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 . 4 , N O . 9 , 2 0 1 1

© 2 0 1 1 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 I N C . D O I : 1 0 . 1 0 1 6 / j . j c m g . 2 0 1 1 . 0 6 . 0 1 3

MR Imaging of Carotid Plaque CompositionDuring Lipid-Lowering TherapyA Prospective Assessment of Effect and Time Course

Xue-Qiao Zhao, MD,* Li Dong, MD,† Tom Hatsukami, MD,‡ Binh An Phan, MD,*Baocheng Chu, MD, PHD,† Andrew Moore, BS,* Trevor Lane, BS,*Moni B. Neradilek, MS,§ Nayak Polissar, PHD,§ Duane Monick, MD,� Colin Lee, MD,¶Hunter Underhill, MD,† Chun Yuan, PHD†

Seattle and Yakima, Washington; and Boise, Idaho

O B J E C T I V E S The purpose of this study was to test the lipid depletion hypothesis and to establish

the time course of change in carotid plaque morphology and composition during lipid therapy using

high-resolution magnetic resonance imaging (MRI).

B A C K G R O U N D Lipid therapy is thought to improve plaque stability and reduce cardiovascular

events by targeting the plaque rupture risk features such as large lipid core, thin fibrous cap, and high

level of inflammatory infiltrates. However, the plaque stabilizing process during lipid therapy has not

been clearly demonstrated in humans and in vivo.

M E T H O D S Subjects with coronary or carotid artery disease, apolipoprotein B �120 mg/dl, and lipid

treatment history �1 year, were randomly assigned to atorvastatin monotherapy or to atorvastatin-

based combination therapies with appropriate placebos for 3 years. All subjects underwent high-

resolution, multicontrast bilateral carotid MRI scans at baseline and annually for 3 years. All images were

analyzed for quantification of wall area and plaque composition blinded to therapy, laboratory results,

and clinical course.

R E S U L T S After 3 years of lipid therapy, the 33 subjects with measurable lipid-rich necrotic core

(LRNC) at baseline had a significant reduction in plaque lipid content: LRNC volume decreased from 60.4 �

59.5 mm3 to 37.4 � 69.5 mm3 (p � 0.001) and %LRNC (LRNC area/wall area in the lipid-rich regions) from

14.2 � 7.0% to 7.4 � 8.2% (p � 0.001). The time course showed that %LRNC decreased by 3.2 (p � 0.001)

in the first year, by 3.0 (p � 0.005) in the second year, and by 0.91 (p � 0.2) in the third year. Changes in LRNC

volume followed the same pattern. Percent wall volume (100 � wall/outer wall, a ratio of volumes) in the

lipid-rich regions significantly decreased from 52.3 � 8.5% to 48.6 � 9.7% (p � 0.002). Slices containing LRNC

had significantly more percent wall volume reduction than those without (�4.7% vs. �1.4%, p � 0.02).

C O N C L U S I O N S Intensive lipid therapy significantly depletes carotid plaque lipid. Statistically

significant plaque lipid depletion is observed after 1 year of treatment and continues in the second

year, and precedes plaque regression. (Using Magnetic Resonance Imaging to Evaluate Carotid

Artery Plaque Composition in People Receiving Cholesterol-Lowering Medications [The CPC Study];

NCT00715273). (J Am Coll Cardiol Img 2011;4:977–86) © 2011 by the American College of

Cardiology Foundation

From the *Department of Medicine, Division of Cardiology, University of Washington, Seattle, Washington; †Department of

Radiology, Vascular Imaging Lab, University of Washington, Seattle, Washington; ‡Department of Surgery, Vascular Surgery,

hsi(dmC

m

A

A

PWV � percent wall volume

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 . 4 , N O . 9 , 2 0 1 1

S E P T E M B E R 2 0 1 1 : 9 7 7 – 8 6

Zhao et al.

MRI of Plaque Lipid Depletion

978

The hypothesis of lipid depletion for plaque

and clinical stability has been established onthe basis of observations from animal studies(1–4), correlation of histological features in

uman coronary plaque with unstable clinical epi-odes (5–9), and angiographic regression and clin-cal event reduction with lipid-lowering therapy10,11). In a case-control study of patients ran-omly allocated to 3 months of pravastatin treat-ent or placebo before carotid endarterectomy,risby et al. (12) provided histological evidence of

See page 987

the efficacy of therapy on plaque lipid content.However, in vivo verification of the lipid depletionhypothesis and establishment of the time course ofplaque regression has not been possible until re-cently (13,14).

Development of high-resolution mag-netic resonance imaging (MRI) tech-niques in recent years has made directassessment of plaque tissue compositionpossible. Numerous studies have shownthat MRI exhibits high contrast forinternal plaque features, and that com-bined information from multiple con-trast weightings is critical for distin-guishing all plaque components (15–20).With confirmation by histological stud-ies, high-resolution multicontrast MRIcan accurately assess plaque tissue con-tents (21–23).

