMeier Estimates of the Probability of New-Onset Cardiovascular Disease (CVD).

Data are shown for all 2946 participants overall (Panel A) and according to the category of Framingham risk score

for CVD: low risk (0 to 20%) (Panel D). In thelow-risk category, 134 of the 1191 persons had plaque, for a prevalence of 11.3%, and the 8-year rates of cardiovas-

cular disease were 2.5% and 11.0% among persons without plaque and those with plaque, respectively (P

8/3/2019 NEJMoa1012592

8/9



A limitation of our study is the white race ofour population, such that our results may not beapplicable to other races or ethnic groups. How-ever, the Framingham risk factors and the riskscore have been successfully applied to variousethnic groups,32,33 and intimamedia thicknessseems to be similarly applicable.34 Another limi-

tation is our 7.2-year follow-up period, which isshorter than the 10-year period for which theFramingham risk score is calculated.35 This dis-crepancy may have decreased the overall power ofour observations. In addition, we relied on a sin-gle experienced and supervised sonographer toobtain high-quality measurements during carotid-artery ultrasonography, but this might affect theimplementation of our findings in primary preven-tion, since we believe that the sonographers judg-ment and experience affect the assessment forplaque in the internal carotid artery.19 One benefit

of using the sonographer, however, was that dataon intimamedia thickness in the internal carotidartery were missing for only 19 of our 2965 par-ticipants.

Our results may affect how intimamediathickness is assessed for the primary prevention ofcardiovascular disease. The recent American Col-lege of CardiologyFoundationAmerican HeartAssociation guidelines13 give carotid intimamediathickness a level IIa recommendation for cardio-vascular risk evaluation (the same as the recom-mendation level for the anklebrachial index andcoronary-artery calcium scoring), with an empha-sis on an indication of high risk if the common-carotid-artery intimamedia thickness is abovethe 75th percentile.10 The 75th-percentile thresh-old is also adopted in the report of the Screeningfor Heart Attack Prevention and Education TaskForce,12 and its use is currently reimbursed in onestate.36 However, these two guidelines lack quan-titative criteria for the intimamedia thickness ofthe internal carotid artery.

Our results show that plaque in the internalcarotid artery, either measured as part of the con-tinuous intimamedia thickness or assumed tobe present if the thickness exceeds a set point of1.5 mm, offers modest incremental value to theFramingham risk score in predicting cardiovas-cular events. We believe the intimamedia thick-

ness of the internal carotid artery should be mea-sured in addition to the thickness of the commoncarotid artery for purposes of cardiovascular riskassessment. One limitation, however, may be themethod used to measure the intimamedia thick-ness of the internal carotid artery. We used a con-tinuous tracing of the contour of the wall and anautomatic algorithm to determine the maximumvalue.19 The results of these offline measurementsmay differ from the results of measurements madewith the calipers available on an ultrasonograph-ic imaging device.

We conclude that the intimamedia thicknessof the common carotid artery and the intimame-dia thickness of the internal carotid artery areindependent predictors of cardiovascular eventsamong participants in the Framingham OffspringStudy. The maximum intimamedia thickness ofthe internal carotid artery, as either a continuousmeasurement or a surrogate for the presence ofplaque (above a threshold of 1.5 mm), contributedsignificantly but modestly to the predictive powerof the risk factors used in calculating the Fram-ingham risk score and improved risk classificationon the basis of the Framingham risk score.

Supported by grants from the National Heart, Lung, andBlood Institute (R01 HL069003 and HL081352, to Dr. Polak) andits Framingham Heart Study (N01-HC-25195).

Dr. Pencina reports receiving fees for board membership fromAbbott. No other potential conflict of interest relevant to thisarticle was reported.

Disclosure forms provided by the authors are available withthe ful l text of this art icle at NEJM.org.

We thank Allison Meisner for performing additional verif ica-tion of the data presented in the manuscript.

References

1. Hodis HN, Mack WJ, LaBree L, et al.Reduction in carotid arterial wall thick-ness using lovastatin and dietar y therapy:a randomized controlled clinical trial.Ann Intern Med 1996;124:548-56.2. OLeary DH, Polak JF, Kronmal RA, etal. Thickening of the carotid wall: a mark-er for atherosclerosis in the elderly? Stroke1996;27:224-31.3. OLeary DH, Polak JF, Kronmal RA,Manolio TA, Burke GL, Wolfson SK Jr.

