(dysnomia) is the most common and early clinical presentation of PPA (2). Other language impairment, such as readingand writing, decreased comprehension of spoken word, andpoverty of words, may also be present (2). Dementing illnesses,such as Pick's disease and Alzheimer's disease (AD), maypresent with language impairment;however, patients with ADusually show decreases in comprehension and written expression to a greater extent than they show “word-finding―difficulty (3). Additionally, spontaneous speech is significanfly more impaired in patients with PPA in comparisonwith patients with AD. Aphasia in PPA tends to affectanterior parts of the language-dominant cortex (4). However, at the late stages there is an overlap of symptoms, andthe clinical distinctions between AD, Pick's disease, andPPA are clinically less clear.

Neuroimaging studies using MR and CT have shownselective atrophy in the left perisylvian region in mostpatients with PPA (5). Previous studies using FDG PET and

@‘@‘Tc-hexamethylpropyleneamine oxime (HMPAO) SPECTshowed frontal and temporal lobe abnormalities beforechanges were apparent on MRI (6—8).However, none ofthese studies performed correlational analysis between cliical dysnomia and the degree of functional deficit on FDGPET or @Tc-HMPAOSPECT. The objective of this studywas to determine whether there is a relationship between thedegree of word-finding difficulty in PPA and the degree ofdecreased frontotemporal @°Tc-HMPAOuptake using semiquantitative brain SPECT.

MATERIALSAND METHODS

StudyPopulationSeven right-handedpatients(4 men, 3 women; age range59—77

y; mean age, 65.6 y) diagnosed with PPA were referredto theDivision of Nuclear Medicine for @“Tc-HMPAObrainSPECTtoassess disease severity.Thedurationofillness atthetimeof SPECTimaging was 2—7y (mean, 2.9 y). The patients presented withvarying degrees of dysnomia (Table 1). All patients satisfied the

criteriafor PPA, i.e., language impairmentwithout accompanyingcognitive disorder in at least the first 2 y of disease (1).

BostonNamingTestAll subjects were evaluated for dysnomia using the Boston

Naming Test (BNT), which is a part of the Boston DiagnosticAphasia Examination(9). It is a visual confrontationnaming task

Primary progressive aphasia (PPA) is an uncommon degenerafivedementiacharacterizedbygradualimpairmentoflanguagefunction with initial sparing of the memory domain. Using semiquantitative @“Tc-hexamethylpropyleneamineoxime (HMPAO)brainSPECTas a measureof regionalcerebralbloodflow(rCBF),we investigatedthe relationshipbetween reducedtomTc@HMPAOuptakeandtheseverityofdysnomiain PPA.Methods:Sevenright-handedpatientswithPPAhadtheirdysnomiaassessedby the BostonNamingTest(BNT),a subtestof theBoston Diagnostic Aphasia Examination. Neuroimaging studies,including @Tc-HMPAObrainSPECT,CT,and MRI,wereperformed.CorrelationalanalysisbetweenreducedrCBFand BNTwasperformed.Results:BrainSPECTshoweda reductionin

@Tc-HMPAOuptakeinvolvingthefrontalandtemporallobesinall 7 patients.CT and MRI showedmildto moderatecerebralatrophyin4 patients.LowscoresontheBNTcorrelatedwithlowfrontotemporal@°“Tc-HMPAO(Spearmanr = 0.97,P = 0.004)inthe 5 patientswith left-hemisphereinvolvement.Conclusion:DecreasedrCBFtothefrontotemporalregioncharacterizedthecerebralabnormalitiesassociatedwithPPA.ThefindingoffocalrCBF abnormalities in the right hemisphere of 2 right-handedwomen corroboratesthat PPA symptomsmay arise from a“non—left-dominant―-hemispheredegenerativeprocess.Our resuits support the usefulness of rCBF SPECT imaging as adiagnosticaidinPPA.KeyWords:progressiveaphasia;@‘Tc-HMPAObrainSPECT;BostonNamingTest

J NucIMed2000;41:228-233

rimary progressive aphasia (PPA) is an uncommon typeof degenerative dementia characterized by gradual impairment of language function that remains neuropsychologically focal for several years with sparing of the memorydomain (1). Compared with other neurodegenerative disorders, which initially affect cognition followed by languageimpairment, many patients with PPA retain their cognitivefunctions, allowing them to continue with their activities ofdaily living (1). “Word-finding―or “naming―difficulty

ReceivedOct.13,1998;revisionacceptedJul.21, 1999.For correspondenceor reprintscontact:James M. Mountz,MD, PhD,

University of Alabama at Birmingham Medical Center, DMsion of NuclearMedicine,619 S. 19thSt., JTJ26O, Birmingham,AL35233-6835.

