普伐他汀联合普罗布考

EffectsofProbucolandPravastatinonCommonCarotidAtherosclerosisin

PatientsWithAsymptomaticHypercholesterolemia

FukuokaAtherosclerosisTrial(FAST)

YasunoriSawayama,MD,PHD,*ChieShimizu,MD,*NaoyasuMaeda,MD,*MasafumiTatsukawa,MD,*NaokoKinukawa,MS,†SamonKoyanagi,MD,PHD,‡SeizaburoKashiwagi,MD,PHD,‡JunHayashi,MD,PHD*Fukuoka,Japan

Thisstudyinvestigatedtheeffectofreducingserumlipidsoncarotidarteryintima-mediathickness(IMT)inasymptomaticpatientswithhypercholesterolemiafromFukuoka,Japan.

BACKGROUNDCarotidatherosclerosisisastrong,independentpredictorofmorbidityandmortalityin

patientswithcoronaryheartdisease(CHD).

METHODSAtotalof246asymptomatichypercholesterolemicpatients(meanage66years)were

randomizedtoreceiveeitherprobucol(500mg/day,nϭ82)orpravastatin(10mg/day,nϭ83)ortoenteracontrolgroup(dietalone,nϭ81);theywerefollowedfortwoyears.ThechangeinIMTinthecommoncarotidarterywastheprimaryendpointmeasure,andtheincidenceofmajorcardiovasculareventswasthesecondarymeasure.

RESULTSOverthetwo-yearperiod,serumlow-densitylipoprotein(LDL)cholesterolwassignificantly

reducedinthepravastatingroup(36%),theprobucolgroup(29%)andthecontrolgroup(12%)(pϽ0.0001,pϽ0.0001andpϽ0.05,respectively).Aftertwoyears,theprobucolandpravastatingroupsshowedasignificantreductioninIMT(Ϫ13.9%andϪ13.9%andpϽ0.01andpϽ0.01,respectively),buttherewassignificantIMTthickening(23.2%;pϽ0.05)inthecontrolgroup.ProbucolreducedtherateofIMTincrease,independentlyofitsreductionofLDLorhigh-densitylipoproteincholesterol.Moreover,therewasasignificantlylowerincidenceofcardiaceventsintheprobucolgroup(2.4%)thaninthecontrolgroup(13.6%)(pϭ0.0136).

CONCLUSIONSProbucolreducedcholesterollevelsandstabilizedplaque,leadingtoalowerincidenceof

cardiaceventsinthesehypercholesterolemicpatients.(JAmCollCardiol2002;39:610–6)©2002bytheAmericanCollegeofCardiologyOBJECTIVES

Overwhelmingevidencefromepidemiologicandclinicalstudies(1)hasdemonstratedthatlow-densitylipoprotein(LDL)cholesterolisakeyelementinthedevelopmentofatherosclerosis,andthatreducingLDLcholesterollevelsleadstoaloweredriskofcoronaryheartdisease(CHD).CarotidatherosclerosisisastrongindependentpredictorofmorbidityandmortalityinpatientswithCHD(2).HighplasmatotalLDLcholesterolconcentrationsareassociatedwithanincreasedprevalenceofcarotidatherosclerosis,anddrugtherapytoreduceLDLcholesterolhasbeenshowntodecreasetheextentofcarotidatherosclerosis(3).

Measurementofthecarotidarteryintima-mediathick-ness(IMT)byhigh-resolutionB-modeultrasonographyhaspreviouslybeenusedforthenoninvasivedetectionofearlycarotidatherosclerosis.TheIMTisalsoareliableendpointforinterventiontrialsassessingdiseaseprogression.Further-more,ultrasonographycandirectlyquantifyearlyathero-scleroticchangesandtheresponsetoriskfactormodifica-tion.Inaddition,themeasurementofcarotidarteryIMTby

FromtheDepartmentsof*GeneralMedicinel,and†MedicalInformatics,KyushuUniversityHospitaland‡NationalKyushuMedicalCenter,Fukuoka,Japan.ThisstudywassupportedbyagrantfromtheJapaneseMinistryofEducation,Science,andCulture,Tokyo,Japan.

ManuscriptreceivedApril3,2001;revisedmanuscriptreceivedNovember5,2001,acceptedNovember16,2001.high-resolutionultrasoundshowslessvariabilitythanthatwithangiographicmeasurementofthecarotidarteries(4),soasmallersamplesizeisrequiredtodeterminethebenefitsoftreatmentortoaccuratelyassessthepresenceofearlyatherosclerosis.

