EfMID

clLlsn

vwtr1brssm

D(5csapheusdtpsta6lms

A

ASca

©P

AD

ULT

CA

RD

IAC

fficacy of Aggressive Lipid Controlling Therapyor Preventing Saphenous Vein Graft Diseaseitsumasa Hata, MD, PhD, Tadateru Takayama, MD, Akira Sezai, MD,

samu Yoshitake, MD, Atsushi Hirayama, MD, and Kazutomo Minami, MD

epartments of Cardiovascular Surgery, and Cardiology, Nihon University School of Medicine, Tokyo, Japan

ms0g1ptnw

adlCl

Background. We assessed the efficacy of aggressive lipidontrolling therapy (ALCT), which maintains low-densityipoprotein cholesterol (LDL-C) below to 80 mg/dL andDL/high-density lipoprotein cholesterol (HDL-C) ratio

ess than 1.5 for preventing postcoronary bypass (CABG)aphenous vein graft (SVG) diseases by using intracoro-ary angioscopy.Methods. Twenty-one patients after CABG were di-

ided into two groups: group I consisted of 10 patientshose serum LDL-C level and LDL/HDL could be con-

rolled less than 80 mg/dL and 1.5, respectively, byosuvastatin for about one year; group II consisted of1 patients whose LDL-C level and LDL/HDL haveeen higher than 100 mg/dL and 2.5, respectively,egardless of having medication of pravastatin. Twenty-even SVGs were assessed by intravascular ultra-ound (IVUS) and angioscopy on postoperative 12 to 16

onths.

ttt

cdhstpNaainca

P

Ttsootd

hi Itabashi-ku, Tokyo, 173-8610, Japan; e-mail: mihata@med.nihon-u.c.jp.

2009 by The Society of Thoracic Surgeonsublished by Elsevier Inc

Results. The serum LDL-C level (I: 64.1 vs II: 130.2g/dL) and LDL/HDL (I: 1.36 vs II: 2.64), and high

ensitive C-reactive protein (I: 0.045 � 0.100 vs II: 0.116 �.020 mg/dL) were significantly lower in group I. Inroup II, IVUS detected eccentric plaques in 11 (78.6%) of4 SVGs. Furthermore the angioscope showed yellowlaque in all 14 SVGs (100%) and 11 (78.6%) of them had

hrombi. On the other hand, in group I, all 13 SVGs hado eccentric, yellow plaques or thrombi and the intimaas entirely clear white.Conclusions. Prophylactic treatment for yellow plaque

nd thrombus formation are extremely important in theevelopment of early and late SVG disease. Aggressive

ipid controlling therapy is quite attractive to avoid postABG SVG disease and may be effective to maintain the

ong-term graft patency.(Ann Thorac Surg 2009;88:1440–4)

© 2009 by The Society of Thoracic Surgeons

espite the advances in the relevant technologies andtechniques, nearly 25% of saphenous vein grafts

SVG) become occluded within 1 year after surgery, and0% of SVGs fail within 10 years [1, 2]. Gansera andolleagues [3] reported that the incidence of SVG occlu-ion was twofold higher than that of the internal thoracicrtery in symptomatic patients [3]. Even though theatient has no coronary events, more than 20% of SVGave been occluded for 10 years after surgery [4]. How-ver, the ready availability of SVGs still accounts for theirse in over 70% of coronary artery bypass graft (CABG)urgery, particularly for the emergency or multiple vesselisease patients. Recently, several investigators reported

hat aggressive risk factor control, together with appro-riate secondary preventive medication, has shown tolow progression of SVG atherosclerosis and the long-erm patency of SVG was not very different than otherrterial conduits except for the internal thoracic artery [5,]. Furthermore, Souza and colleagues [7] showed theong-term SVG patency was 90% by harvesting with

inimum trauma. Because of the need of the cardiacurgeon to perform complete revascularization for mul-

ccepted for publication June 1, 2009.

ddress correspondence to Dr Hata, Department of Cardiovascularurgery, Nihon University School of Medicine, 30-1 Ooyaguchi Kamima-

ivessel diseases, it is increasingly important to maximizehe long-term SVG patency in order to optimize long-erm surgical outcome.

