Feasibility of planar fluorine-18-FDG imaging after recent myocardial infarction to assess...

tion is enhanced in ischemic but viable myocardium. Thisincreased myocardial glucose utilization can be traced noninvasively by PET using ‘8F-fluorodeoxyglucose(FDG)(Z3). Tilhisch et al. and Tamaki et al. (4,5) showed recovery of wall motion abnormalities in patients undergoingcoronary bypass surgery could be predicted by normal orenhanced myocardial FDG uptake before revascularization. These studies have validated FDG as a valuabletracer of myocardial viability.

It is possible to visualize myocardialFDG uptakewith astandardgamma camera and a 511-keY collimator (6—8).Because gammacamerasystems aremore readilyavailablethan PET systems, development of planarFDG scintigraphy to assess myocardialviability may enable the application of FDG in routine clinical practice. We performed astudy in patients shortly after myocardial infarction to define the clinical feasibility of planar myocardial FDG imaging and to investigate the relationship between @°‘TlandFDG in myocardial areas with normal and abnormal leftventricular function.

MATERIALS AND METHODS

PatientsThestudywas approvedby theEthicalCommitteeon Human

Researchof theFreeUniversityHospital.Fifty-oneconsecutivepatients (41 men; mean age 60 ±12yr) admittedto the coronarycare unit with myocardialinfarctionwere studied. All patientswere admitted with a typical histoiy of prolonged chest pain,transiently elevated serum enzyme levels including CK-MB fractions, and electrocardiographicsigns of either a 0-wave or anon-Q-waveinfarction.Forty-fourpatientshad 0-wave infarctionandseven patientshadnon-0-waveinfarctionby electrocardiographiccriteria.Thirty-threepatientsweretreatedwiththrombolytic agents in the acute stage of infarction.

Exclusioncriteriaforenteringthestudywereunstableanginapectoris, insulin-dependentdiabetes mellitus,severe congestiveheart failure(Killipclass III—IV)and major noncardiacdisease.Afterinformedconsentwas obtained,allpatientsunderwentscmtigraphicevaluationby @°@TlandFDOat reston the sameday;which was within 5 ±2 days of infarction.

Theaimofthisstudywasto definethedinicalfeasibilityofplanarmyocardial 18F-fluorodeoxyglucose(FDG) imagingand to asseas the relationbetween @°i1,FDGand leftventricularfunctionearly after rnyocardial infar@on.Methods: Fifty-onepatientswere studied 5 ±2 days after infar@on.Sdntigraphic imageswere ViSUallyand quantitativalyanalyzed usinga drcumferentialprofilestechnique. FDGuptake was normalizedto the area withma)dmal201fl uptake. ScintigraphiCdata were compared withIeftventricularwall motion as assessed byventriculography in 22patients. Relativeregional201@fluptake was categOriZedas normel ( 75%ofpeak activity),mOderatelyreduced (50%-75%)orsevere'y reduced (< 50%). These tracer defects were conskiered viable it FDG uptake exceeded 201fl uptake by 20%and/or@ffFDGuptakewas normal( 75%).AllregionswithFDGuptake 20% less than 2OFfluptake were considered nonviable.Results: Fourhundredforty-onemyocardialregionswerearialyzed; 200 showed normal @°iiuptake, 241 had reduced uptake, 191 had moderatalyreduced @°i1uptake and 50 regionshad severely reduced uptake. Viabilityfor mOderatelyand asveralyreduced regionswas observed in 62% and 48%, respectivaly.A concordance between flow and metabolismwas observed in 38% and 52%, respectivaly. Conclusion: MyocardialFDGimagingisfeasiblewithstandardgammacamerasystemsand enables the identificationof regionswithpreserved glucosemetabolism in patients shortlyafter infarction.

Key Words: planar myocardial imaging; fluonne-18-fluorodeoxyglucose; myocardial infarction; rnyocardial viability

J Nuci Med 1995; 36:975-981

nder normal physiologic conditions, and particularlyin the fasting state, fatty acids are the preferredsubstratesfor myocardial energy production (1). Only a minority ofthe energy is obtained from glucose and lactate. Both animal and clinical studies have shown that glucose utiliza

ReceivedJun. 13, 1994;revisionaccepted0@ 24, 1994.Forcorrespondenceorreprintscont@F.C.NAsser,MD,PhD,Departmentof

Cardiology,Free UniversityHospital,De Boelelaan1117,1007MBAmsterdam,TheNstherIw@ds.