Developmentofthe3-hydroxy-3-methylglutarylcoen-zymeAreductaseinhibitors(statins)hasledtothewide-spreaduseoflipid-loweringdrugs.Theclinicalbenefitofstatinsforprimaryandsecondarypreventionofthecardio-vascularcomplicationsofatherosclerosis,especiallyinasso-ciationwithcoronaryarterydisease,hasbeeninvestigatedinseverallargeclinicaltrials(5).TheKuopioAtherosclerosisPreventionStudy(6)andtheCarotidAtherosclerosisItal-ianUltrasoundStudy(7)haverecentlyevaluatedtheeffectofpravastatinoncarotidatherosclerosis,asmeasuredbyB-modeultrasonography.BothstudiesshowedasignificantreductionintherateofcarotidarteryIMTprogression(6,7).

Probucol,4,4Ј-(isopropylidenedithio)bis(2,6-di-tert-butylphenol),hasbeenreportedtopreventatherogenesisbyactingasanantioxidantandsuppressingtheoxidativemodificationofLDL,inadditiontoitsrecognizedactioninloweringcholesterollevels(8).Ithasbeenshowntocausemarkedregressionofcutaneous(9)andtendon(10)xan-

JACCVol.39,No.4,2002February20,2002:610–6AbbreviationsandAcronyms

CETPϭcholesterolestertransferproteinCHDϭcoronaryheartdiseaseHDLϭhigh-densitylipoproteinIMTϭintima-mediathicknessLDLϭlow-densitylipoproteinMIϭmyocardialinfarction

thomasinfamilialhypercholesterolemia.Retardationoftheprogressionofcoronary(11)andfemoral(12)atherosclero-sishasbeenrelatedtoadecreaseofLDLcholesterolandanincreaseofhigh-densitylipoprotein(HDL)cholesterollevels,inagreementwiththegeneralconceptoftheroleofHDLinreversecholesteroltransport(13).Becauseprobu-colisknowntodecreaseHDLcholesterol,itseemsimpor-tanttostudytheeffectofthisdrugonatherosclerosis.Theaimofthepresentstudywastodeterminewhethertreatmentofhypercholesterolemicpatientswithprobucolorpravastatincouldretardtheprogressionandpromotetheregressionofcarotidatherosclerosis,bymeasuringtheeffectoftwoyearsoftreatmentwithprobucolorpravastatinoncarotidarteryIMTinhypercholesterolemicpatients.Tothebestofourknowledge,thisisthefirststudytouseB-modeultrasonographytocomparetheeffectsofprobucolandpravastatinoncarotidatherosclerosis.

METHODS

Studygroup.BetweenFebruary1996andFebruary2000,asymptomatichypercholesterolemicpatientswhometthefollowingcriteriawereeligibletobeenrolledinthestudy:1)primaryhypercholesterolemia(definedasaserumtotalcholesterollevelofatleast220mg/dl);and2)treatmentwitheitherprobucolorpravastatin.Atotalof246Japanesepatientsbetweentheagesof30andyearswereenrolled.Ն350Exclusionmg/dl;uncontrolledcriteriaincludedheartafailure;serumrecenttriglyceride(Ͻ6months)levelmyocardialinfarction(MI);severeorunstableanginapec-toris;hypothyroidism/hyperthyroidismorotherendocrinediseases;secondaryhyperlipidemia;uncontrolleddiabetesmellitus;uncontrolledhypertension;heavydrinking;obesepatientsonweightreductionprograms;diseasesthatmightinterferewithdrugabsorption;anysevereillness;andtreatmentwithcertaindrugs,includingcorticosteroids,otherlipid-loweringagentsorantacidscontainingalumi-numsalts.Written,informedconsentwasobtainedfromeachpatient,andthetrialwasapprovedbytheEthicsCommitteeofKyushuUniversityHospital.

Studydesign.Patientswererandomlyassignedtooneofthefollowingthreegroups:1)aprobucolgroup(nϭ82,age41to80years)thatreceivedprobucolat500mgtwicedailyaftermeals;2)apravastatingroup(nϭ83,age41toyears)thatreceivedpravastatinat10mg/dayaftertheeveningmeal;and3)acontrolgroup(nϭ81,age30toyears)thatwasondietalone.Randomizationwasdoneby

Sawayamaetal.611

ProbucolandPravastatinforCarotidAtherosclerosis

theminimizationmethod,controllingforthefollowingfour

factors:totalcholesterollevel,age,genderandIMT.Hos-pitalvisitsformonitoringwerescheduledaftertwoweeksandeveryfourweeksthereafter.Ateachvisit,abriefphysicalexaminationwasdone,andthenumberoftabletswascountedtoassesscompliance.Measurementoflipids,lipoproteinsandsafetylaboratoryvariableswasalsodoneateachvisit.Thirty-four(21%)ofthe165patientsintheintent-to-treatpopulationdidnotcompletethestudy.DEFINITIONOFENDPOINTS.