Carotid-artery intima and media thicknessas a risk factor for myocardial infarctionand stroke in older adults. N Engl J Med1999;340:14-22.4. Chambless LE, Heiss G, Folsom AR, etal. Association of coronary heart diseaseincidence with carotid arterial wall thick-ness and major risk factors: the Athero-sclerosis Risk in Communities (ARIC)Study, 1987-1993. Am J Epidemiol 1997;146:483-94.

5. Bots ML, Hoes AW, Koudstaal PJ, Hof-man A, Grobbee DE. Common carotidintima-media thickness and risk of strokeand myocardial infarction: the RotterdamStudy. Circulation 1997;96:1432-7.6. Chambless LE, Folsom AR, Clegg LX,et al. Carotid wall thickness is predictiveof incident clinical st roke: the Atheroscle-rosis Risk in Communities (ARIC) study.Am J Epidemiol 2000;151:478-87.7. Hodis HN, Mack WJ, LaBree L, et al.




8/3/2019 NEJMoa1012592

9/9



The role of carotid arterial intima-mediathickness in predicting clinical coronaryevents. Ann Intern Med 1998;128:262-9.8. Pignoli P, Tremoli E, Poli A, Oreste P,Paoletti R. Intimal plus medial thicknessof the arterial wall: a direct measurementwith ultrasound imaging. Circulation 1986;74:1399-406.9. Dalager S, Paaske WP, Kristensen IB,

Laurberg JM, Falk E. Artery-related differ-ences in atherosclerosis expression: impli-cations for atherogenesis and dynamicsin intima-media thickness. Stroke 2007;38:2698-705.10. Stein JH, Korcarz CE, Hurst RT, et al.Use of carotid ultrasound to identify sub-clinical vascular disease and evaluate car-diovascular disease risk: a consensusstatement from the American Society ofEchocardiography Carotid Intima-MediaThickness Task Force: endorsed by theSociety for Vascular Medicine. J Am SocEchocardiogr 2008;21:93-111. [Erratum,J Am Soc Echocardiogr 2008;21:376.]11. Touboul PJ, Hennerici MG, Meairs S,

et al. Mannheim intima-media thicknessconsensus. Cerebrovasc Dis 2004;18:346-9.12. Naghavi M, Falk E, Hecht HS, et al.From vulnerable plaque to vulnerable pa-tient part III: executive summary of theScreening for Heart Attack Preventionand Education (SHAPE) Task Force re-port. Am J Cardiol 2006;98:2H-15H.13. Greenland P, Alpert JS, Beller GA, etal. 2010 ACCF/AHA guideline for assess-ment of cardiovascular risk in asymptom-atic adults: executive summary: a reportof the American College of CardiologyFoundation/American Heart AssociationTask Force on Practice Guidelines. Circu-

lation 2010;122:2748-64.14. del Sol AI, Moons KG, Hollander M, etal. Is carotid intima-media thickness use-ful in cardiovascular disease risk assess-ment? The Rotterdam Study. Stroke 2001;32:1532-8.15. Nambi V, Chambless L, Folsom AR, etal. Carotid intima-media thickness andpresence or absence of plaque improvesprediction of coronary heart disease: theARIC (Atherosclerosis Risk in Communi-ties) study. J Am Coll Cardiol 2010;55:1600-7.16. Kannel WB, Feinleib M, McNamaraPM, Garrison RJ, Castelli WP. An investi-gation of coronary heart disease in fami-

lies: the Framingham Offspring Study.Am J Epidemiol 1979;110:281-90.17. McNamara JR, Schaefer EJ. Automat-ed enzymatic standardized lipid analysesfor plasma and lipoprotein fractions. ClinChim Acta 1987;166:1-8.18. Warwick GR, Benderson J, Albers JJ.Dextran sulfate-magnesium precipitationprocedure for quantitation of high-density-

lipoprotein cholesterol. Clin Chem 1982;28:1379-88.19. Polak JF, Pencina MJ, Meisner A, et al.Associations of carotid artery intima-media thickness (IMT) with risk factorsand prevalent cardiovascular disease: com-parison of mean common carotid arteryIMT with maximum internal carotid ar-tery IMT. J Ultrasound Med 2010;29:1759-68.20. Polak JF, OLeary DH, Kronmal RA, etal. Sonographic evaluation of carotid ar-tery atherosclerosis in the elderly: rela-tionship of disease severity to stroke andtransient ischemic attack. Radiology 1993;188:363-70.