Plaque rupture with thrombus formation is the majorause of long-term SVG disease [8, 9]. Lowering of lowensity lipoprotein cholesterol (LDL-C) below 100 mg/dLas been shown to be effective for reducing the progres-ion of atherosclerosis in SVG [10]. High density lipopro-ein cholesterol (HDL-C) has also been associated with arotective effect on atherosclerosis in carotid arteries [11].o studies to date have investigated the efficacy of

ggressive statin therapy on angioscopic progressionfter CABG, SVG disease. The aim of this study was todentify the morphologic changes in SVG by intracoro-ary ultrasound and angioscopy, and to assess the effi-acy of aggressive lipid controlling therapy with rosuv-statin versus moderate therapy with pravastatin.

atients and Methods

wenty-one patients who have been followed-up morehan 12 months after CABG were angiographically as-essed. All procedures were carried out with traditionaln-pump CABG using the left internal thoracic artery,ne radial artery, and SVGs. In CABG, we usually usedhe left internal thoracic artery for the left anterior

escending artery or big diagonal branch, and one radial

0003-4975/09/$36.00doi:10.1016/j.athoracsur.2009.06.009

aotdmstndmusIpmfrat1wb(esipSb

1brwcLsasJtw

nswrtgaeaar

SRcNst0

F

T

V

AMHDSNT

CR

T

V

SLHLTHh

a

Llp

1441Ann Thorac Surg HATA ET AL2009;88:1440–4 AGGRESSIVE LIPID CONTROL FOR SVG DISEASE

AD

ULT

CA

RD

IAC

rtery for the proximal right coronary artery or firstbtuse marginal branch of circumflex territory becausehe radial artery length is limited. The SVG is used for theistal part of the right coronary artery, second obtusearginal branch, or distal part of the left anterior de-

cending artery. The SVG was harvested by standardechnique with scissors and clips. An endoscope wasever used for graft harvesting. Furthermore, we tried toistend the harvested SVG with just a little pressure andinimize the intimal damage of SVG. All the patients

nderwent CABG during 2007 in our Institution. Thistudy was a case-controlled observational study andnstitutional Review Board approval was provided beforeublication of this manuscript and reporting of the infor-ation. The matched criteria between the groups were as

ollows; noninsulin user, age less than 80 years old,eceiving both aspirin 100 mg and ticlopidine 200 mg daily,nd receiving some calcium antagonists. Seventeen pa-ients (81.0%) were male, and the average age was 65.6 �0.4 years, ranging from 31 to 78. Written informed consentas given to all patients to receive the graft assessmenty coronary angiography, intravascular ultrasound

IVUS), and angioscopy. The coronary angioscopic cath-ter (Fibertech Co, Tokyo, Japan) used in the presenttudy is 0.73 mm in outer diameter with a fiber contain-ng 6,000 pixels. After the SVG angiography, we carefullyut the angioscope into the distal anastomosis site ofVG. We then observed the entire SVG intima by pullingack the angioscope to the proximal anastomosis site.We divided them into two groups. Group I consisted of

0 patients whose LDL-C level and LDL/HDL ratio haveeen strictly maintained lower than 80 mg/dL and 1.5,espectively, after surgery by aggressive lipid controlith rosuvastatin (5 mg) at an outpatient clinic. Group II

onsisted of 11 matched patients whose LDL-C level andDL/HDL ratio have exceeded 100 mg/dL and 2.5, re-pectively, for one year after surgery treated with prav-statin 10 mg at general practitioners. Both doses oftatins were the usual starting dose of medication forapanese patients. In this series, all patients had startedaking statins approximately 14 days after the surgeryhen they first visited the outpatient clinic. Statins were

able 1. Patient Profiles

ariable Group I Group II p Value

ge 62.3 � 11.8 63.7 � 6.9 0.5624ale/female 9/1 8/3 0.3141ypertension (%) 80 63.6 0.4071iabetes (%) 50 45.5 0.835moking (%) 40 45.5 0.8008umbers of grafts 3.30 � 0.48 3.27 � 0.47 0.1315arget vessel for SVGs 0.7887LAD 1 2CX 4 5RCA 8 7

aX � circumflex artery; LAD � left anterior descending artery;CA � right coronary artery; SVG � saphenous vein graft.

ever discontinued until angioscopic assessment. Theerum level of LDL-C, HDL-C, and the LDL/HDL ratioas assessed at every month. No patients in either group

eceived beta-blockers. We compared the two groups inerms of serum level of cholesterols, triglyceride, hemo-lobin A1c, hypersensitive C-reactive protein (hsCRP),nd the incidence of SVG diseases such as graft stenosis,ccentric plaque, unstable yellow plaque, and thrombi byngiography, IVUS, and angioscopy, respectively. Wessessed 13 and 14 SVGs from groups I and II,espectively.