We designed and conducted a ran-domized, double blind, and partial

placebo-controlled study entitled The CPCStudy (Carotid Plaque Composition by MRIDuring Lipid-Lowering) to test 2 hypotheses: 1)intensive lipid therapy will deplete plaque lipidcontent; and 2) low-density lipoprotein choles-terol (LDL-C) lowering plus high-density lipo-protein cholesterol (HDL-C) raising, comparedto LDL-C lowering alone, or similar HDL-Craising with further LDL-C lowering will lead tomore plaque depletion. The primary endpoint isthe carotid plaque lipid content as measured by

University of Washington, Seattle, Washington; §The Mountain-WhisHeart Center, Yakima, Washington; and ¶St. Luke’s Idaho Cardiology Aby a National Institutes of Health grant (R01 HL63895) from the NatioInc. The study medications are provided by Pfizer Inc., Abbott Laborareported they have no relationships relevant to the contents of this paper

a

tein

re

Manuscript received December 15, 2010; revised manuscript received May 2

MRI. The secondary endpoints include plaqueburden and other plaque tissue contents. Thisreport describes the study results of the firsthypothesis.

M E T H O D S

Study subjects. Details of the study design havebeen previously published (24). Briefly, participantswere recruited from the University of WashingtonMedical Center Cardiac Catheterization Labora-tory, the Yakima Heart Center Cardiac Catheter-ization Laboratory, and the St. Luke’s Idaho Car-diology Associates in Boise. Subjects who met thefollowing criteria were approached for the study:1) angiographically confirmed coronary artery dis-ease (defined as having at least 1 50% stenosis or 330% coronary lesions, or being post–myocardialinfarction, post–percutaneous coronary interven-tion, or post–coronary artery bypass graft surgery)or carotid disease (defined as having a �15%stenosis by ultrasonography); 2) apolipoprotein(Apo) B �120 mg/dl; 3) duration of lipid therapybefore enrollment �1 year; and 4) no contraindi-cation for gadolinium-based contrast-enhanced ca-rotid MRI. The study procedures and consentforms were approved by the University of Wash-ington institutional review board.

Among the 123 subjects enrolled in the study,mean age was 55 years, mean body mass index was30 kg/m2, 73% were male, 43% had a family historyof premature cardiovascular disease, 37% had aprevious myocardial infarction, 80% had clinicallyestablished coronary artery disease, 52% were hy-pertensive, 12% had diabetes mellitus, 23% werecurrent smokers, and 47% met the criteria formetabolic syndrome. The mean baseline total cho-lesterol was 238 mg/dl, triglycerides were 202mg/dl, LDL-C was 163 mg/dl, HDL-C was 41mg/dl, HDL2 was 7.4 mg/dl, ApoB was 131 mg/dl,ApoA1 was 129 mg/dl, lipoprotein(a) was 69nmol/l, fasting glucose level was 103 mg/dl, insulinlevel was 25 �U/dl, and median high-sensitivityC-reactive protein was 1.8 mg/l. The mean (� SD)lipid treatment history before enrollment was 7 � 4

onths.

Light Statistics, Seattle, Washington; �Yakimaciates, Boise, Idaho. This research is supportedHeart, Lung, and Blood Institute, and by Pfizeres, and Daiichi Sankyo, Inc. The authors havedisclose.

per-sso

naltorito

B B R E V I A T I O N S

N D A C R O N YM S

APO � apolipoprotein

CIMT � carotid intima-medi

thickness

ER � extended release

HDL-C � high-density

lipoprotein cholesterol

LDL-C � low-density lipopro

cholesterol

LRNC � lipid-rich necrotic co

MRI � magnetic resonance

imaging

6, 2011, accepted June 8, 2011.

bGec1rcr

M

crwsfsflwccemmtaw

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 . 4 , N O . 9 , 2 0 1 1

S E P T E M B E R 2 0 1 1 : 9 7 7 – 8 6

Zhao et al.

MRI of Plaque Lipid Depletion

979

Of the 123 subjects, 33 with acceptable imagequality, measurable lipid-rich necrotic core(LRNC) on baseline carotid MRI and LRNCmeasurements at year 3 were selected to test thehypothesis that intensive lipid therapy will depleteplaque lipid content.Lipid treatments and clinical follow-up. All subjectswere randomly assigned to 1 of 3 treatment groups:1) single therapy: atorvastatin (10 to 80 mg/day)alone, placebos for extended release (ER) niacin andcolesevelam; 2) double therapy: atorvastatin plusER-niacin (2 g/day), and placebo for colesevelam;3) triple therapy: atorvastatin, ER-niacin plus co-lesevelam (3.8 g/day). The treatment target forLDL-C was �80 mg/dl for the single-therapy anddouble-therapy groups and �60 mg/dl for thetriple-therapy group. The HDL-C target for the 2niacin-treated groups was to increase 10 mg/dl frombaseline.

All subjects were followed up monthly for thefirst 6 months and then bimonthly for the remain-ing 30 months of the 36-month protocol. A total of21 visits per subject was conducted for this study.At these visits, subjects underwent questioningabout side effects and symptomatic state as well as atargeted cardiovascular examination. Any adverseevents, both cardiovascular and noncardiovascular,were collected. At approximately 50% of the visits,blood samples were drawn (�50 ml). In all treat-ment arms, dietary counseling was targeted atweight reduction with goals based on the level ofobesity, with additional emphasis on the use ofmonounsaturated fats in a fat-limited (25% to 30%)diet.