21. DAgostino RB Sr, Vasan RS, PencinaMJ, et al. General cardiovascular risk pro-file for use in primary care: the Framing-ham Heart Study. Circulation 2008;117:743-53.22. Pencina MJ, DAgostino RB. Overall Cas a measure of discrimination in survivalanalysis: model specific population valueand confidence interval estimation. StatMed 2004;23:2109-23.23. Pencina MJ, DAgostino RB Sr,DAgostino RB Jr, Vasan RS. Evaluatingthe added predictive ability of a newmarker: from area under the ROC curve toreclassification and beyond. Stat Med2008;27:157-72.

24. Lorenz MW, Markus HS, Bots ML,Rosvall M, Sitzer M. Prediction of clinicalcardiovascular events with carotid intima-media thickness: a systematic reviewand meta-analysis. Circulation 2007;115:459-67.25. Lorenz MW, von Kegler S, SteinmetzH, Markus HS, Sitzer M. Carotid intima-media thickening indicates a higher vascu-lar risk across a wide age range: prospec-tive data from the Carotid AtherosclerosisProgression Study (CAPS). Stroke 2006;37:87-92.26. Rosvall M, Janzon L, Berglund G,Engstrm G, Hedblad B. Incident coro-nary events and case fatality in relation to

common carotid intima-media thickness.J Intern Med 2005;257:430-7.27. Touboul PJ, Labreuche J, Vicaut E,Amarenco P. Carotid intima-media thick-ness, plaques, and Framingham risk scoreas independent determinants of strokerisk. Stroke 2005;36:1741-5.28. Rundek T, Arif H, Boden-Albala B, El-kind MS, Paik MC, Sacco RL. Carotid

plaque, a subclinical precursor of vascularevents: the Northern Manhattan Study.Neurology 2008;70:1200-7.29. Tzoulaki I, Liberopoulos G, Ioannidis JPA. Assessment of claims of improvedprediction beyond the Framingham riskscore. JAMA 2009;302:2345-52.30. Li R, Duncan BB, Metcalf PA, et al. B-mode-detected carotid artery plaque in ageneral population. Stroke 1994;25:2377-83.31. Executive summary of the third reportof the National Cholesterol EducationProgram (NCEP) Expert Panel on Detec-tion, Evaluation, and Treatment of HighBlood Cholesterol in Adults (Adult Treat-

ment Panel III). JAMA 2001;285:2486-97.32. DAgostino RB Sr, Grundy S, SullivanLM, Wilson P. Validation of the Framing-ham coronary heart disease predictionscores: results of a multiple ethnic groupsinvestigation. JAMA 2001;286:180-7.33. Liu J, Hong Y, DAgostino RB Sr, et al.Predictive value for the Chinese popula-tion of the Framingham CHD risk assess-ment tool compared with the ChineseMulti-Provincial Cohort Study. JAMA 2004;291:2591-9.34. Folsom AR, Kronmal RA, DetranoRC, et al. Coronary artery calcificationcompared with carotid intima-media thick-ness in the prediction of cardiovascular

disease incidence: the Multi-Ethnic Studyof Atherosclerosis (MESA). Arch Intern Med2008;168:1333-9. [Erratum, Arch InternMed 2008;168:1782.]35. Wilson PW, DAgostino RB, Levy D,Belanger AM, Silbershatz H, Kannel WB.Prediction of coronary heart disease usingrisk factor categories. Circulation 1998;97:1837-47.36. Texas State Legislature. H.B. No.1290: An act relating to health benefitplan coverage for bariatric surgery and forcertain tests for the early detection of car-diovascular disease. 2009:1-6.Copyright 2011 Massachusetts Medical Society.



nejmoa1012592

Documents