228 THE Joui@.r'@ OF NUCLEARMEDICINE •Vol. 41 •No. 2 •February 2000

Frontotemporal Decreases in rCBF Correlatewith Degree of Dysnomia in PrimaryProgressive AphasiaElmer C. San Pedro, Georg Deutsch, Hong Gang Liu, and James M. Mountz

Department ofRadiology, Division ofNuclear Medicine, University ofAlabama at Birmingham Medical Center@Birmingham, Alabama

by on February 25, 2018. For personal use only. jnm.snmjournals.org Downloaded from

PatientAgeDurationMRIno.(y)Sex of PPA*(y) PresentationBNT score or CT

TABLEIClinicalData on 7 PatientswithPPA

234567

Mean

59M371M260F277M265M264F763F265.572.9DysnomiaDysnomia,I speechfluencyDysnomiaDysnomia,@ speech fluencyDysnomia,I speechfluencyTrippingoverwords,jspeechfluencyDysnomia,I speechfluency

54 CT, atrophy35 CT,normal30 CT, normal18 MRI,atrophy14 CT,atrophy50 MRI,atrophy45 MRI,normal

*Durationof PPAindicatestimeintervalbetweenoccurrenceoffirstsymptomstodiagnosis.Patientspresentedwithvaryingdegreesofdysnomia.

andis scoredaccordingto a standaniprocedure,whereinsubjectsareaskedto nameline drawingsof commonobjects(9). The maximumscoreon the BNT is 60. In this study,dysnomiais consideredmild if theBNT score is 40-60, moderateif3O-40, and severeif <30.

NeuropsychologicalTestingNeuropsychologicalassessmentincludedAphasiaSeries,Apraxia

Series, Beck Depression Inventory,Dementia Rating Scale (Mattis), Weschler Adult Intelligence Scale—Revised(WAIS-R), andWeschlerMemoryScale—Revised.

CTandMRINon—contrast-enhancedCT was performedon 4 patients(Table

1) using a GE CT! scanner (General Electric, Milwaukee, WI).MRI was performedon the remaining3 patients(Table 1) using aGE SignaAdvantage(GeneralElectric) l.5-T scanner.

@‘Tc-HMPAOBrainSPECTAll 7 subjectsunderwent @Tc-HMPAObrainSPECTto assess

regionalcerebralblood flow (rCBF).Approximately740 MBq (20mCi) @mTc@HMPAOwere injected intravenously,with patientsintherestingstatewitheyes closed in a dim,quietroom.Imagingwasperformed on the Picker Prism 3000XP (Picker International,Bedford, OH) triple-head y camera, equipped with low-energy,high-resolutioncollimators yielding an image resolution of approximately 7 mm full width at half maximum.The matrix size was 128 X128 (pixel size, 2.0 mm on edge).Acquisitionparameterswere 120°rotation, 40 stops/head at 45 s/stop (120 total projections). Imageattenuationcorrectionwas performedusing the Changmethod,andreconstnictionwas performedusing a Butterwoithfilterwith frequencycutoffof 0.225andNyquistorderof 6 (10).Transversesectionswerereconstructed parallel to and sequentially above the canthomeatal line(CM) as defined by the Mountz reference system (Hanison Medical,Helena,AL)routinelyusedintheDivisionofNuclearMedicine(11).

rCBFSPECTSemiquantitativeAnalysisRegion oflnterest Definition. Because the pathologic abnormal

ity in PPA predominantly affects the perisylvian regions to a greaterextent than other areas of the brain, semiquantificationwas performedon CM line + 55 cm (2). This level encompasses importantcortical(Brodman)areasof thebrainresponsibleforlanguagefunctionthatareknown to be involvedin early PPA(Fig. 1A).Regionof interest(ROl)count thta from the petisylvian territory (frontotemporal region) reptesented by regions 2-4 (left hemisphere) and 9-1 1 (tight hemisphere)wereobtainedand are shownin Table2.