975Planar MyocardialFDG Imaging@ Huftinkat al.

Feasibility of Planar Fluorine- 18-FDG Imagingafter Recent Myocardial Infarction to AssessMyocardial ViabilityJohannes M. Huitink, Frans C. Visser, Arthur van Lingen, A.B. Johan Groeneveld, Jeroen J. Bax,Guido R. van Leeuwen, Gerard M. Visser, Mart J.A. Van Loon, Gerrit J.J. Teule and Cees A. Visser

Departments of Cardiolo@, Nuclear Medicine, IntenzalMedicine, Fn@eUniversizy Hospita4 and the Radionuclide Center,Free University, Amsterdam@ The Netherlands

ThaIllum-201 ScintigraphyRest @°‘Tlscintigraphywas performedin the morningaftera

lightmeal.Tento 15mmafterintravenousadministrationof74—90MBq201'flviews were obtainedin the standard left anterioroblique 45 (LA045), anteriorand left anterioroblique 70 (LAO7O)positions.The gammacamerawas equippedwith a 9-mmcrystaland a low-energyall-purposecollimator.Imageswere obtainedusing 8 rain per view in the zoom mode in a wordframe matrix of128x 128pixelsand the scintigraphicdata were stored on disk.To ensure correct patient positioning during FDG imaging, thescintigraphicallydeterminedpositionof the center of the leftyentricle was marked on the chest of the patient using a marker (57Co)and a pen with water-insolubleink after each @°‘Tlimage.

FDG ScintigraphyDeoxyglucose was labeled with ‘8Fat the Radionudide Center

of the Free University; no carrierwas added to the 18F.A typicalpurity of greater than 95%was obtained. Transportationto thehospitallasted20mm.Forty-fiveminutesbeforeintravenousadministration of 185MBq FDG, an oralglucose load of2O—50gwasgiven togetherwith a lightcarbohydratemeal. The mealwas toprevent fastingconditions duringthe study, promoteFDG uptakeandminimizethe heterogeneityofmyocardial FDG uptake(9,10).Forty minutesafter injection,scintigraphywas started. StandardLA045, anterior and LAO7Oviews were obtained. The sameconventionalAngergammacameraequippedwith a 511-keVcollimator (van Mullekom, Nuclear Fields, Boxmeer, The Netherlands)was used. Startingwitha 360-keVcollimator,adjustmenttoa 511-keV collimatorwas achieved by increasingthe bore lengthandseptalthickness.The holediameterhad to be increasedto 4.6mm to ascertain that the collimator weight did not exceed 150 kg.The septal thickness of this collimatoris 2.11 mm and the spatialresolution of this imaging system using a line spread functionFWHMhasbeenestimatedto be 14mmfor ‘8Fin 10cmwater(7).

The FDO imageswere acquiredwith an energywmdow of 20%around the photopeak of ‘8F(511 keV) in the zoom mode in awordframe matrix of 128 x 128pixels during 8 min per view. Allscintigraphicdatawere storedon disk. Bloodsampleswere takento measure plasma glucose and insulin levels shortly before theadministration of FDG in 23 patients.

Cardiac CatheteriZatIOnCardiac catheterizationwith coronary angiographyand left

ventriculography was performed in 22 patients using the Judkins'technique. Cardiac catheterization was performed within 6 ±5days of infarction.Left ventriculogramswere obtained in all patientsina30°rightanterioroblique(RAO3O)projectionandinthe50°left anterior oblique(LAO6O)position.The ventriculogramsand coronary angiogramswere reviewed by two experienced ohservers without knowledge of the scintigraphic data. Maximalluminal diameter narrowing of each major coronary arteiy wasestimatedvisually and differencesin interpretationwere resolvedbyconsensuswithathirdobserver.A luminaldiameternarrowinggreaterthan50%was consideredsignificant.

Left ventricularwall motionwas assessedqualitativelyby dividing the left ventricle into anterior, apical and inferior regions inthe RAO3Oview and the septal, inferoapical and posterolateralregionsin the LAO6Oview. Wallmotionwas scoredas normal,hypokinetic, akinetic or dyskinetic.