Theprimaryendpointwas

therateofprogressionofcarotidatherosclerosis,whichwasmeasuredastheslopeofthechangeinthemeanIMTofsixcarotidsegments(3siteseachintheleftandrightcommoncarotidarteries,whichwere2,2.5and3cmproximaltothecarotidbifurcation)onultrasoundexamination.AnincreaseinIMTovertwoyearswasdefinedas“progression”andadecreaseinIMToverthesameperiodwasdefinedas“regression.”Ultrasonographywasperformedatenrollmentandtheneverysixmonthsforthenext24months.Thesecondaryendpointwastheincidenceofmajoratheroscle-roticevents,aseffectedbyeachtreatment.

CLINICALEVENTS.

Clinicalcardiac,cerebrovascularand

peripheralvascularevents,aswellasdeaths,weremoni-tored.Thetimeuntilthefirstcardiacevent(coronaryangioplasty,coronaryarterybypassgraftsurgery,definiteorprobableMIorunstableanginapectorisrequiringhospitaladmission)andthetimeuntildeathfromanycauseweredetermined.

B-MODEULTRASONOGRAPHY.

Theprocedureusedfor

measuringcarotidarteryIMT,aswellasitsreproducibility,hasbeendescribedelsewhere(14).Inbrief,ultrasonographywasperformedwiththepatientinthesupineposition,usinganAlokaSSD-2000system(Aloka,Tokyo,Japan)witha7.5-MHztransducer.Allexaminationswereperformedbyonetrainedphysicianwhohadnoknowledgeoftheclinicalhistoryandriskfactorprofileofthesubjects.TheIMTofthefarwallofthecarotidarterywasmeasuredat2,2.5and3cmproximaltothecarotidbifurcationineachoftherightandleftcommoncarotidarteries.Measurementsweremadeonlongitudinalscansobtainedintheanterioroblique,lateralandposteriorobliqueviews.TheIMTwasdefinedasthedistancebetweentwoechogeniclinesseparatedbyahypoechoicoranechoicspace,withtheouterlinecorre-spondingtothemedia-adventitiaborderandtheinnerlinerepresentingthelumen-intimaborder.ThemeanIMTwascalculatedastheaveragevalueoftheIMTmeasurementsforthesixsitesinthecarotidarteries.Stenosiswasdefinedasplaque(asitewhereIMTwasՆ1.10mm)occupyingmorethanhalfoftheluminalcircumferenceofthearteryonatransversescan.Athree-leadelectrocardiogramwasre-cordedsimultaneously,andvideotaperecordingoftheexaminationwasdonecontinuously.ImagesofeachoftheIMTmeasurementsiteswererecordedonsuperVHSvideotapeforsubsequentoff-lineanalysis.

612

Sawayamaetal.ProbucolandPravastatinforCarotidAtherosclerosis

Table1.BaselineCharacteristicsoftheStudyGroup

Total(n؍246)

Age(years)Men(%)

Bloodpressure(mmHg)SystolicDiastolic

Bodymassindex(kg/m2)Smoking(%)

RecentorformerNeverHistory(%)HypertensionDiabetes

CerebralvesseldiseaseCoronaryheartdiseaseSerumcholesterol(mg/dl)Total

Low-densitylipoproteinHigh-densitylipoproteinSerumtriglycerides(mg/dl)Intima-mediathickness(mm)

66.1Ϯ13.777(31.3)130.8Ϯ21.977.1Ϯ11.423.2Ϯ3.9146(59.3)100(40.7)102(41.5)56(22.8)103(41.9)35(14.2)253.0Ϯ26.3166.1Ϯ31.257.0Ϯ16.6149.2Ϯ72.61.308Ϯ0.597

ProbucolGroup

(n؍82)65.7Ϯ12.925(30.4)132.2Ϯ22.777.8Ϯ13.023.5Ϯ4.248(58.5)34(41.6)37(45.1)15(18.3)35(42.7)9(11.0)252.4Ϯ25.6166.0Ϯ27.057.8Ϯ19.1142.7Ϯ67.91.347Ϯ0.2

PravastatinGroup

(n؍83)

65.6Ϯ14.822(26.5)129.2Ϯ22.775.6Ϯ11.223.5Ϯ4.444(53.0)39(47.0)39(47.0)18(21.7)34(41.0)15(18.1)251.5Ϯ24.8160.7Ϯ31.656.7Ϯ15.6168.7Ϯ84.61.267Ϯ0.2

JACCVol.39,No.4,2002February20,2002:610–6ControlGroup

(n؍81)67.1Ϯ13.430(37.0)130.9Ϯ20.277.8Ϯ9.622.9Ϯ2.754(66.7)27(33.3)26(32.1)23(28.4)34(42.0)11(13.6)255.2Ϯ28.4171.6Ϯ33.956.5Ϯ15.0135.8Ϯ59.11.286Ϯ0.501

DataarepresentedasthemeanvalueϮSDornumber(%)ofpatients.