tatistical Examinationesults were expressed as the mean � SD. Statisticalalculations were conducted using StatView (SAS Inc, Cary,C). Using parametric and nonparametric data, statistically

ignificant differences were determined using the Student test and Fisher exact test, respectively. A p value of less than.05 was considered statistically significant.

ig 1. Yellow plaque in saphenous vein grafts detected by

able 2. Laboratory Data

ariable Group I Group II p Value

tudy durationa (month) 13.7 � 1.3 13.6 � 1.6 0.5345DL (mg/dL) 64.1 � 11.9 130.2 � 66.1 0.0038DL (mg/dL) 50.3 � 17.3 48.7 � 8.2 0.3463DL/HDL ratio 1.36 � 0.34 2.64 � 1.25 0.0036G (mg/dL) 160.2 � 81.0 125.8 � 61.1 0.2018bA1c (mg/dL) 6.0 � 0.9 6.3 � 1.4 0.2007sCRP (mg/dL) 0.045 � 0.100 0.116 � 0.020 0.0125

Study duration from surgery to graft assessment.

DL � low density lipoprotein cholesterol; HDL � high densityipoprotein cholesterol; HbA1c � hemoglobin A1c; hsCRP � hy-ersensitive C-reactive protein; TG � triglyceride.

ngioscopy.

R

TaspmwcT

g(i(dHApda2(yEgtia(5Oee

Fa

Ft

F

T

V

C

I

A

1442 HATA ET AL Ann Thorac SurgAGGRESSIVE LIPID CONTROL FOR SVG DISEASE 2009;88:1440–4A

DU

LTC

AR

DIA

C

esults

here was no difference between the groups in terms ofverage age, sex, prevalence of hypertension, diabetes,moking, and target vessels of SVGs (Table 1). Averageostoperative study duration was approximately 14onths in both groups (Table 2). The serum LDL-C levelas significantly lower in group I (64.1 � 11.9 mg/dL)

ompared with group II (130.2 � 66.1 mg/dL) (p � 0.0038;able 2). The LDL/HDL ratio was significantly lower in

ig 2. Massive thrombus formation on the yellow plaque detected byngioscopy.

ig 3. Entire clear white intima without any yellow plaque and

hrombi detected by angioscopy. Around the distal anastomosis site.

CS

roup I (1.36 � 0.34) than that of group II (2.64 � 1.25)p � 0.0036; Table 2). The Hs-CRP was significantly lowern group I (0.045 � 0.100 mg/dL) than that of group II0.116 � 0.020 mg/dL) (p � 0.0125; Table 2). There was noifference between the groups in terms of the level ofDL, triglyceride, and hemoglobin A1c levels (Table 2).ll SVGs and other arterial grafts in both groups wereatent on graft angiography. In group II, however, IVUSetected one or two large amounts of eccentric plaques inlow-density echoic lesion, which length was more thancm at around the mid portion of SVG intima in 11

78.6%) out of 14 SVGs. Angioscopy demonstrated theellow plaques (Fig 1) on the same points of SVG intima.ven though IVUS did not detect the plaques, the an-ioscopy showed some yellow plaques on the mid por-ion of SVG intima and finally, yellow plaques werendicated in all 14 SVGs (100%). Furthermore, angioscopelso revealed white thrombi on the yellow plaques in 1178.6%) of 14 SVGs (Fig 2). Graft angiography revealed0% to 75% stenosis in five of these SVGs with thrombi.n the other hand, in group I, all 13 SVGs had no

ccentric, yellow plaques or thrombi, and the intima wasntirely clear white (Figs 3 and 4). The incidence of SVG

ig 4. Around the proximal anastomosis site.

able 3. Findings of CAG, IVUS, and Angioscopy

ariable Group I Group II p Value

AGStenosis in SVG 0 5 SVGs (35.7%) 0.032

VUSEccentric plaque 0 11 SVGs (78.6%) �0.0001ngioscopyYellow plaque 0 14 SVGs (100%) �0.0001Thrombi 0 11 SVGs (78.6%) �0.0001

AG � coronary angiography; IVUS � intravascular ultrasound;VG � saphenous vein graft.