As published previously (25), after 1 year of thestudy therapy, on average, ApoB was reduced by40%, 45%, and 51% among single-, double-, andtriple-therapy groups, respectively; and LDL-C waslowered by 47%, 47%, and 57%, respectively; trig-lycerides decreased by 25%, 33%, and 42%, respec-tively; and HDL-C was increased by 12%, 25%,and 29%, respectively.Carotid MRI scans. High-resolution, multicontrastilateral carotid MRI scans were performed on aE 1.5-T Signa whole body scanner using a 5.6

cho speed platform (Waukesha, Wisconsin). Thearotid MR scans were obtained at baseline, and at, 2, and 3 years. All follow-up MRI scans wereepeated using the same imaging protocol andarefully matched for scan coverage using the ca-otid bifurcation as internal landmark.

Subjects were placed in the supine position in the

RI scanner with the neck extended to bring the

arotid arteries into a more superficial locationelative to the skin. A custom designed head holderas used to minimize subject movement. Two

eparate phased-array carotid coils (26) were usedor simultaneous bilateral scan. As described in thetudy methodology (24), 2-dimensional time-of-ight, proton density, and T2-weighted and T1-eighted images were acquired. The longitudinal

overage of this set of images (10 to 12 slices) wasentered at the carotid bifurcation and covered thentire most likely diseased region. Then 20 ml (0.1mol/kg) of gadolinium contrast material was ad-inistrated intravenously though a power injector;

he T1 scan was repeated 5 to 7 min after contrastdministration. The total scan time for each subjectas 40 to 50 min.

MRI analysis for quantification of plaque size and tissuecomposition. Images were analyzed by experiencedreviewers in the Vascular Imaging Laboratory at theUniversity of Washington. A peer-review processwas conducted to reach consensus on each of thereviews. The image analyses were blinded to ther-apy, laboratory results, and clinical course. Oneexperienced reviewer (B.A.P.), blinded to treatmentand time points, matched the axial images fromdifferent time points according to their distancefrom the bifurcation of the carotid artery (carotidbifurcation was always used as the landmark fororientation and coverage). The results were peerreviewed by other experienced reviewers (H.U. andB.C.) who were also blinded to therapy and timepoints. All reviewers separately interpreted eachimage, and the reviewers conferred to reach aconsensus opinion.

Total vessel area was quantified by placing con-tours around the outer wall boundaries of thecarotid artery using CASCADE (27), a custom-designed image analysis tool. Similarly, lumen areawas quantified by placing contours around thelumen boundary, and wall area was calculated bysubtracting the lumen area from the total vesselarea. Lumen, wall, and total vessel volumes werederived by adding the areas across slices and mul-tiplying by the sum by the slice thickness (2 mm).Percent wall volume (PWV) was calculated usingthe following formula: (wall volume/total vesselvolume) � 100%. The PWV provides a measure ofplaque burden that adjusts for variation in arterysize, and is similar to percent atheroma volumedescribed in the intravascular ultrasound literature.

Carotid plaque tissue composition (LRNC, fi-brous tissue, loose matrix, and calcification) was

identified and quantified using previously published

uaspastbtCWTddQccd

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 . 4 , N O . 9 , 2 0 1 1

S E P T E M B E R 2 0 1 1 : 9 7 7 – 8 6

Zhao et al.

MRI of Plaque Lipid Depletion

980

MRI criteria (19,23). Specifically, LRNC was iden-tified using multicontrast weightings plus post-contrast T1-weighted images. LRNC usually appearsisointense to hyperintense on the time-of-flight andpre-contrast T1-weighted images and has variedsignal intensity on protein density weighted andT2-weighted images, has no or slight contrastenhancement compared with the surrounding tissueon post-contrast T1-weighted images. Plaque com-ponent volumes were calculated as described above,and the proportional volume of each componentwas calculated using the formula: (component vol-ume/wall volume) � 100%.

The plaque LRNC, as the primary end point ofthe study, was assessed in 2 different ways: 1) totalLRNC volume; and 2) percent LRNC (%LRNC)relative to the lipid-containing slices only. As illus-trated in Figure 1, the total LRNC volume wasmeasured by multiplying the slice thickness by thesum of the LRNC areas (circled in yellow) on the 8consecutive images from common carotid throughbifurcation to the internal carotid arteries. The

Figure 1. Example of Carotid Artery in Multicontrast MRI

Magnetic resonance imaging (MRI) example of 8 consecutive imageinternal carotid arteries (ICA) from 5 contrast weightings: time of fliT1-weighted (T1W), and post-contrast T1W. The lumen in red and ooutlined. Plaque tissue components within the arterial wall are idenpurple and lipid content in yellow). For example, lipid content in yand hypointense on PDW and T2W images, and has no contrast enT1W images. Total lipid-rich necrotic core (LRNC) volume is calculatcircled in yellow color on the 8 consecutive images. Percent LRNC

the lipid-containing slices, from slice #2 to #7.