Laboratorytests.Bloodsampleswerecollectedbetween8and9AMaftera12-hfast.Serumcholesterolandtriglyc-eridelevelsweremeasuredenzymatically.High-densitylipoproteincholesterolwasmeasuredinthesupernatantafterprecipitationofapolipoproteinB–containinglipopro-teins,accordingtothecalciumheparinmethod,andLDLcholesterolwascomputedusingFriedewald’sformula(15).Allmeasurementsweremadeonthedayofbloodcollection,inwhichcasethebloodwasstoredfornolongerthanthreedaysatϪ4°Cuntilassay.

Statisticalanalysis.Alldatawererecordedonstandard-izedformsandwereenteredintoadatabase.ResultsareexpressedasapercentageormeanvalueϮSD.Differencesinfrequencieswerecomparedusingthechi-squaretest,asappropriate.ComparisonofmedianvaluesbetweenmorethantwogroupswasdonebythenonparametricKrusal-Wallistest.Differencesbetweenserumlipidlevelsatbase-lineandafter6,12,18and24monthswerecomparedusingFriedman’stest,withtheBonferronicorrectionformultiplecomparisons.

TheprimaryendpointofthisstudywasthechangeinIMT(mm)after24months.DifferencesinIMTamongthethreegroupswereevaluatedbythenonparametricKrusal-Wallistest.MultiplecomparisonsofIMTchangesatvarioustimesofmeasurementwereevaluatedwiththeBonferronicorrection.ForwardstepwisemultiplelinearregressionanalysiswasperformedtodetectfactorsthatinfluencedtherateofchangeinIMT,anddifferencesinserumlipidslevelsfrombaselinewereanalyzedandselectedasvariablesforentryintothemodelbytheforwardstepwisemethod.CorrelationsbetweenthepercentreductioninIMTincreaseandLDLorHDLcholesterollevelaftertwoyearsoftreatmentwereevaluatedbycalculatingSpearman’scorrelationcoefficients.

Thesecondaryendpointwastheincidenceofmajoratheroscleroticevents,aseffectedbyeachtreatment.Differ-encesbetweenthethreegroupswereassessedbythelog-ranktest,withtheBonferronicorrectionformultiplecomparisons.ApvalueϽ0.05wasconsideredstatisticallysignificantinallanalyses.Alldatawereanalyzedonanintent-to-treatbasis.

RESULTS

Baselinecharacteristics.ThebaselinecharacteristicsofthesubjectsaresummarizedinTable1.Themeanageofthepatientswas66.1years,and31.3%weremen.Averagesystolicanddiastolicbloodpressureswere130.8and77.1mmHg,respectively.Ofthe246patients,146(59.3%)wererecentorformersmokers,102(41.5%)hadahistoryofhypertensionand56(22.8%)haddiabetesmellitus.BaselineserumtotalcholesterolandLDLcholesterollevelswere253.0and166.1mg/dl,respectively.TheHDLcholesterolandserumtriglyceridelevelswere57.0and149.2mg/dl,respectively.ThemeanIMTwas1.308mm.Therewerenostatisticallysignificantdifferencesinanyofthesebaselinecharacteristicsamongthethreegroups.

Drugtreatmentandserumlipids.Thechangesintotalcholesterol,LDLcholesterol,HDLcholesterolandtriglyc-eridesareshowninFigure1.

Therewasasignificantdecreasein

theserumtotalcholesterollevelafter12monthsofthera-py—by20.0%intheprobucolgroupandby20.1%inthepravastatingroup(bothpϽ0.001byFriedman’stest).After24monthsoftherapy,therewasafurtherdecreaseinbothgroups—24.1%and23.0%totalreductions,respec-tively,whencomparedwithbaseline(bothpϽ0.00l;Friedman’stest).Inthecontrolgroup,thetotalcholesterol

TOTALCHOLESTEROL.

JACCVol.39,No.4,2002February20,2002:610–6Sawayamaetal.ProbucolandPravastatinforCarotidAtherosclerosis

613

Figure1.Serumlipidlevelsintheprobucol(triangles),pravastatin(squares)andcontrol(circles)groups.DataarepresentedasthemeanvalueϮSD.*pϽ0.05;***pϽ0.001vs.baseline.

levelwasreducedattheendofthestudy,butthedifferencewasnotsignificant.