dcTn

C

Ifaahdtbraptc

wtcSosprSodnbnfpspavvrctmd

sLmoftugNTim

pbttmgavlkop

httls“glaappeawsaclsSttalHlwbec

STawgolL8m[eu

1443Ann Thorac Surg HATA ET AL2009;88:1440–4 AGGRESSIVE LIPID CONTROL FOR SVG DISEASE

AD

ULT

CA

RD

IAC

iseases such as yellow plaque and thrombi was signifi-antly higher in group II than in group I (p � 0.0001;able 3). During follow-up periods, there was no coro-ary event in either group.

omment

t is evident from this angioscopic study that the majoreatures of SVG diseases are the presence of unstabletherosclerotic plaques and thrombi. Especially, SVGtherosclerosis predisposes to thrombosis because of theigh content of lipids and tissue factor, chronic flowisturbances, and associated impairment of vasodilata-

ion. The SVG atheromas are more diffuse and vulnera-le to rupture, and the major consequences of plaqueupture in SVGs seem to be rapid platelet aggregationnd certain thrombotic occlusion [12]. Therefore, anti-latelet agents and cholesterol lowering therapy are

heoretically attractive options for the prevention of suchonsequences.

In the present study, all patients received ticlopidine asell as aspirin. Our previous study [9] showed that

iclopidine reduced the development of SVG thrombiaused by nonlaminar or sluggish flow within the SVG.everal investigators have also reported that the additionf ticlopidine to aspirin significantly inhibited high sheartress which induced platelet aggregation [13]. In theresent study, although the patients in both groups haveeceived the same antiplatelet therapy, more than 70% ofVGs in group II still had thrombi; no thrombi, on thether hand, was detected in all group I SVGs. It should beemonstrated that thrombi on the yellow plaques wereot only caused by platelet aggregation due to sluggishlood flow, but also by multiple plaque rupture of vul-erable plaques [14, 15]. The predominant mechanisms

or SVG disease are intimal hyperplasia appearing fromostoperative early phase and accelerated atherosclero-is. Saphenous vein graft atherosclerosis is especiallyrone to thrombi because of the high content of lipidsnd tissue factor, chronic flow disturbance, and impairedasodilation. These SVG atheroma are more diffuse andulnerable to rupture, and the consequences of plaqueupture in SVGs seem to be associated with almostertain thrombotic occlusion [12]. Therefore, regular an-iplatelet therapy is not enough to avoid SVG thrombi; it

ay be more important and essential to interrupt theevelopment of vulnerable yellow plaques.A post-CABG trial with angiographic assessment has

hown that aggressive lowering and maintenance of theDL-C level below 100 mg/dL was more effective thanoderate lowering of LDL-C in reducing the progression

f atherosclerosis in SVGs. It defined the prognosticactors for risk of atherosclerotic progression in SVGs inhe large and well-characterized post-CABG patient pop-lation [10]. In the present study, the LDL-C level ofroup I was maintained below 80 mg/dL. However, theational Cholesterol Education Program (NCEP) Adultreatment Panel III guidelines now recommend achiev-

ng more aggressive target levels (an LDL-C level �70

g/dL) in certain very high-risk secondary prevention t

atients [14]. O’Keefe and colleagues (15) also proposed,ased on accumulating data from multiple populations

hat do not develop atherosclerosis or coronary events,hat normal LDL-C levels are approximately 50 to 70

g/dL. Von Birgelen and colleagues [16] further sug-ested that an LDL-C level of 75 mg/dL is, on average,ssociated with plaque regression. Moreover, many in-estigators have recommended “aggressive lipid-

owering therapy” as a recent international consensus toeep serum LDL-C level between 70 and 80 mg/dL inrder to avoid the development of atheroscleroticlaques [17, 18].Nissen and colleagues [19] reported that rosuvastatin

ad a strong effect on lowering the LDL-C level to lesshan 70 mg/dL. Harley and colleagues [20] also describedhat rosuvastatin was more effective at reducing LDL-Cevels and attaining a NCEP LDL-C goal than othertatins. In the present study, however, we mentioned theaggressive lipid-controlling therapy” rather than “ag-ressive lipid-lowering therapy.” Campeau and col-

eagues [21] observed that elevated plasma cholesterolnd LDL-C, as well as low HDL-C, was associated withtherosclerotic progression in SVGs. Recently, it is im-ortant to control not only LDL-C but also HDL-C forreventing atherosclerosis development and coronaryvents. Astor and colleagues [22] suggested that a gradedssociation exists between lower levels of HDL-C andere associated with increased risk of heart attack irre-

pective of the LDL-C levels. In fact, serum HDL-C levelsre variable. Therefore, it is essential to examine theombined effects of higher HDL-C levels with variousevels of LDL-C, and the LDL/HDL ratio was the mostensitive marker for the atherosclerosis progression [23].everal studies proposed an LDL/HDL ratio of 2.0 or less

o halt the progression of atherosclerosis [24, 25], whilehat of 1.5 or less produced remarkable regression oftherosclerosis [19, 26]. In the present study, aggressiveipid controlling therapy could keep the patient’s LDL/