Data AnalysisFDG images were projected on a monitor screen without back

ground correction or smoothing for visual interpretationby two

experiencedobservers.Scintigramswerevisuallyscoredas good,moderate or poor. Anterior views were used to draw regions ofinterest(ROI)over myocardiumwith qualitativelynormal @°@Tland FDG uptake. Scintigraphicinformationcombined with datafrom coronary angiographywas used to assure that ROIs weredrawn in flow areas of normal coronary arteries when these angiographicdatawere available In addition,ROIswere drawnoverthelungareaadjacentto theleftventricleandovertheliverarea. The numbersof pixelsper myocardial,lungand liver ROIswere100,425and115,respectivelyinallpatients.Heart/hingandheart/liverratiosfor @°‘Tland FDOwere calculatedfromthecountsperpixelin theROTs.

Quantftatlve Analysis of Myocardlal Thallium andFDG lm@ee

The imageswere quantitativelyanalyzedwith an operatorinteractive computer program. This program involved the following steps A circle was drawnaroundthe left ventricle.Theinterpolative background subtraction technique as described byGoriset al. (11)and modifiedby Watsonet al. (12)usingaproximity weighting function was applied. After background subtraction, nine-point image smoothing was performed. The leftventricle was divided into 60 radii spaced 6°apart, originatingfromthecenterof theventricle.Alongeachradius,thepixelwithmaximum 201Tl or FDO activity was chosen and plotted as thenumber of counts versus angle. The combinationof all 60 radiiresultedin a circumferentialprofileof myocardialuptake. Profileswere then aligned such that the 90°radius in each view corresponded to the visually located apex. The outflow tract radii,chosen between 204 and 330°(13), were excluded from furtherdata analysis. In addition, the left ventricle without the outflowtractwas divided into 13sectors of 18°.Of each myocardialview,the myocardialsector with the maximal @°@Tluptakewas used asthe individualreferencesector for the 2OFfland FDG study. FDGuptakewithinthissectorwasconsidered100%andFDGactivitiesof othersectorswere normalizedto this maximal @°@Tluptake.Thus, relativemyocardial @°@‘fland FDG uptake in sectors wereexpressed as a percentage of this reference region. The threeplanarimages were divided into a total of nine regions (Fig. 1).Averagerelative20111and FDG uptakefor these regionswerecalculated. Relative regional 201'fluptake was categorized intonormal( 75%of peakactivity),moderatelyreduced(50%—75%)andseverelyreduced(< 50%).Thesetracerdefectswereclassifled as viable and nonviable. Regions were considered viable ifFDG uptake exceeded @°@Tluptake by 20% or more or if FDGuptake was normal ( 75%). All regions with FDG uptake lessthan20%of 20―fluptakewereconsiderednonviable.

Imagingtwo tracers with differentphoton energiesmay resultin differentcircumferentialproffles. For this purpose, a phantomstudy was performed with an ellipsoidal sphere with an inner andouter wall. In the 1-cm space between these walls, a small compartment(2 x 2 cm)was introducedto create a tracerdefect. Boththe sphere andthe defect were filledwith @°‘Tland FDG activity.If the defectactivitywas less than 33%,the normalizedprofflesof20111and FDG differed less than 7% from each other. If the defectactivitywas more than 60%,the normalized @°@‘fland FDG proifies differed less than 2% from each other. Hence, a 20% criterionfor the differencebetween normalized @°@Tland FDG profilescansafetybe used for the detectionof myocardialviability.

976 The Journalof NudearMedicine•Vol.36 •No.6 •June 1995

FIGURE1. Myocardialregionsusedforthequantitativeanaly&softhesdntigrams.ANT= anterio@LA045= leftanterioroblique,450;LAO7O= leftanterioroblique,70°.

Comparison of LeftVentricularWallMotionwithSdntlgraphlc Data

For correlationof wall motionandscintigraphicdata,correspondingmyocardialregionswerecompared.Theanteriorscintigramswere comparedwith the RAO3Oventriculogramsand theLA045 scintigramswith the LAO6Oventriculograms(14,15).

Statistical AnalysisValues are expressed as mean ±standarddeviation. Statistical

analysis was performedwith the Student's t-test for paired andunpaired data. A p value < 0.05 was considered statisticallysignificant.