SerumLDL

cholesterollevelswerealsoreducedafter12months,fallingby24.2%intheprobucolgroup(pϽ0.001byFriedman’stest),by32.7%inthepravastatingroup(pϽ0.001byFriedman’stest)andby5.1%inthecontrolgroup.After24months,therewasafurtherdecreaseinallthreegroups,withadeclineof28.6%,35.9%and8.5%,respectively,whencomparedwithbaseline(pϽ0.00l,pϽ0.001andpϽ0.05,respectively,byFriedman’stest).

LOW-DENSITYLIPOPROTEINCHOLESTEROL.

Intima-mediathickness.TheIMTchangesintheprobu-col,pravastatinandcontrolgroupsareshowninFigure2.Intheprobucolgroup,IMTwassignificantlyreducedafter12monthsoftherapy(8.3%;pϽ0.01byFriedman’stest).After24monthsoftherapy,therewasafurthersignificantreduction,foratotalreductionof13.9%,comparedwithbaseline(pϽ0.0lbyFriedman’stest).NosignificantreductioninIMTwasfoundduringthefirst18monthsof

Theserum

HDLcholesterollevelofthepravastatingroupwasin-creasedsignificantly(by6.4%)after24months(pϽ0.05byFriedman’stest).Inthecontrolgroup,HDLcholesterolalsoincreasedbetweenbaselineand24months(by5.4%),butthechangewasnotsignificant.Incontrast,theHDLcholesterolleveloftheprobucolgroupwassignificantlyreducedafter6months(21.8%;pϽ0.00lbyFriedman’stest)andafter24months(20.7%;pϽ0.05byFriedman’stest).

HIGH-DENSITYLIPOPROTEINCHOLESTEROL.TRIGLYCERIDES.

Triglyceridelevelsshowednosignificant

changesthroughoutthecourseofthestudy.

Figure2.Changesinintima-mediathickness(IMT)intheprobucol(triangles),pravastatin(squares)andcontrol(circles)groups.DataarepresentedasthemeanvalueϮSD.*pϽ0.05;**pϽ0.01vs.baseline.

614

Sawayamaetal.ProbucolandPravastatinforCarotidAtherosclerosis

Table2.SeriousAdverseEventsReportedDuringtheTwo-YearFollow-UpPeriod

ClinicalEventTotalcardiovasculareventsFatalMINonfatalMIPTCA/CABG

TotalcerebrovasculareventsTotalothereventsTotaldeaths

ProbucolGroup

(n؍82)

2(2.4%)2(2.4%)0000

1(2.4%)

PravastatinGroup

(n؍83)

4(4.8%)3(3.6%)1(1.2%)1(1.2%)0

2(2.4%)5(6.0%)

JACCVol.39,No.4,2002February20,2002:610–6ControlGroup

(n؍81)11(13.6%)8(9.9%)3(3.7%)00

1(1.2%)9(11.1%)

Dataarepresentedasthenumber(%)ofpatients.

CABGϭcoronaryarterybypassgraftsurgery;MIϭmyocardialinfarction;PTCAϭpercutaneoustransluminalcoronaryangioplasty.

therapyinthepravastatingroup,buttherewasasignificantreductionby13.9%after24months(pϽ0.01byFriedman’stest).Inthecontrolgroup,IMTincreasedsignificantlyby23.3%after24months(pϽ0.05byFriedman’stest).ThechangeinIMTwassignificantlygreaterintheprobucolandpravastatingroupsthaninthecontrolgroup(bothpϽ0.001bytheKruskal-Wallistest).TherewasnosignificantdifferenceinthechangeinIMTbetweentheprobucolandpravastatingroupsat24weeksaftertheendoftreatment.MultipleregressionanalysisofthereductioninIMT.ForwardstepwisemultiplelinearregressionanalysiswasperformedtoassessthefactorsinfluencingIMTregression.Lipid-loweringtherapywasfoundtobethemostimportantindependentfactorassociatedwiththerateofIMTregres-sion(Fϭ13.60,pϽ0.0001).ThereductionofLDLcholesterolaftertwoyearsoftreatmentwasalsoindepen-dentlyassociatedwiththerateofIMTregression(Fϭ12.28,pϭ0.0007).Incontrast,ageshowedanindependentassociationwiththerateofIMTprogression(Fϭ11.53,pϽ0.0009).TherewerenosignificantassociationsbetweenIMTchangesandanyoftheothervariablesinvestigated.CorrelationbetweenpercentreductioninIMTprogres-sionandLDLcholesterol.Therewasaweakbutsignif-icantandpositivecorrelationbetweentheabsolutechangeinIMTandthechangeinLDLcholesterolintheprava-statingroup(rϭ0.363,pϭ0.0051bySpearman’smethod),buttherewasnosuchcorrelationintheprobucolorcontrolgroup(rϭ0.065,pϭ0.52andrϭ0.130,pϭ0.3321,respectively,bySpearman’smethod).TherewasnocorrelationbetweentheabsolutechangeinIMTandthechangeinHDLcholesterolinanyofthegroups.