DL ratio less than 1.5. Furthermore, a serum LDL-Cevel was maintained at approximately 60 mg/dL, whichas almost similar to the LDL-C level of that of a neworn baby [15]. Therefore, SVG endoscopy shown asntire clear white intima in all SVGs by aggressive lipidontrol looked like “baby skin.”

tudy Limitationhis study has several limitations. It is a case controlnalysis of a single institution. Therefore, the sample sizeas slightly small, quantitative coronary analysis in an-iography has not been conducted, and we have no dataf coagulation factors in each patient. In the aggressive

ipid controlling therapy, however, serum LDL-C and theDL/HDL ratio were quite strictly controlled lower than0 mg and 1.5, respectively. These lipid controls wereuch more aggressive than those of a previous report

10]. Furthermore, as far as we know, the data from SVGndoscopy in very high-intensity statin therapy is quitenique and results were obviously better than regular

reatment. We have to keep observing the patients and

fo

CPfegHitt

R

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

1444 HATA ET AL Ann Thorac SurgAGGRESSIVE LIPID CONTROL FOR SVG DISEASE 2009;88:1440–4A

DU

LTC

AR

DIA

C

uture studies are needed to determine the effect of thebserved changes on long-term outcome.

onclusionrophylactic treatment for yellow plaque and thrombus

ormation are extremely important in the development ofarly and late SVG disease. Use of more effective neweneration statins is vitally important to achieve a LDL/DL ratio below 1.5. Aggressive lipid controlling therapy

s quite attractive to avoid post-CABG SVG disease andhe study indicated that it may be effective to maintainhe long-term graft patency.

eferences

1. Bulkley BH, Hutchins GM. Accelerated “atherosclerosis”: amorphologic study of 97 saphenous vein coronary arterybypass grafts. Circulation 1977;55:163–9.

2. Kalan JM, Roberts WC. Morphologic findings in saphenousveins used as coronary artery bypass conduits for longerthan 1 year: necropsy analysis of 53 patients, 123 saphenousveins, and 1865 five-millimeter segments of veins. AmHeart J 1990;119:1164–84.

3. Gansera B, Schmidtler F, Angelis I, Kiask T, Kemkes BM,Botzenhardt F. Patency of internal thoracic artery comparedto vein grafts–postoperative angiographic findings in 1189symptomatic patients in 12 years. Thorac Cardiovasc Surg2007;55:412–7.

4. Campbell PG, Teo KSL, Worthley SG, et al. Non-invasiveassessment of saphenous vein graft patency in asymptom-atic patients. Brit J Radiol 2009;82:291–5.

5. Knot UN, Friedman DT, Pettersson G, Smedira NG, Li J, EllisSG. Radial artery bypass grafts have an increased occur-rence of angiographically severe stenosis and occlusioncompared with left internal mammary arteries and saphe-nous vein grafts. Circulation 2004;109:2084–91.

6. Shah PJ, Bui K, Blackmore S, et al. Has the in situ rightinternal thoracic artery been overlooked? An angiographicstudy of the radial artery, internal thoracic arteries andsaphenous vein graft patencies in symptomatic patients. EurJ Cardiothorac Surg 2005;27:870–5.

7. Souza DSR, Johansson B, Bojö L, et al. Harvesting thesaphenous vein with surrounding tissue for CABG provideslong-term graft patency comparable to the left internalthoracic artery: results of a randomized longitudinal trial.J Thorac Cardiovasc Surg 2006;132:373–8.

8. Murphy GJ, Angelini GD. Insights into the pathogenesis ofvein graft disease: lessons from intravascular ultrasound.Cardiovasc Ultrasound 2004;2:8.

9. Hata M, Sezai A, Niino T, et al. What is the optimalmanagement for preventing saphenous vein graft diseases?Early results of intravascular angiographic assessment. CircJ 2007;71:286–7.