Cortical Circumferential Profile. Cortical circumferential profiles were obtainedby delineatingan annularringof cortex 1.6 cm(8 pixels) wide using a computer-automated algorithm. The outerboundaryof this annuluswas definedby a pixel thresholdof 50%of the average pixel value for the entire section under analysis.Individual cortical ROIs were created by subdividing this annulusinto 12 equal angularsectors (Fig. 1C). The rCBF value for eachROIwas calculatedby a normalizationprocedure,in an which thetotal pixel counts in each ROI were divided by the average of thetop 90%of the cerebellarpixel counts (Fig. 1B) andcomparedto 9age- and sex-matchedhealthycontrolsanalyzedin an identicalmanner(12,13).Wehaveshownthatthemajorlobarregionsofthe braincanbeassessedusingthismethodofROI characterization(14).

On section CM + 5.5 cm, regions 1—3represent the left frontallobe and regions 10—12represent the right frontal lobe. Region 4representsthe left superiortemporal lobe and region 9, the rightsuperior temporal lobe. Regions 5-6 and 7-8 represent the occipitallobes in the left and right hemispheres, respectively (Fig. IC) (14).Regions 5—6and 7—8,which representthe left and right occipitalregions, were not included in the analysis, because this area is rarelyinvolvedin PPA(2). Regions 1and 12werealsoexcludedbecausetheydo not correspond to known Bmdman language areas (Fig. lA).

Because the typical presentation of PPA involves the languagedominant left hemisphere, statistical analysis was performed onlyon a subgroupof 5 patientswith left frontotemporaldiminutionof

@Tc-HMPAOuptake (Table 2).

RESULTSBostonNamingTestandNeuropsychologicalTesting

Table 1 shows the results of the 60-item confrontationnaming test; a score of <45 is indicative of significantdysnomia (15). Our patient population had mild (n = 3),moderate (n = 2), and severe (n = 2) degrees of dysnomia(16). Five of the 7 patients had nonfluent aphasias. None of

the patients showed depression on the Beck Depression Inventory. None of the patients showed memory or cognitive declineon Maths, WAIS-R (selected subtests), or Weschler MemoryScaler-Revised.

CTand MRIThree patients (Table I) had normal CT or MRI scans for

age. Two of the 4 patients (Table 1) who underwent CT had

Bi@m@SPECT w@PRooiu@ssivE APHAsIA •San Pedro et al. 229

by on February 25, 2018. For personal use only. jnm.snmjournals.org Downloaded from

Patientno.Left

hemisphereAOlRighthemisphereROl23 4 Mean9 1011 Mean

Patientmeanwascalculatedonlyforpatients1—5,whoshowedunilateralleftfrontotemporaldiminutionon @Tc-HMPAOSPECT.Patients6and7wereexcludedbecausetheyshowedrighthemisphericandbilateralinvolvement,respectively.Meancountratiosforfirst5patientsarelowerthanage-matchednormalcontrolsanalyzedinanidenticalmanner(13).

A B

R@ 6 L

FIGURE I . (A) LeftlateralbraindiagramwfthBrodmanareasconsideredto have languagefunction,showingextentof corticalinvolvement(asterisks)inprogressiveaphasia.AreasurroundingperisylvianregionextendingfromfrontaltotemporallobeshasbeenreportedtobeinvolvedinPPA.Diagramalsoillustrateslevelofsection,5.5cmaboveCM(CM+ 5.5cm),wheresemiquantificationwasperformed.Sectionispositionedatcenterofbrainandallowsmeasurementofcountdatainfrontallobe,perisyManregion,andtemporallobe.(B)ExampleofcerebeilarAOlusedforcountnormalization.(C)TwelvecorticalROIs,clockwise(1—12),usedtodividetransverseSPECTslice.ROls1—2representleft frontallobe;region3 representsfrontotemporalcortex;region4 representslefttemporallobe.Regions5 and6 representoccipitallobe,whichisrarelyreportedtobe involvedinPPA.

mild to moderate atrophy. Two of the 3 patientswho underwent MRI were also found to have mild atrophy commensurate for age (Table 1). A CT scan of a59-y-old man (Table 1) with progressive aphasia is shown inFigure 2. The CT scan showed mild age-related cerebralatrophy.

Semlquantftative rCBF SPECTPatients 1—5(Table 2) represent a subgroup who showed

left-hemispheric hypoperfusion on SPECT. Cortical-tocerebellar count ratios for regions 2—4(left perisylvian) and9—11(right perisylvian) are shown. Reduction in @‘@Tc

HMPAO uptake in the left perisylvian, frontal, and temporallobe regions was seen in this subgroup. Figure 2 is arepresentative @Tc-HMPAOSPECT scan.