Cumulativeincidenceofmajorcardiovasculareventsandmortality.Amongthe82patientsintheprobucolgroup,twohadamajorcardiovascularevent(2diedofCHD),comparedwith4ofthe83patientsinthepravastatingroup(3diedofCHDand1hadanonfatalMI)and11ofthe81patientsinthecontrolgroup(8diedofCHDand3hadanonfatalMI)(Table2,Fig.3).Ofthe16deathsthatoccurredduringthisstudy,twowereintheprobucolgroup,5wereinthepravastatingroupand9wereinthecontrolgroup.Amongthese16deaths,13werefromcardiovascularcauses,whereastheotherswerefromgastrointestinalbleed-ingandinfection.Therewasasignificantlylowerincidence

ofcardiaceventsintheprobucolgroupthaninthecontrolgroup(pϽ0.05bythelog-ranktest).Therewasalsoalowerincidenceofdeathintheprobucolgroupthaninthecontrolgroup,butthedifferencewasnotsignificant.

DISCUSSION

Thepresentstudyaddsnewinformationtotheever-growingpoolofdataonanti-atherogenicregimens.TheFukuokaAtheroSclerosisTrial(FAST)isthefirststudytodemonstratethebenefitofprobucolinpatientswithhyper-cholesterolemiaandalsotodemonstrateaneffectofprobu-colontheincidenceofcardiovasculardisease.

Probucolisamongthemostextensivelyusedlipid-loweringdrugsandhasbeenassociatedwithmanypoten-tiallybeneficialoutcomes.Despitethis,probucoltherapylostsomecredibilityaftertheapparentlynegativefindingsoftheProbucolQuantitativeRegressionSwedishTrial(PQRST)(16).ThechangesinIMTinourprobucolgroupareinstrikingcontrasttothosereportedinPQRST,inwhichthetargetvesselwasthefemoralartery.However,reportsofIMTregressioninthefemoralarteryarelimited,andthearterialcalibermonitoringmethodusedinPQRSTcouldpotentiallybeinfluencedbyarterialremodeling(17).

Figure3.Kaplan-Meierestimatesoftheoccurrenceofallcoronaryevents.Boldsolidlineϭcontrolgroup;thinsolidlineϭpravastatingroup;brokenlineϭprobucolgroup.Therewasasignificantdifferencebetweenthecontrolgroupandprobucolgroups(*pϽ0.05bythelog-ranktest).

JACCVol.39,No.4,2002February20,2002:610–6B-modeultrasonography.TheB-modeultrasonographymodalityusedinthisstudyisanattractivetoolforassessingtheeffectofdrugsonatherosclerosis.Theamountofdataobtainedbythismethodisgrowingthroughanumberofotherclinicaltrialsnowinprogress.Moreover,increasinglypreciseestimatesoftheprogressionrateundervariousclinicalconditionsandtheeffectoftreatmentonprogres-sionhavebecomeavailable(16).ThismaybebecausecardiovasculareventsinfluencethevesselwallratherthanthelumenandbecausetheassociationbetweenarterialIMTandthelumendiameterisimprecise(17).Insupportofthisview,wefoundthattheIMTprogressioninthecommoncarotidarterywassignificantlyreducedintheprobucolandpravastatingroups,butnotinthecontrolgroup.ThechangesincarotidarteryIMTinthepravastatingroupresembledthosereportedpreviously(6,7).

Lipid-loweringeffect.EvidencefromepidemiologicandclinicalstudieshasdemonstratedthatLDLcholesterolisakeyelementinthedevelopmentofatherosclerosis,andthatreducingLDLcholesterollevelsleadstoalowerriskofCHD.Recentstudiesusingpravastatinhavedemonstratedasubstantialreductionincholesterol,alongwithasignifi-cantreductionincoronaryandall-causemortality(18).ThethreegroupsinthisstudyshowedasubstantialreductionintheLDLcholesterollevel.WefoundthereductioninLDLcholesteroltobemoresignificantinthepravastatinthanintheprobucolandcontrolgroups.AlthoughthecontrolgroupshowedasignificantreductionofLDLcholesterolbydietalone,progressionofIMTstilloccurred,unliketheoutcomeintheactivetreatmentgroups.Thissuggeststhatlife-stylemodificationalonedidnotadequatelyretardIMTprogression.