%LRNC was calculated based only on the slicescontaining a LRNC (slices #2 to #7 in this exam-ple).Statistical analysis. In all analyses, the artery wassed as the unit of observation. That is, in a givennalysis, slice values of the outcome within a givenubjects were pooled (as described previously) toroduce 1 value per subject. Data from 1 carotidrtery (the index artery with more lesion) perubject were used in the analysis. Descriptive sta-istics are presented as mean � SD. Changesetween carotid scans were annualized because ofhe variation in time between visits across subjects.hanges over time were compared to zero using theilcoxon signed-rank test or the 1-sample t test.

he t test was used for carotid wall measurementata, which were reasonably similar to the normalistribution. Normality of outcomes was assessed by-Q plots. The Wilcoxon signed-rank test was

arried out for comparisons involving plaque tissueomponents, which sometimes had non-normalistributions. Change in %LRNC and PWV were

the common carotid artery (CCA) through bifurcation to the(TOF), proton density weighted (PDW), T2-weighted (T2W),r wall boundary in blue of the carotid artery are identified andd using an established algorithm and quantified (loose matrix inw appears isointense on the TOF and pre-contrast T1W imagescement compared with the surrounding tissue on post-contrasty multiplying the slice thickness (2 mm) by the sum of the areasRNC) is the proportion of the wall occupied by the LRNC only in

s ofghtutetifieellohaned b(%L

sm3qjAb33

vo08eL

Lt3

ajda

pidbitl(

ar

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 . 4 , N O . 9 , 2 0 1 1

S E P T E M B E R 2 0 1 1 : 9 7 7 – 8 6

Zhao et al.

MRI of Plaque Lipid Depletion

981

correlated with changes with lipid variables usingSpearman correlation. Calculations were carried outin R, version 2.9.0 (28) All p values �0.05 wereconsidered statistically significant.

R E S U L T S

Change in plaque lipid content. Among the 123ubjects enrolled in this study, 33 subjects witheasureable LRNC at baseline who completed

-year follow-up scans and had acceptable imageuality were included in this comparison. Six sub-

ects were excluded because of poor image quality.nother 81 subjects had no measurable LRNC ataseline and were excluded from the analysis. Finally,subjects with LRNC at baseline did not complete

-year follow-up scans and were excluded, too.After 3 years of intensive lipid therapy, LRNC

olume significantly decreased from a mean � SDf 60.4 � 59.5 mm3 to 37.4 � 69.5 mm3 (p �.001) and %LRNC from 14.2 � 7.0% to 7.4 �.2% (p � 0.001), as summarized in Figure 2. Anxample of a carotid plaque with reduction inRNC size is shown in Figure 3A.There were 5 subjects who had new measurable

RNC at some point during the 3 years. Only 1 ofhese 5 subjects had measurable LRNC at the-year scan. Among the 75 study subjects with all 4

% L

RN

C

Baseline Year 1 Yea

p<0.001*

p

* Compared to Baseline

§ Year 2 compared to year 10

5

10

15

Figure 2. Carotid Plaque Lipid Depletion and Time Course Duri

After 3 years of intensive lipid therapy, lipid-rich necrotic core voluto 37.4 mm3. Percent LRNC (%LRNC) (pink circles) also significantlycourse over 3 years showed that %LRNC for pooled slices containinyear, significantly decreased by 3.0% in the second year, and decre

statistically significant). Bars around the estimates are standard error ba

nnual measurements, the overall number of sub-ects with measurable LRNC was significantly re-uced over the 3 years: 44% at baseline versus 33%t 3 years (p � 0.03, McNemar test).Plaque lipid depletion time course. The year-by-year

laque lipid depletion time course is also illustratedn Figure 2. The mean LRNC volume significantlyecreased by 12 mm3 (p � 0.007) in the first year,y 13 mm3 (p � 0.004) in the second year, andncreased nonsignificantly by 0.4 mm3 (p � 0.3) inhe third year. Similarly, mean %LRNC amongipid-containing slices significantly decreased by 3.2p � 0.001) in the first year, by 3.0 (p � 0.005) in

the second year, and reduced nonsignificantly by 0.9(p � 0.2) in the third year. Figure 3B demonstratesn MRI example of plaque lipid depletion andegression in plaque volume over 3 years.Changes in other plaque tissue components. Figure 4shows plaque fibrous tissue, calcium, and loosematrix changes among lipid-containing slices over 3years. Compared to baseline, the volume of fibroustissue significantly decreased (571 mm3 vs. 547mm3, p � 0.007), but percent of plaque composedof fibrous tissue significantly increased (83.4% vs.90.1%, p � 0.001) during the 3 years. The meanvolume of fibrous tissue decreased slightly in thefirst year and significantly in the second year. Mean

Year 3

LR

NC

-V (

mm

3 )05§

0

10

20

30

40

50

60

70

80

LRNC-V

% LRNC

-Year Lipid Therapy

LRNC-V) (green squares) significantly decreased from 60.4 mm3

reased from 14.2% to 7.4%. The plaque lipid depletion timeid at any time point significantly decreased by 3.2% in the firstby 0.9% in the third year (the change from year 2 to 3 was not

r 2

=0.0

ng 3

me (decg lipased

rs.

warLyanuiy

y0prtc

scTsisvta

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 . 4 , N O . 9 , 2 0 1 1

S E P T E M B E R 2 0 1 1 : 9 7 7 – 8 6

Zhao et al.