0. The Post Coronary Artery Bypass Graft Trial Investigators.The effect of aggressive lowering of low-density lipoproteincholesterol levels and low-dose anticoagulation on obstruc-tive changes in saphenous-vein coronary-artery bypassgrafts. N Engl J Med 1997;336:153–63.

1. Phan BA, Chu B, Polissar N, Hatsukami TS, Yuan C, Zhao

XO. Association of high-density lipoprotein levels and caro-

tis atherosclerotic plaque; noninvasive MR imaging forcharacterization. Int J Cardiovasc Imaging 2007;23:337–42.

2. Safian RD. Accelerated atherosclerosis in saphenous veinbypass grafts: A spectrum of diffuse plaque instability. ProgCardiovasc Dis 2002;44:437–48.

3. Isaka N, Tanigawa T, Nishikawa M, Nakano T. High shearstress induced platelet aggregation (h-SIPA) and effects ofantiplatelet therapy. Nippon Rinsho 1998;56:2624–9.

4. Grundy SM, Cleeman JI, Merz CN, et al. Implications ofrecent clinical trials for the National Cholesterol EducationProgram (NCEP) Adult Treatment Panel III. J Am CollCardiol 2004;44:720–32.

5. O’Keefe JH Jr, Cordain L, Harris WH, Moe RM, Vogel R.Optimal low-density lipoprotein is 50 to 70 mg/dl lower isbetter and physiologically normal. J Am Coll Cardiol 2004;43:2142–7.

6. Von Birgelen C, Hartmann M, Mintz GS, et al. Relationbetween progression and regression of atherosclerotic leftmain coronary artery disease and serum cholesterol levels asassessed with serial long-term (� or �12 months) follow-upintravascular ultrasound. Circulation 2003;108:2757–62.

7. Nissen SE, Tuzcu EM, Schoenhagen P, et al. Effect ofintensive compared with moderate lipid-lowering therapyon progression of coronary atherosclerosis. JAMA 2004;291:1071–80.

8. Okazaki S, Yokoyama T, Miyauchi K, et al. Early statintreatment in patients with acute coronary syndrome. Dem-onstration of the beneficial effect on atherosclerotic lesionsby serial volumetric intravascular ultrasound analysis dur-ing half a year after coronary event: The ESTABLISH Study.Circulation 2004;110:1061–8.

9. Nissen SE, Nicholls SJ, Sipahi I, et al. Effect of very high-intensity statin therapy on regression of coronary athero-sclerosis: the ASTEROID trial. JAMA 2006;295:1556–65.

0. Harley CR, Gandhi SK, Heien H, McDonough K, Nelson SP.Lipid levels and low-density lipoprotein cholesterol goalattainment in diabetic patients: rosvastatin compared withother statins in usual care. Expert Opin Pharmacother 2008;9:669–76.

1. Campeau L, Enjalbert M, Lespérance J, et al. The relation ofrisk factors to the development of atherosclerosis in saphe-nous-vein bypass grafts and the progression of disease in thenative circulation. A study 10 years after aortocoronarybypass surgery. N Engl J Med 1984;311:1329–32.

2. Astor BC, Lee F, Munter PM. Association of HDL cholesterolwith coronary heart disease risk across categories of LDLcholesterol: the Atherosclerosis Risk in Communities (ARIC)Study. J Am Coll Cardiol 2007;Mar 6:326A.

3. Phan BA, Chu B, Polissar N, Hatsukami TS, Yuan C, ZhaoXO. Association of high-density lipoprotein levels and caro-tis atherosclerotic plaque; noninvasive MR imaging for char-acterization. Int J Cardiovsc Imaging 2007;23:337–42.

4. Noike H, Hitsumoto T, Sakurai T, et al. Relationships be-tween intravascular ultrasonographic findings and coronaryrisk factors in patients with normal coronary angiography.[In Japanese] J Cardiol 2005;45:1–10.

5. Crouse JR III, Raichlen JS, Riley WA, et al. Effect of rosuv-astatin on progression of carotid intima-media thickness inlow-risk individuals with subclinical atherosclerosis: theMETEOR Trial. JAMA 2007;297:1344–53.

6. Nicholls SJ, Tuzch EM, Sipahi I, et al. Statins, high-densitylipoprotein cholesterol, and regression of coronary athero-

sclerosis. JAMA 2007;297:499–508.