RESULTS

Image QualftyVisually, the image qualityof 27 (53%)FDG studies was

good, 17(33%)were moderate and 5 (10%)were poor. Twoof the five poor FDG studies were of such bad quality thatthey were uninterpretableand were excluded from furtherdata analysis. Heart-to-lung ratios of 201'flimages were2.37 ±0.44 and of FDG, 2.45 ±0.39 (p = ns). Heart-toliver ratios were 1.06 ±0.19 and 1.78 ±0.34 (p < 0.001),respectively.

Regional AnalysisFigure 2 shows the relation of relative regional @°@Tland

FDG uptake (FDG = 0.6811 + 35.59; r = 0.63; n = 441).FDG uptake in normal myocardium was homogeneouslydistributed with a FDGfl°'Tlratio of 1.10 ±0.15. TheFDGf°'Tl ratio in abnormal myocardium was 1.30 ±0.27(p < 0.05 versus normal myocardium). Figure 3 shows aflow chart of the analyzed myocardial regions. Four hundred forty-one myocardial regions was analyzed for 201@fland FDG uptake:200 regions showed normal @°‘Tluptake,241 regions had reduced uptake, 191 had moderately reduced 20111uptake (50—75%)and 50 regions had severelyreduced uptake (< 50%).

Viability was preserved in 62% and 48% (p = ns) of

moderately reduced and severely reduced regions. A concordance between blood flow and metabolism was observed in 38% and 52%, respectively. The averageand FDG uptake in relation to the severity of reduction inflow is expressed in Table 1. Thalliumuptake differsin thenormal, moderately and severely reduced regions, but itcannot differentiate between viable and nonviable regionswithin severely reduced regions(43.1%versus 42.0%).

Coronary Anglography and RegionalWallMotionNine patients had single-vessel disease, seven patients

had double-vessel disease and six patients had triple-vesseldisease. The mean left ventricular ejection fraction was49% ±9%. Ninety myocardial areas showed normalwallmotion on ventriculography (Fig. 4). The averageuptake was 75.7% ±11.2%(p < 0.01 versus hypokineticand akinetic regions) and FDG uptakewas 86.2% ±13.0%

.1t125 .1@•.@-

liE@@ 25

“0 25 50 75 100 12520111uptake(%of p@k activity)

FiGURE 2. Scatterplotand regressionanalysis of relative @°iiand relativeFDG uptake In myocardlalregionswithnormalandreduced20111uptake.

977Planar MyocardialFDG Imaging•Huitinkat al.

ST INF

ANT LA045 LAO7O

AverageAverage20111uptake18FDGuptakeRegions

(%)(%)

normal hypokh*tlc akinticn—gO flz29 n@13

left ventricularwall motion

4 p=NS

I100

1@80

@ 60

.140

20

FiGURE4. Bargraphshowingrelative2OFfl@ •@=uptakeInrelationto leftventricularwallmotion.

Figure 5. Figure 6 shows an example of a mismatch between flow and metabolism.

Plasma Glucose and Insulin LevelsPlasma glucose and insulin levels were, respectively,

6.8 ±1.4 mmole/liter and 43.7 ±19.1 mU/liter.

DISCUSSION

We have described the use of planar myocardial FDGimaging in patients shortly after myocardial infarction using a rest 201Tland FDG protocol. Regional left ventricularfunction was correlated to regional @°@Tland FDG uptake.

PlanarFDG imagingyielded images of adequate qualityin most patients. Seventy-six percent of the patients hadnormal or relatively increased FDG uptake within the infarct area, which probablyindicates the presence of viablemyocardial tissue. This incidence is in line with previousFDG studies after myocardial infarction. Signs of residualischemia after infarction were reported in 55%-77% ofpatients (2,8,16,17).

Table 1 and Figure 3 show that FDG uptake can bepreserved (79.3%), especially in regions with moderatelyreduced 20111uptake. This is in contrast to FDG uptake inregions with severely reduced @°‘Tluptake (63.2%), but itis similar to the results of a study by Kalif et al. (18). Theyfound that a blood flow reduction of 50% did not changerelative glucose utilization in ischemic canine myocardiuin.In contrast, low flow conditions resulted in marked impairment of FDG uptake. Our findings also correlate with thestudy by Dilsizian et al. (19) who found that the magnitudeof @°“fluptake after reinjection could distinguish betweenviable and nonviable myocardium Mild-to-moderate @°@Tldefects may represent viable myocardium. With planarscintigraphy, we found a high incidence of residual tissueviability in ventricular regions with reduced flow and impaired function during the subacute phase of myocardialinfarction. These results indicate, as do previous FDGPET studies, that (1)viable myocardiumcan be characterized by a small decrease in 20@Tluptake combined with anormal FDG uptake, and (2) viable myocardium can be

FiGURE3. Rowchartoftheanalyzedmyocardialregions.