MRI of Plaque Lipid Depletion

982

percent fibrous tissue increased significantly duringeach of the first 2 years. Both volume and percent offibrous tissue did not change significantly in thethird year. Calcium and loose matrix, both involume and percent composition, remained at asimilar level over the 3 years.Carotid arterial wall burden change and its timecourse. Three-year and annual changes in total

all volume, PWV, and lumen volume weressessed to describe and test carotid atheroscle-otic burden change in subjects with measurableRNC at baseline. As shown in Table 1, after 3ears of intensive lipid therapy, total wall volumend PWV decreased significantly, but there waso statistically significant change in lumen vol-me. Time course showed no statistically signif-cant change in wall volume or PWV in the first

Figure 3. MRI Examples of Plaque Lipid Depletion and Time Co

(A) An example of significant lipid content reduction (yellow arrowcarotid artery. Overall, 11% of study subjects had completed plaqueexample of the plaque lipid depletion time course. Regression in lip1-year, and 2-year MRI scans. Regression in plaque volume seemedyears 1 to 3. CE � contrast enhanced; T1W � T1-weighted.

ear, but significant changes during the second s

ear, as wall volume decreased by 42 mm3 (p �.001 for 2-year volume compared to baseline;� 0.01 compared to year 1) and PWV was

educed by 2.1% (p � 0.001 for 2-year comparedo baseline, p � 0.01 compared to year 1). Thehanges in the third year were minimal.Difference in wall volume change between slices withand without LRNC. As also shown in Table 1, thelices with LRNC showed a significant mean de-rease in total wall volume and PWV over 3 years.he reduction in wall burden measurements in

lices without LRNC showed the same direction asn slices with LRNC but was smaller and nottatistically significant. Furthermore, both for wallolume and PWV, the 2 measures of wall burden,he differences in reduction between the slices withnd without LRNC at baseline were statistically

nd plaque regression at 3 years compared to baseline in the leftid depletion over 3 years. (B) Magnetic resonance imaging (MRI)ich necrotic core (LRNC) size was notable between the baseline,ollow plaque lipid depletion and was most pronounced from

urse

s) alipid-rto f

ignificant.

msavmsfcptH

rror

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 . 4 , N O . 9 , 2 0 1 1

S E P T E M B E R 2 0 1 1 : 9 7 7 – 8 6

Zhao et al.

MRI of Plaque Lipid Depletion

983

Relation between plaque and lipid changes. Spear-an correlation analysis showed no statistically

ignificant association between change in %LRNCnd PWV as assessed by MRI and change in lipidariables over 3 years among the 33 subjects witheasurable LRNC at baseline in this report. Ob-

erved Spearman correlations ranged in magnituderom 0.00 to 0.31. The lipid variables includehanges in total cholesterol, very low density lipo-rotein cholesterol, intermediate-density lipopro-ein cholesterol, LDL-C, HDL-C, HDL-C2,

DL-C3, ApoB, ApoA1, and triglycerides.

D I S C U S S I O N

The most widely accepted hypothesis is that lipid-lowering therapy targets the plaque rupture riskfeatures such as large lipid core, thin fibrous cap,and high level of inflammatory infiltrates and activ-ity; therefore, the plaque stability is improved andcardiovascular events are reduced. The evidencefrom diet-induced atherosclerosis animal models

% C

A

Baseline Year

1.0

1.5

2.0

% F

T

Baseline Year 1 Year 2 Year 3

FT-

V (

mm

3 )

80

85

90

95 600

580

560

540

520

500

FT-V% FT

Figure 4. Carotid Plaque Tissue Composition Change Over 3-Ye

Changes in plaque fibrous tissue (FT), calcium (CA), and loose matrcles), over 3 years. CA � calcium; FT � fibrous tissue; LM � loosepercent of fibrous tissue increased. These changes were significantsignificantly over 3 years. Bars around the estimates are standard e

Table 1. Carotid Wall Measurements Change Over 3 Years and

Comparison: Baseline vs. 3 Yrs

n Baseline 3 Yrs p Value‡

Wall volume, mm3 33 641 � 200 597 � 198 �0.001

Percent wall volume, % 33 46.5 � 7.1 44.3 � 7.2 0.006

Lumen volume, mm3 33 750 � 249 764 � 258 0.110

Values are mean � SD for the comparison of baseline versus 3 years and mean �times between the visits for different subjects. The calculation of changes wasversus 3 years. †The smaller sample size (n � 31) was the results of 2 patients

0.001 when compared to null hypothesis of zero change. �p � 0.01 when compare

(1–4) showed that lowering cholesterol can de-crease foam cells and cholesterol ester content at 6months, diminish cholesterol crystals and necrosisat 12 months, and reduce plaque size at 24 to 48months. The CPC study, 1 of the first studiestesting the lipid depletion hypothesis in humansand in vivo, demonstrated that intensive lipid ther-apy significantly depletes plaque lipid content over 3years. We found a 6.8% absolute reduction inlipid-containing segment (14.2% at baseline vs.7.4% at 3 years) with intensive lipid therapy and11% lower frequency of subjects with measurableLRNC (44% at baseline vs. 33% at 3 years). Thesereductions were statistically significant.