Figure 3 (Table 2) shows an atypical right-hemispherichypoperfusion of the frontal and temporal region in PPA.The patient is a 64-y-old schoolteacher who had PPA for 7 ywith gradual decrease in fluency but with intact cognitivefunction (Mini-Mental State Examination [MMSE] score,28/30). The brain SPECT shows hypoperfusion to the rightfrontal, perisylvian, and temporal lobe regions.

In the 3 patients (Table 1) who had normal CT or MRI

TABLE2Cortical-to-CerebellarROlCountRatiosin 7 Patientswith ProgressiveAphasiaObtainedat CM + 5.5cm

10.780.770.820.790.810.830.800.8120.740.710.800.750.840.860.960.8930.720.840.700.750.820.830.810.8240.740.670.740.720.850.870.860.8650.730.660.710.700.860.830.810.8360.840.820.860.840.670.740.720.7170.820.790.690.770.680.730.760.72

Patientmean±1SD0.76 ±0.020.75 ±0.070.75 ±0.060.84 ±0.020.84 ±0.020.85 ±0.07Controlmean ±1 SD0.88 ±0.070.88 ±0.060.85 ±0.030.87 ±0.080.90 ±0.090.88 ±0.09

230 THEJoui@i. OFNUCLEARMEDICINE•Vol. 41 •No. 2 â€F̃ebruary 2000

by on February 25, 2018. For personal use only. jnm.snmjournals.org Downloaded from

COMPUTEDTOMOGRAPHY

Tc-99m HMPAOBrain SPECT

FIGURE2. CT (A) and @mTc-HMPAObrain SPECT scans (B) of 59-y-oldmanwithmilddysnomia(patient1; Table1).Patient presentedwith word-findingdifficultyof 2-y duration.Hehasnocomprehensionproblemwithwrittenor spokenwords;however,he reportsthat “wordscannotcome out―when he tries to speak. Hisneurologicalandcognitiveexaminationsarenormal,and he is able to perform activitiesofdailylivingwithoutassistance.Hescored54 on BNT. Reduction in rCBFto left pensyIvian,frontal,and temporallobes(arrows)characteristic ofprogressive aphasia is identified on @Tc-HMPAOSPECT. CT scanshowsmildatrophy.

scans for age, there was decreased @mTc@HMPAOuptake onbrain SPECT (Table 2).

Figure 4 (Table 1) exemplifies the reduction in @°@TcHMPAO uptake accompanied by a normal MRI. The patientis a 63-y-old woman with mild dysnomia (BNT = 45). Shepresented with increasing difficulty in saying certain words.She had a tendency to hesitate when speaking and haddifficulty writing, particularly when spelling multisyllablewords.

BrainrCBFSPECTandBNTCorrelations between rCBF ratios and BNT scores were

performed for the 5 patients with unilateral left-hemisphericinvolvement. Spearman rank correlation test demonstrated aparallel relation between mean left rCBF count ratio (Table

2) and the severity of dysnomia (Table I) (r = 0.97; P =0.004).

DISCUSSION

Word-finding difficulty or dysnomia is the failure toretrieve a specific word to express oneself in an ongoingconversation. It is the earliest and most common presentationofprogressive aphasia.As the diseaseprogresses,characteristic language impairment, such as impaired fluency, disintegration of syntax, and phonemic paraphasias, emerge (4,17).Left-hemispheric involvement in PPA is well documentedby neuroimaging and histopathological studies in accordwith the fact that language function is primarily localized inthe left hemisphere among right-handed individuals.

MRI Tc—99mHMPAOSPECT

FIGURE3. MR image(A) and @mTcHMPAObrainSPECT scan(B) of 65-y-oldwomanwithmilddegreeof dysnomia(patient 6; Table1). She presentedwith 7-yhistoryof “tripping―overwords. Initially,shehad troublewith multisyllablewords butcurrentlyhas difficultyeven with singlesyllablewords.She reportsproblemswithdecreasingfluency.She claims to knowwhatshewantstosaybutis“notabletogetthewordsout.―Sheoncedplayedthepianoand sang with accompanimentbut latelyreports loss of interest because of herinabilityto get right tune. She scored 30 of30 on MMSE. Right frontotemporal regionshowsatrophyonMRimageandreducedrCBF on SPECT. There is larger area ofrightfrontotemporalhypopertusiononrCBFSPECT (lowerarrow)than atrophyshownonMR image.