Anothermechanismoftheanti-atherogeniceffectofprobucol.Ourstudyrevealedthatprobucolnotonlyre-ducedtheLDLlevel,butalsocausedadecreaseinHDLcholesterol.ThisreductioninHDLcholesterolbyprobucolmayreflectanincreaseinreversecholesteroltransport,mostlikelyresultingfromthedirectinductionofcholesterolestertransferprotein(CETP),consequentlyhavingananti-atherogeniceffect(19),becauseprobucolhasbeenreportedtoenhancereversecholesteroltransportfromtheperipherytotheliverbyHDL(20),andthisactivityappearstobedependentonincreasedCETPactivity(19).However,probucoltherapyretardedtheprogressionofIMT,inde-pendentofitsLDLorHDLcholesterol-loweringeffectinthepresentstudy.Moreover,theIMTreductionoccurredearlierwithprobucolthanwithpravastatin.ThesefindingssuggestthattheremaybeanothermechanisminvolvedintheeffectonIMTprogressionthatisindependentofthelipid-loweringeffectofprobucol.

Stabilizationofplaque.Numerouspathophysiologicob-servationsinhumansandanimalshaveledtotheresponse-to-injuryhypothesisofatherosclerosis,whichemphasizesendothelialdysfunctionasthefirststepintheatheroscle-roticprocess.EndothelialdysfunctionispossiblyrelatedtoanincreaseofLDLandmodifiedLDL,geneticalterations,

Sawayamaetal.615

ProbucolandPravastatinforCarotidAtherosclerosis

elevationofplasmahomocysteineorinfectionwithmicro-organisms,oracombinationoftheseorotherfactors(21).Theatheroscleroticprocessleadstotheenlargementofvulnerableplaques,whicharecharacterizedbyalipid-richcoreandathinfibrouscap.Theadvancedstagesareassociatedwithaccumulationofmonocytesandlympho-cytes,plateletadhesionandapoptosis.Cardiacevents,suchasacutecoronarysyndromes(e.g.,unstableanginaandMI),appeartoresultfromplaquerupture(22).Probucolhasalsobeenshowntomodifyvascularremodelingafterballoonangioplasty,andremodelingisanimportantcomponentoftherestenoticprocess(23).Inaddition,probucolhasaradical-scavengingeffectandinhibitstheproductionofplatelet-derivedgrowthfactorandinterleukin-1,givingitananti-inflammatoryeffect(24).Suchreportssuggestthatprobucolmaynotonlyhavealipid-loweringeffect,butmayalsostabilizeplaque.Interestingly,thepresentstudydem-onstratesthattheincidenceofcardiaceventswaslowerintheprobucolgroupthaninthecontrolorpravastatingroup.Studylimitations.Becausethesamplesizewassmallinthepresentstudy,adifferenceintheeffectsofprobucolandpravastatinonIMTcouldnotbedemonstrated.Therefore,alarge-scaleinvestigationwillbenecessarytodeterminewhetherprobucolandpravastatinshowanydifferencesintheireffectonIMT.Thelackofaplacebocontrolgroupwasanotherlimitation.However,theuseofquantitativeB-modeultrasonographyallowedustoobtainunbiaseddata.

Conclusions.Probucolwaseffectiveinreducingthecho-lesterollevelandstabilizingplaqueinhypercholesterolemicpatients,whichprobablyexplainsareductionintheinci-denceofcardiacevents.

Reprintrequestsandcorrespondence:Dr.JunHayashi,Depart-mentofGeneralMedicine,KyushuUniversityHospital,Higashi-ku,Fukuoka812-8582,Japan.E-mail:hayashij@genmedpr.med.kyushu-u.ac.jp.

REFERENCES

1.LevineGN,KeaneyJF,Jr.,VitaJA.Cholesterolreductionincardio-vasculardisease:clinicalbenefitsandpossiblemechanism.NEnglJMed1995;332:512–21.

2.CriquiMH,LangerRD,FronekA,etal.Mortalityoveraperiodof10yearsinpatientswithperipheralarterialdisease.NEnglJMed1992;326:381–6.

3.HodisHN,MackWJ,LaBreeL,etal.Reductionincarotidarterialwallthicknessusinglovastatinanddietarytherapy:arandomizedcontrolledclinicaltrial.AnnInternMed1996;124:548–56.