(p = ns versus hypokinetic regions; p < 0.01 versus aidnetic regions), respectively, in these areas.

Twenty-nine hypokinetic myocardial areas were observed. The average @°‘Tluptake was moderately reducedin these regions (65.8% ±12.9%, p = ns versus akineticregions), but FDG was comparatively increased (86.0% ±18.7%, p = nsversusakinetic regions).Twenty-four (83%)hypokinetic areas were considered viable using the viability definitions, and five regions (17%) were considerednonviable.

In the 13 areas with akinesia, the average @°‘Tluptakewas 65.7% ±16.9% and the FDG uptake was 78.4% ±17.1% (p = ns versus hypokinetic regions). Nine (69%)regions were viable using the definitions for viability andfour regions (31%)were nonviable. Viable myocardial regions were detected in 37 (76%)patients. Thirty patientswith ECG signs of 0-wave infarctionhad myocardialareaswith viability. An example of scintigraphic images of apatient with a match in @°“fland FDG uptake is shown in

TABLE IAverage Regional 20111and FDG Uptake in Regions with

Normal,ModeratelyRedUCed @0—75%)and SeverelyRedUCed (< 50%) Ffl Uptake

Normal83.6±5.592.2±11950%-75%T@algroup 64.5 ±7.379.3 ±14.4*Nonviable

61.1±6.965.4±7.3*Viable66.6 ±6.8@87.8 ±10.5<

50% Total group 42.5 ±6.1632 ±13.6Nonviable42.0±6252.8±8.1*Viable

43.1 ±6.074.5 ±8.2*p

< 0.01 vs. FDGuptakeinnormalregkns.tp< 0.001vs. @°i1uptakeinnonviablemoderatelyreducedregions.

978 The Journalof NudearMedicine•Vol.36 •No. 6 •June 1995

•201T1@FDG

AA

Ui

0.

UiN-I

0z

REGIONS REGIONS

B

0@@

.-

@ /

FIGURE 6@ Quantitativeanalyals (i@4and leftanterior ob@que,45°view (B) of 201T1(left) and FDG (right) sdntigram. A concordantreductioninthe inferioruptakeof @°@T1and FDGcan be observed.The relative enhanced septal uptake of FDG to @°@11exceeds thecriteria for viability.

coronary artery disease was shown by Eitzman et al. (21)and Tamaki et al. (22). Their findings suggested that patients with impaired'left ventricular function and metabolicevidence of myocardial ischemia appeared to have themost clinical benefit from a revascularization procedure.Whether these findings also apply to patients with recentmyocardial infarction has to be established.

Left Ventricular FunctionWilliams et al. (8) performed planar FDG imaging in

patients after myocardial infarction who were evaluatedbefore and afterrevascularization.0f46 fixed @°‘Tldefects(exercise 201T1and 24-hr redistribution imaging) preopera

tively, 30 defects demonstrated improvement in regionalexercise 20111 uptake or regional systolic function afterrevascularization. FDG@Tl mismatches were present in83%of these regions but were absent in the majorityof thefunctionally unimproved regions. Therefore, their datasuggest that viable myocardium can be identified usingplanar FDG scintigraphy. Our data indicate also that thedetection of viability with the planar approach if feasible.Most regions with normal left ventricular function had anormal 201T1 and FDG uptake. Hypokinetic regionsshowed a clear discordance between flow and metabolism,indicating the shift in energy production from free fattyacids to glucose. Three regions with a severely reduced

FIGURE 5. QuantItativeanalysis (A)and leftanterior oblique,70°@ñew(B)of 20111Qeft)and FDG (rIght@sontigram. The reduction Inboth @°@TIand FDG uptake Inthe anterior region of the leftventricleshowedconcordance(nonviabletissue).

characterized by a FDG/2°'Tl mismatch ( 20%) with a

more profound perfusion deficit and relative increase inFDG uptake. Nonviable myocardium is characterized by amatched FDG/201Tldefect.