BRAIN SPECT IN PROGRESSIVE APHASIA •San Pedro et al. 231

by on February 25, 2018. For personal use only. jnm.snmjournals.org Downloaded from

MRI Tc-99m HMPAOB SPECT

FIGURE4. MR image(A) and @°TcHMPAOSPECTscan(B)of63-y-oldwomanwith mild dysnomia(BNT = 45). Patientpresentedwithhistoryofslowlyprogressivedifficultywith expressivelanguage, withdifficuityfindingnounsandproducingwordsforpast2y.Shereportsincreasingdifficultysayingcertainwords.She hastendencytohesitatewhen speakingand has difficultywriting,particularlywhenspellingmultisyllablewords.Batteryof neuropsychologicalexaminationsfailedto showany cognitiveimpairmentexcept for languagedeficits.SPECT showsbilateraltemporallobeandrightposteriorfrontalreduction(arrows)in99mTcHMPAQuptake.MR imageisnormal.

LanguageImpairmentand @“Tc-HMPAOSPECTIn this series of 7 right-handed patients, 5 showed

left-hemispheric reduction in @Tc-HMPAO uptake, 1showed unilateral right-hemispheric reduction, and 1 showedbilateral temporal lobe and right frontotemporal reduction.

Among the patients with left-hemispheric involvement,the strong association between scores on the confrontationnaming test and the reduction of @“@Tc-HMPAOuptake onSPECT indicates that the focal rCBF abnormalities observedindeed reflect the degenerative process underlying theprogressing aphasia. @Tc-HMPAOSPECT may be usefulin diagnosing ambiguous cases and in following the progression of the disease. Involvement of the frontal lobe regionhas been shown to accompany PPA. In this study, reductionin @°@Tc-HMPAOuptake in this region correlates with lowBNT scores.

Language impairment arising from right-hemispheric degeneration in a right hander is very unusual. Patient 6 mayrepresent a case of right-hemisphere language dominance ina right hander, a relatively rare situation with less than 5%incidence (18). Aphasia arising from bilateral brain injury isclearly not unexpected; the presence of focal bilateraldegeneration resulting in primarily aphasic symptoms inpatient 7 is atypical of most previously reported cases ofPPA. The finding of focal rCBF abnormalities in the righthemisphere of 2 right-handed patients suggests that PPAsymptoms may arise from other than a left-hemisphere(language-dominant) degenerative process.

It should be noted that both the right (patient 6) andbilateral (patient 7) PPA cases reported here are women,raising the possibility that the language deficit in these casesis,at least,in part,explainedbythegreaterbilateralrepresentation of language in women (15). More recentstudies using the Wada amobarbital technique suggest thatmany previously reported cases of right-hemisphere speechactually represent cases of bilateral speech organization(19). Patient 6 may have extensive language representation

and this may also explain her relatively mild languagedeficit, despite the fact that she had the disease for 7 y.

The more severe language impairment of patient 7 arisingfrom less severe but bilateral hemispheric rCBF deficits is

consistent with bilateral language representation amongwomen.

A recent study using stereological volume measurementshas shown that women have proportionally larger languageassociated regions (Wernicke's and Broca's areas) comparedwith men (20). This may offer an alternative explanation asto why men are more affected by PPA than women.

The 99mTc44Lr@ftJ@AOSPECT study showed cortical abnormalities of various degrees around the perisylvian region inall patients. Structural imaging with CT or MRI, however,identified only 4 patients with some degree of atrophy.

@‘°Tc-HMPAOSPECT is more sensitive than MRI or CT incharacterizing the cortical abnormalities in PPA. The cortical abnormalities observed on SPECT in cases in which CTor MR images are normal or just mildly atrophic probablyrepresent early neurometabolic defects involving the frontaland temporal (perisylvian) regions in this disease.

rCBF is related to neuronal activity, whereas brainatrophy is the result of neuronal loss. In Figure 4 there isreduction of rCBF in areas in which there is no cerebral

atrophy. In Figure 2 there is very mild atrophy on CT.However, there is significant reduction of rCBF to the leftside of the brain. Finally, in Figure 3, there is greaterhypoperfusion to the right posterior temporal lobe than theatrophy shown on MRI. These cases suggest that there maybe a gradual progression of cerebral changes in PPA from adecrease in neuronal activity (as indicated by a decrease inrCBF), without accompanying cerebral atrophy, to a moreadvanced stage involving cerebral atrophy. Because 99mTc@HMPAO SPECT is a sensitive indicator of both decreasedneuronal activity and atrophy, it has the capability for earlydisease detection.