4.RileyWA,BarnesRW,ApplegateWB,etal.Reproducibilityofnoninvasiveultrasonicmeasurementofcarotidatherosclerosis:theAsymptomaticCarotidArteryPlaquestudy.Stroke1992;23:1062–8.5.TheLong-termInterventionwithPravastatininIschaemicDisease(LIPID)StudyGroup.Preventionofcardiovasculareventsanddeathwithpravastatininpatientswithcoronaryheartdiseaseandabroadrangeofinitialcholesterollevels.NEnglJMed1998;339:1349–57.6.SalonenR,NyyssonenK,PorkkalaE,etal.KuopioAtherosclerosisPreventionStudy(KAPS):apopulation-basedprimarypreventivetrialoftheeffectofLDLloweringonatheroscleroticprogressionincarotidandfemoralarteries.Circulation1995;92:1758–.

7.MercuriM,BondMG,SirtoriCR,etal.Pravastatinreducescarotidintima-mediathicknessprogressioninanasymptomatichypercholes-

616

Sawayamaetal.ProbucolandPravastatinforCarotidAtherosclerosisterolemicMediterraneanpopulation:theCarotidAtherosclerosisItal-ianUltrasoundStudy.AmJMed1996;101:627–34.

8.StrandbergTE,TilvisRS.Probucol:physiologicalandpharmacolog-icalregulationofsmallintestinalcholesterolsynthesis.GenPharmacol1988;19:321–9.

9.YamamotoA,MatsuzawaY,KishinoB,HayashiR,HirobeK,KikkawaT.Effectsofprobucolonhomozygouscasesoffamilialhypercholesterolemia.Atherosclerosis1983;48:157–66.

10.YamamotoA,MatsuzawaY,YokoyamaS,FunahashiT,YamamuraT,KishinoB.Effectofprobucolonxanthomataregressionfamilialhypercholesterolemia.AmJCardiol1986;57:29H–35H.

11.BrownBG,ZhaoXQ,SaccoDE,AlbersJJ.Lipidloweringandplaqueregression:newinsightsintopreventionofplaquedisruptionandclinicaleventsincoronarydisease.Circulation1993;87:1781–91.12.BlankenhornDH,AzenSP,CrawfordDW,etal.Effectsofcolestipol-niacintherapyonhumanfemoralatherosclerosis.Circula-tion1991;83:438–41.

13.MillerNE.HDLmetabolismanditsroleinlipidtransport.EurHeartJ1990;11SupplH:1–3.

14.PignoliP,TremoliE,PoliA,OresteP,PaolettiR.Intimalplusmedialthickenssofarterialwall:adirectmeasurementwithultrasoundimaging.Circulation1986;74:1399–401.

15.

WerbaJP,SafaO,GianfranceschiG,MichelagnoliS,SirtoriCR,FranceschiniG.Plasmatriglyceridesandlipoprotein(a):inverserela-tionshipinahyperlipidemicItalianpopulation.Atherosclerosis1993;101:203–11.

JACCVol.39,No.4,2002February20,2002:610–616.WalldiusG,EriksonU,OlssonAG,etal.Theeffectofprobucolon

femoralatherosclerosis:theProbucolQuantitativeRegressionSwedishTrial(PQRST).AmJCardiol1994;74:875–83.

17.CrouseJR,GoldbourtU,EvansG,etal.Arterialenlargementinthe

AtherosclerosisRiskInCommunities(ARIC)cohort:invivoquanti-ficationofcarotidarterialenlargement.Stroke1994;25:1354–9.

18.ShepherdJ,CobbeSM,FordI,etal.Preventionofcoronaryheart

diseasewithpravastatininmenwithhypercholesterolemia.NEnglJMed1995;333:1301–7.

19.BlankenhornDH,AzenSP,CrawfordDW,etal.Effectsof

colestipol-niacintherapyonhumanfemoralatherosclerosis.Circula-tion1991;83:438–47.

20.FranceschiniG,WerbajP,CalabresiL.Drugcontrolofreverse

cholesteroltransport.PharmacolTher1994;61:2–324.

21.RossR.Atherosclerosis:aninflammatorydisease.NEnglJMed

1999;340:115–26.

22.FusterV,BadimonL,BadimonJ,ChesebroJH.Thepathogenesisof

coronaryarterydiseaseandtheacutecoronarysyndromes(2).NEnglJMed1992;326:310–8.

23.TardifJC,CoteG,LesperanceJ,etal.Probucolandmultivitaminsin

thepreventionofrestenosisaftercoronaryangioplasty.NEnglJMed1997;337:365–72.

24.ChangMY,SasaharaM,ChaitA,RainesEW,RossR.Inhibitionof

hypercholesteromia-inducedatherosclerosisinthenonhumanprimatebyprobucol.II.Cellularcompositionandproliferation.ArteriosclerThrombVascBiol1995;15:1631–40.