Histological studies of human coronary plaqueshave demonstrated an association between featuressuch as the presence of a large lipid core, thinfibrous cap, and high concentration of inflamma-tory infiltrates with prior unstable clinical events(5–9). In the carotid artery, MRI has provided anessential tool for identifying potential high-risk

% L

M

Baseline Year 1 Year 2 Year 3

LM

-V (

mm

3 )

0.5

1.0

1.5

7

6

4

5

2

1

3

LM-V% LM

Year 2 Year 3

CA

-V (

mm

3 )

12

10

8

6

4

-VCA

ipid Therapy

M), both volume (V) (green squares) and composition (pink cir-ix. Compared to baseline, volume of fibrous tissue decreased, butach of the first 2 years. Calcium and loose matrix did not changebars.

erential Changes Between Slices With and Without LRNC

Annual Changes*Comparison: Chang

LRNC (�) vs

n 1st Yr 2nd Yr 3rd Yr n LRNC (�) L

30 �16 � 62 �42 � 57§� 3 � 34 31 �45 � 55¶ �

30 �0.6 � 2.9 �2.1 � 3.2§� �0.2 � 2.5 31 �4.7 � 7.1¶ �

30 2 � 50 18 � 52 5 � 42 31 12 � 41

or each annual change during the 3 years. *Changes between the scans were annued to patients with all 4 annual visits. Thus, the sample size is smaller than for thhaving both lipid-rich necrotic core (LRNC) positive (�) and LRNC negative (�) s

1

CA%

ar L

ix (Lmatrat e

Diff

e Over 3 Yrs Slices. LRNC (�)†

RNC (�) p Value‡

11 � 32 0.004

1.4 � 3.2 0.020

7 � 40 0.300

SE f alized because of differentlimit e comparison of baselinenot lices. ‡Paired t test. §p �

d to year 1. ¶p � 0.001 when compared to zero change.

att

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 . 4 , N O . 9 , 2 0 1 1

S E P T E M B E R 2 0 1 1 : 9 7 7 – 8 6

Zhao et al.

MRI of Plaque Lipid Depletion

984

plaque features associated with more rapid progres-sion and future ischemic events. In a recent pro-spective natural history study of carotid atheroscle-rosis, Underhill et al. (29) showed that carotidplaques with larger LRNC at baseline were associ-ated with a significantly higher risk of plaquesurface disruption on 3-year follow-up carotidMRI. Furthermore, in a prospective study of sub-jects with initially asymptomatic 50% to 79% ca-rotid stenosis, Takaya et al. (30) found that plaqueswith larger LRNC were associated with a higherrisk of cerebrovascular events during a meanfollow-up of 3 years (hazard ratio for every 10%increase in LRNC size: 1.6, p � 0.004). The plaquelipid depletion demonstrated in this study suggest apotential mechanism by which aggressive lipid ther-apy results in plaque stabilization and a lower riskfor future ischemic events, and provides compellingevidence to justify larger prospective clinical trials tothis hypothesis.

This study also demonstrated that significantplaque lipid depletion can be achieved after 1 yearof intensive treatment, continues with a similar rateof change during the second year, and begins toplateau during the third year. The time course forplaque lipid depletion demonstrated in this studyprovides a vascular biological explanation for the onsetand persistence of clinical benefit seen in theplacebo-controlled lipid-lowering trials (10). Thecardiovascular event reduction in these trials beganat 1 to 2 years after initiation of lipid-loweringtherapy, which corresponds with the timing ofsignificant plaque lipid depletion and increase inpercent fibrous tissue at 1 and 2 years observed inthis study. The temporal relationship between theplaque biological changes and clinical event reduc-tion during lipid therapy will be further examined inour on-going studies.

In addition to reduction in lipid core volume,regression in overall plaque burden was observed,primarily after 2 to 3 years of intensive lipid therapy,consistent with the observations from animal stud-ies (1–4), and consistent with plaque burden regres-sion seen in other lipid-lowering studies in humans(13,31,32). We believe that the regression observedin plaque burden was primarily induced by lipiddepletion because:1) substantial and statistically sig-nificant carotid wall burden reduction only occurredin the slices containing a LRNC; and 2) the timecourse of wall burden reduction seemed to followplaque lipid depletion, with the largest changes

occurred during the second year. a

The temporal relationship between plaque lipidand plaque burden regression, and the observedregression only at sites with a LRNC may also helpto explain the inconsistent results seen in studiesusing carotid intima-media thickness (CIMT) as amarker for regression during LDL-C–loweringtherapy. In the METEOR (Effect of Rosuvastatinon Progression of Carotid Intima-Media Thicknessin Low-Risk Individuals With Subclinical Athero-sclerosis) study (33), where relatively statin-naïvesubjects with thick CIMTs were recruited (baselineCIMT of 1.16 mm, and hence more likely to havelipid-rich plaques), statistically significant differ-ences between treatment groups were observed(0.014 mm/year). In contrast, the ENHANCE(Simvastatin With or Without Ezetimibe in Famil-ial Hypercholesterolemia) study (34), where subjectshad been treated for an extended period of time, andhad a relative normal CIMT at baseline (0.68 mm andmore likely to have lipid-poor plaques) showed a verysmall difference (0.005 mm per 2 years) betweengroups. Once atherosclerotic plaque lipid has beendepleted and regression has achieved with therapy,further treatment is certainly necessary to keep plaquestable without significant progression, but maybe un-likely to induce further regression.

Although there are no statistically significant cor-relations between the plaque changes, in both%LRNC and PWV, and the lipid variables, this doesnot indicate there is no differential treatment effectamong the 3 different lipid therapies, because only 33subjects were included in the current correlation anal-ysis. The differential treatment effect will be investi-gated fully when the CPC study is closed, and it willbe further examined in the carotid MRI substudy inthe AIM-HIGH (Atherothrombosis Intervention inMetabolic syndrome with low HDL/high triglycer-ides: Impact on Global Health outcomes) trial inapproximately 200 subjects (35).