Early ImagingEarly identification of viable but jeopardized myocar

dium would support aggressive therapy to salvage endangered myocardial tissue. We found a high incidence oftissue viability in the subacute phase of infarction. As wasshown by Schwaiger et al. (2), absence of residual tissuemetabolism shortly after infarction is associated with irreversible injury, while remaining of metabolic activity indicates regions with a variable outcome. Although polymor

phonuclear leukocytes accumulate in necrotic tissue asearly as 4 hr after reperfusion of ischemic canine myocardium, preserved FDG uptake in the area of infarction islikely to reflect the metabolism of viable myocytes and isnotjust an inflammatory reaction, as was shown by a studyof Wijns et al. (20). Ourresults show that FDG imagingtoassess viability during the subacute phase of infarction isfeasible; however, the need for complementaryrevascularization has to be determined.

Clinical RelevanceThe clinical relevance and prognostic importanceof the

detection ofmyocardial viability with FDG in patients with

Pianar Myocardlal FDG Imaging •Huitink at al. 979

Ui

0.

UiN-I

0z

B

- - 2OITL

— FDG

S

- - 2OITL

— FDG

20111uptake and a concordant low FDG uptake (< 50%)were akinetic at ventriculography.These regions probablyrepresent real scar tissue. The other akinetic myocardialregions had either a mild @°‘Tldefect or increased FDGuptake. Thus akinetic regions may represent scarred, hibernating or stunned myocardium. It supports previousfindings that evaluation of regional function shortly afterthe onset of infarction may fail to distinguish necrotic fromviable myocardium. Additional informationaboutviabilityfrom metabolic studies or from @°@Tlactivity are neededbefore proper clinical decisions can be made.

MethOdOIOgIC ConsiderationsFor viability assessment, myocardial FDG uptake

should be related to a blood flow tracer (4,17). For thispurpose, we used 20111because early 20111imaging reflectsblood flow distribution in the myocardium (23).

The metabolic tracer, FDG, competes with glucose forphosphoiylation by the enzyme hexokinase. In the myocardial cell, FDG is phosphoiylated and trapped becauseFDG-6-P04 c@1IIflotbe furthermetabolized (24,25). In ischemic myocardium, glucose becomes the major source ofenergy and may therefore lead to increased FDG uptakerelative to nonischemic myocardium. This is the pathophysiological basis for normalizationof FDG uptake to theregion revealing maximal @°‘Tluptake. Because @°@Tluptake depends on blood flow, oxygen consumption andmyocardialmass, the region that shows maximal @°@Tluptake represents the most “normal―myocardium.

Although planar 511-keV scintigraphy will not meet thestandards of PET in terms of resolution and quantitation,the application of planar imaging systems seems justified,because, as Tilhisch et al. (4) reported, only large myocardial areas with flow/metabolism mismatch improved aftersurgical revascularization.

Although overlap of myocardial activity from normalmyocardium may mask areas with ischemia or infarction,from the data of Williams et al. (8), it can be inferredthatviability can be demonstrated with planar imaging.

The fact that most myocardial areas with normal function had 20111as well as FDG uptake of 75% or moresupports the cut-off point of the lower limit of normal at75% and the definition ofviability. These criteria have beenpreviously used by others (26) and are also supported byour experience with 201'fl and FDG studies in 12 normalindividuals (unpublished data), although it is acknowledgedthat revascularization data demonstrating improved regional function are needed to confirm these findings. Seventeen percent of hypokinetic regionswas considered nonviable using the criteria for viability. This may have beencaused by qualitative visual analysis of the wall motiondata.

It has been shown that @°@Tlreinjection studies(19,27,28)can be performedto detectmyocardialtissueviability. Our findings suggest that rest imaging, withoutthe need of exercise-induced ischemia, may also identifyviability.

The lack of 20111rest-redistributiondata is a majorurnitationofour study; however, thiswas not the primarygoalof our study. In theory, many of the regions with reduced20111uptake but preserved FDG uptake would show @°@Tlredistribution if late 201'flimaging had been performed.Thus, the relative value of planar FDG scintigraphy cornpared to PET and compared to @°@Tlreinjection and restredistribution studies needs to be established. Also, theadditionalvalue of planar FDG scintigraphyto rest-redistributionimagingprotocols needs investigation.