The effects of the 3 treatment therapies may haveintroduced some heterogeneity into the responses.This increase in variability, if it occurred, would tendto decrease power to detect changes and identifycorrelates of change. Thus, statements of statisticalsignificance in this paper may be conservative. It isunlikely that statistically significant differences notedhere are artifacts. Statistically significant differencesand associations observed in this study—with a samplesize of �40 subjects—can be attributed to severalspects of the study, including: 1) a relativelyreatment-naive population; 2) potent lipid modifica-ion on both LDL-C and HDL-C, and; 3) highly

ccurate assessment of plaque change by MRI.

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 . 4 , N O . 9 , 2 0 1 1

S E P T E M B E R 2 0 1 1 : 9 7 7 – 8 6

Zhao et al.

MRI of Plaque Lipid Depletion

985

C O N C L U S I O N S

In conclusion, intensive lipid therapy significantlydepletes carotid atherosclerotic plaque lipid inhumans. Statistically significant plaque lipid de-pletion is observed after 1 year of treatment andcontinues in the second year. Regression in over-all plaque burden was observed primarily at loca-tions with a LRNC, and its time course followsplaque lipid depletion. These findings suggest a

Macrophages and atherosclerotic

1

1

1

1

1

1

1

1

MR imaging for id

sociated with intensive lipid therapy. Further-more, these results highlight the importance ofselection of subjects with a LRNC in future trialsexamining the effect of therapy on plaque regres-sion.

Reprint requests and correspondence: Dr. Xue-Qiao Zhao,University of Washington, Cardiology, Box 354800,1107 Northeast 45th Street, Suite 114, Seattle, Wash-

potential mechanism for plaque stabilization as- ington 98105. E-mail: xueqiao@u.washington.edu.

1

1

2

2

2

2

2

R E F E R E N C E S

1. Armstrong ML, Megan MB. Lipiddepletion in atheromatous coronaryarteries in rhesus monkeys after re-gression diets. Circulation Res 1972;30:675–80.

2. Armstrong MC, Megan MB. Arterialfibrous protein in cynomolygous mon-keys after atherogenic and regressiondiets. Circulation Res 1975;36:256–61.

3. Clarkson TB, Bond MG, BullockBC, Marzetta CA. A study of athero-sclerosis regression in Macaca mulatta.IV. Changes in coronary arteries fromanimals with atherosclerosis inducedfor 19 months and then regressed for24 or 48 months at plasma cholesterolconcentrations of 300 or 200 mg/dl.Exp Mol Pathol 1981;34:345–68.

4. Small DM, Bond MG, Waugh D,Prack M, Sawyer JK. Physiochemicaland histological changes in the arterialwall of nonhuman primates during pro-gression and regression of atherosclero-sis. J Clin Invest 1984;73:1590–605.

5. Davies MJ. Stability and instability:two faces of coronary atherosclerosis.The Paul Dudley White Lecture1995. Circulation 1996;94:2013–20.

6. Van der Wal AC, Becker AE, Van derLoos CM, Das PK. Site of intimal rup-ture or erosion of thrombosed coronaryatherosclerotic plaques is characterized byan inflammatory process irrespective ofthe dominant plaque morphology. Circu-lation 1994;89:36–44.

7. Moreno PR, Falk E, Palacios IF,Newell JB, Fuster V, Fallon JT. Mac-rophage infiltration in acute coronarysyndromes: implications for plaquerupture. Circulation 1994;90:775–8.

8. Davies MJ, Richardson PD, Woolf N,Katz DR, Mann J. Risk of thrombosisin human atherosclerotic plaques: roleof extracellular lipid, macrophages,and smooth muscle cell content. BrHeart J 1993;69:377–81.

9. Libby P, Geng YJ, Aikawa M, et al.

plaque stability. Curr Opin Lipidol1996;7:330–5.

0. Brown BG, Zhao X-Q. Lipid-Lowering Therapy. In: HennekensCH, editor. Clinical Trials in Cardio-vascular Disease (A Companion toBraunwald’s Heart Disease). Philadel-phia, PA: WB Saunders, 1999:199–216.

1. Brown BG, Zhao XQ, Chait A, et al.Simvastatin and niacin, antioxidantvitamins, or the combination for pre-vention of coronary disease. N EnglJ Med 2001;345:1583–92.

2. Crisby M, Nordin-Fredriksson G, ShahPK, Yano J, Zhu J, Nilsson J. Pravasta-tin treatment increases collagen contentand decreases lipid content, inflamma-tion, metalloproteinases, and cell deathin human carotid plaques: implicationsfor plaque stabilization. Circulation2001;103:926–33.

3. Zhao XQ, Chun Y, Hatsukami TS, etal. Effects of intensive lipid-loweringtherapy on carotid atheroscleroticplaque composition in vivo by MRI: acase-control study. ArteriosclerThromb Vasc Biol 2001;21:1623–9.

4. Underhill HR, Yuan C, Zhao XQ, etal. Effect of rosuvastatin therapy oncarotid plaque morphology and com-position in moderately hypercholes-terolemic patients: a high-resolutionmagnetic resonance imaging trial. AmHeart J 2008; 155:584.e1–8.