CONCLUSION

Myocardial FDG imaging is feasible with a standardgamma camera in patients shortly after myocardial infarction. Uptake of FDG in myocardial regions with normal20111uptake and normal left ventricular function is relatively homogeneously distributed; in regions with reduced20111 uptake and abnormal function, it was significantlymore heterogeneous due to the presence of relatively increased FDG uptake to flow. A considerable number ofinfarctpatients (76%)had viable regions within the infarctarea. Because an earlier planar FDG study has indicatedthat increased FDG uptake is related to functional recovery after revascularization, further studies are justified toevaluate the clinical relevance of planarmyocardial FDGimaging.

Finally, because gamma cameras are used in the majority of community hospitals and because the physical halflife of ‘8Fallows offsite use of FDG, application in routineclinical practice may be possible for viability assessment,risk-stratification and prediction of functional recovery after myocardial infarction.

ACKNOWLEDGMENTS

The authorsthankHenry Greuter for the skilifullpreparationofthe FDO, and nuclear medicine technologists Mhjam Schelen,Suzettevan Balen, Annet Pierikand Ton de Vries for technicalassistance.

This study was presentedin part at the EuropeanAssociationofNuclear MedicineCongress, October 1993,Lausanne, Switzerland and at the Workshopon AlternativeImagingTechniques forFluorine-18-FDG, April 13—141994, Amsterdam, The Netherlands.

REFERENCES1. Camici P, Ferrannini E, Opie LP. Myocardial metabolismin ischemicheart

disease: basic principlesand applicationto imagingby positron emissiontomography.Flog CardiovascDis 1989;32:217—238.

2. Schwaiger M, Brunken R, Grover-McKay M, et al. Regional myocardialmetabolism in patients with acute myocardial infarction assessed bypositronemissiontomography.lAin CoilCardiol 1986;8:800-808.

3. Camici P, Araujo L, Spinks T, Ct al. Increased uptake of ‘Ffluordeoxyglucosein postischemicmyocardiumof patients with exerciseinduced angina. Circulation 1986;74:81—88.

4. TillischJ, MarshallR. Reversibilityof cardiacwallmotionabnormalitiespredictedby positrontomography.NEnglJMed 1986;314:884-888.

5. TamakiN, YonekuraY, YamashitaK, et al.Positronemissiontomographyusing fluorine-18deoxyglucosein evaluation of coronaiy artety bypassgrafting.AmI Cardiol 1989;64:860-865.

6. Hoeflin F, Ledermann H, Noelpp U, Weinreich R, Roesler H. Routine

980 The Journalof NuclearMedicine•Vol.36 •No. 6 •June 1995

‘8F-deoxy-2-fluoro-d-glucose(‘@FDG)myocardialtomographyusinga normal largefieldofview gammacamera.Angidogy 1989;140:1058—1064.

7. VanLingenA, HuijgensPC,VisserFC,Ctal.Performancecharacteristicsof a 511-keV collimator for imaging positron emitters with a standardgammacamera.EurlNud Med 1992;19:315—321.

8. WilliamsI(A, TaillonLA, StarkVJ. Quantitativeplanarimagingof glucosemetabolic activity in myocardial segments with exercise thallium-201 perfusiondefects in patientswith myocardialinfarction:comparisonwith late(24hr)redistributionthalliumimagingfordetectionofreversibleischemia.Am Hea,fJ 1992;124:294-304.

9. Gropler Ri, SiegelBA, Lee KJ, et al. Nonuniformity in myocardial accumulation offluorine-18-fluorodeoxyglucose in normal fasted humans.JNuclMed 1990;31:1749—1756.

10. Hicks RI, Herman WH, Kalif V, et al. Quantitative evaluation of regionalsubstratemetabolismin the humanheartby positronemissiontomography.JAm Coil Cardiol 1991;18:1O1—111.

11.GorisML, DaspitSO,McLaughlinP,KrissJP.Interpolativebackgroundsubstraction.JNuclMed 1976;17:744—747.

12.WatsonDD,CampbellNP,ReadEK,GibsonRS,TeatsCD,BelierGA.Spatial and temporal quantitation of plane thallium myocardial images. INuciMed 1981;22:577-584.