5. Shinnar M, Fallon JT, Wehrli S, et al.The diagnostic accuracy of ex vivoMRI for human atherosclerotic plaquecharacterization. Arterioscler ThrombVasc Biol 1999;19:2756–61.

6. Yuan C, Mitsumori LM, FergusonMS, et al. In vivo accuracy of multi-spectral magnetic resonance imagingfor identifying lipid-rich necroticcores and intraplaque hemorrhage inadvanced human carotid plaques. Cir-culation 2001;104:2051–6.

7. Mitsumori LM, Hatsukami TS, Fer-guson MS, Kerwin WS, Cai J, YuanC. In vivo accuracy of multisequence

entifying unstable

fibrous caps in advanced human ca-rotid plaques. J Magn Reson Imaging2003;17:410–20.

8. Cappendijk VC, Cleutjens KB, Kes-sels AG, et al. Assessment of humanatherosclerotic carotid plaque compo-nents with multisequence MR imag-ing: initial experience. Radiology2005;234:487–92.

9. Saam T, Cai JM, Cai YQ, et al.Carotid plaque composition differsbetween ethno-racial groups: an MRIpilot study comparing mainland Chi-nese and American Caucasian pa-tients. Arterioscler Thromb Vasc Biol2005;25:611–6.

0. Clarke SE, Hammond RR, MitchellJR, Rutt BK. Quantitative assessmentof carotid plaque composition usingmulticontrast MRI and registered his-tology. Magn Reson Med 2003;50:1199–208.

1. Cai JM, Hatsukami TS, FergusonMS, Small R, Polissar NL, YuanC. Classification of human carotidatherosclerotic lesions with in vivomulticontrast magnetic resonanceimaging. Circulation 2002;106:1368 –73.

2. Takaya N, Cai J, Ferguson MS, et al.Intra- and interreader reproducibilityof magnetic resonance imaging forquantifying the lipid-rich necroticcore is improved with gadoliniumcontrast enhancement. J Magn ResonImaging 2006;24:203–10.

3. Cai JM, Hatsukami TS, FergusonMS, et al. In vivo quantitative mea-surement of intact fibrous cap andlipid rich necrotic core size in athero-sclerotic carotid plaque: a comparisonof high resolution contrast enhancedMRI and histology. Circulation 2005;112:3437–44.

4. Zhao XQ, Phan BA, Chu B, et al.Testing the hypothesis of atheroscle-rotic plaque lipid depletion duringlipid therapy by magnetic resonanceimaging: study design of CarotidPlaque Composition Study. Am

Heart J 2007;154:239–46.

2

2

2

2

3

3

3

c

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 . 4 , N O . 9 , 2 0 1 1

S E P T E M B E R 2 0 1 1 : 9 7 7 – 8 6

Zhao et al.

MRI of Plaque Lipid Depletion

986

25. Moore AB, Phan BA, ChallenderCM, Williamson J, Marcovina S,Zhao XQ. Effects of adding extended-release niacin and colesevelam to sta-tin therapy on lipid levels in subjectswith atherosclerotic disease. J ClinLipidol 2007;1:620–5.

6. Hayes CE, Mathis CM, Yuan C. Sur-face coil phased arrays for high resolu-tion imaging of the carotid arteries. JMagn Reson Imaging 1996;1:109–12.

7. Kerwin W, Xu D, Liu F, et al. Mag-netic resonance imaging of carotidatherosclerosis: plaque analysis. TopMagn Reson Imaging 2007;18:371–8.

8. R: A Language and Environment forStatistical Computing. Version 2.9.0.Vienna, Austria: The R Foundationfor Statistical Computing, 2009.

9. Underhill HR, Yuan C, Yarnykh VL,et al. Predictors of surface disruption

with MR imaging in asymptomatic

carotid artery stenosis. AJNR Am JNeuroradiol 2009;31:487–93.

0. Takaya N, Yuan C, Chu B, et al.Association between carotid plaquecharacteristics and subsequent isch-emic cerebrovascular events: a pro-spective assessment with MRI—initialresults. Stroke 2006;37:818–23.

1. Corti R, Fuster V, Fayad ZA, et al.Lipid lowering by simvastatin inducesregression of human atheroscleroticlesions: two years’ follow-up by high-resolution noninvasive magnetic reso-nance imaging. Circulation 2002;106:2884–7.

2. Corti R, Fuster V, Fayad ZA, et al.Effects of aggressive versus conventionallipid-lowering therapy by simvastatin onhuman atherosclerotic lesions: a pro-spective, randomized, double-blind trialwith high-resolution magnetic reso-

nance imaging. J Am Coll Cardiol2005;46:106–12. m

33. Crouse JR III, Raichlen JS, Riley WA,et al., for the METEOR Study Group.Effect of rosuvastatin on progression ofcarotid intima-media thickness in low-risk individuals with subclinical athero-sclerosis: the METEOR trial. JAMA2007;297:1344–53.

34. Kastelein JJ, Akdim F, Stroes ES, etal., for the ENHANCE Investigators.Simvastatin with or without ezetimibein familial hypercholesterolemia.N Engl J Med 2008;358:1431–43.

35. Carotid Plaque Characteristics by MRI inAIM-HIGH. Available at: http://clinicaltrials.gov/show/NCT01178320.Accessed August 3, 2011.

Key Words: atherosclerosis yarotid artery y lipids y

agnetic resonance imaging.