13. Garcia E, Maddahi J, Berman D, Waxman A. Space/time quantitation ofthaffium-201myocardialscintigraphy.INuci Med 1981;22:309-317

14. Gerwitz H, Beller GA, Strauss HW, et al. Transient defects of restingthaffiumscans in patientswith coronaiy arteiy disease. Circulation1979;59:707—713.

15. Berger BC, Watson DD, Burwell LR, et al. Redistribution ofthallium at restin patients with stable and unstable angina and the effect of coronary arteiybypass surgery. Circulation 1979;60:1114-1125.

16. MarshallRC,TiIIiSChJH,PhelpsME,etal. Identificationanddifferentiationof restingmyocardialischemiaandinfarctionin manwith positroncomputedtomography,‘8F-labeledfluorodeoxyglucoseandN-13ammonia.Circulation 1983;67:766—777.

17. BrunkenR, TillishJ, SchwaigerM, et al. Regionalperfusion,glucosemetabolismand wall motion in patientswith chronicelectrocardiographic0-wave infarctions:evidenceforpersistenceofviable tissuein someinfarctregions by positron emission tomography. Circulation 1986;73:951—963.

18. KaIf V, SchwaigerM, NguyenN, McQanahanTB, GallagherKP. Therelationship between myocardial blood flow and glucose uptake in ischemic

caninemyocardiumdeterminedwithfluotine-18-deoxyglucoseJNuclMed1992;33:1346—1353.

19. DilsizianV, FreedmanNMT, BacharachSL, Perrone-FilardiP, BonowRO.Regionalthalliumuptakein irreversibledefects.Magnitudeof changein thallium activity after reinjection distinguishes viable from nonviablemyocardium. Cfrculation 1992;85:627-634.

20.WijnsW, MelinJA,LenersN, etal.Accumulationofpolymorphonuclearleukocytesin reperfusedischemiccaninemyocardium:relationwith tissueviabilityassessedby fiuorine-18-2-deoxyglucoseuptake.INuci Med 1988;29:1826—1832.

21. EitzmanD,Al-AouarZ,KanterHL,etal.Clinicaloutcomeofpatientswithadvancedcomnaiy arteiy diseaseafterviabilitystudieswithpositronemissiontomography.JAm CoilCardiol1992;20:559-565.

22. Tamaki N, Kawamoto M, Takahashi N, Ct al. Prognostic value of anincrease in fluorine-18deoxyglucoseuptake in patients with myocardialinfarction:comparisonwith stress thallium imaging.I Am Coil Cardiol1993;22:1621—1627.

23. Maddahi J, Garcia EV, Berman DS, Waxman A. Swan HJC, Forrester JS.Improvednoninvasiveassessmentof coronary arteiy diseaseby quantitative analysis of regional stress myocardial distribution and washout ofthaffium-201.Circulation1981;64:924-931.

24. Phelps ME, Hoffman El, 5dm C, et al. Investigation of [‘8F]2-fluoro-2-deoxyglucosefor the measureof myocardialglucosemetabolism.I NuciMed 1978;19:1311—1319.

25. Gallagher BM, FowleriS, Gutterson NI, MacGregor RR, Wan C, Wolf AP.Metabolictrappingas a principleof radiopharmaceuticaldesign:somefectors responsible for the biodistribution of [“F]2-deoxy-2-fluoro-D-glucose.INuciMed 1978;19:1154—1161.

26. AltehoeferC, vom DaMJ, Bud U, Uebis R, KleinhansE, HanrathP.Comparisonof thallium-201single-photonemissiontomographyafter restinjectionandfluorodeoxygiucosepositronemissiontomographyfor assessmentof myocardialViabilityin patientswithchroniccoronaiyartezydisease. EurlNuciMed 1994;21:37-45.

27. BonowRO,DilsizianV.CuocoloA,BacharachSL.Identificationofviablemyocardium in patients with chronic coronaiy artery disease and left yentricular dysfunction. Comparison of thallium scintigraphy with reinjectionandPETimagingwith‘F.fluorodecxyglucose.Cüvulation1991;83:26-37.

28. Dilsizian V, Rocco TP, Freedman NMT, Leon MB, Bonow RO. Enhanceddetection of ischemic but viable myocardium by the reinjection of thaffiumafter stress-redistributionimaging.N EngilMed 1990;323:141-146.

981Planar MyocardialFDG Imaging•Huitinkat al.