232 Tm@JOURNALOF NUCLEARMEDICINE•Vol. 41 •No. 2 •February 2000

by on February 25, 2018. For personal use only. jnm.snmjournals.org Downloaded from

6. ChawluklB. MesulamMM.HurtigH,etal.Slowlyprogressiveaphasiawithoutgeneralized dementia: studies with positron emission tomography. Ann Neural.1986;19:68—74.

7. Delecluse F, Andersen AR, Waldemar 0, et al. Cerebral blood flow in progressiveaphasia without dementia. Brain. l990;ll3:l395—l404.

8. McDaniel KD, Wagner MT. Greenspan BS. The role of brain single photonemission computed tomography in the diagnosis of primary progressive aphasia.Arch Neural. 199l;48:1257—l260.

9. GoodglassH, KaplanE. TheAssessmentof Aphasia.Philadelphia,PA:Lea&Febiger; 1983.

10. Liu HG, Harris JM, Inampudi C. Mountz JM. Optimal reconstructionfilterparameters for multi-headed brain SPECT: dependence on count activity. J NucIMed Technol.1995;23:251—254.

11. Mountz JM. Wilson MW, Wolff CG. Deutsch G, Harris JM. Validation of areference method for correlation ofanatomic and functional brain images. ComputMedimaging Graph. 1994:18:163—174.

12. Liu HG, Mountz JM, lnampudi C, San Pedro EC, Deutsch G. A semiquantitativecortical circumferential normalization method for clinical evaluation of ICBF

brain SPECT. Clin Nucl Med. 1997;22:596-604.13. DeutschG, MountzJM,KatholiCR,LiuHG,HarrellLE.Regionalstabilityof

cerebral blood flow measured by repeated technetium-99m-HMPAO 5PECT:implications for the study of state-dependent change. J NucI Med. l997;38:6—l3.

14. MountzJM,TolbertLC,LillDW.KatholiCR,LiuHG.Functionaldeficitsinautistic disorder: characterization by technetium-99m HMPAO and SPECF. JNuclMed. 1995:36:1156—1162.

15. Springer SP, Deutsch G. Left Brain, Right Brain: Perspectives from CognitiveNeuroscience. 5th ed. New York, NY: WH Freeman; 1998.

16. WelchLW,DoineauD, Johnson5, KingD. Educationalandgendernormativedata for the Boston Naming Test in a group of older adults. Brain Lang.1996;53:260—266.

17. Beland R, Ska B. Interaction between verbal and gestural language in progressiveaphasia: a longitudinal case study. Brain Lang. 1992:43:355—385.

18. Rasmussen 1, Milner B. The role of early left.brain injury in determininglateralization of cerebral speech functions. In: Dimond S. Blizzard D. eds.Evolutionand Lateralizationofthe Brain.New York, NY: New YorkAcademy ofSciences; 1977.

19. LoringDW.MeadorK,LeeG,et al.Cerebrallanguagelateralization:evidencefrom intracarotid amobarbital testing. Neuropsychologia. 1990;28:83l—838.

20. HarastyJ,DoubleKL, HalIidayGM, Kril ii, McRitchieDA. Language-associatedcortical regions are proportionally larger in the female brain. Arch Neural.1997:54:171—176.

Bi@mr SPECT m@PROGRESSIVEAPHASIA•San Pedro et al. 233

CONCLUSION

@ SPECT may be useful in evaluatingpatients with progressive language impairment in whomearly PPA or another dementia is being considered. Sincelanguage impairment can be a predominant sign in variousforms of dementia, rCBF SPECT is useful in characterizingthe blood flow abnormalities that can distinguish differenttypes of dementias. Clinical dysnomia may have a neuropathologic correlate in the frontotemporal lobe region, theseverity of which may be evaluated with semiquantitativerCBF SPECT. In addition, our data show that PPA maydemonstrate rCBF abnormalities not only in the languagedominant left hemisphere but may also involve the righthemisphere and can be bilateral.

ACKNOWLEDGMENTS

This work was supported in part by National Institutes ofHealth grants RO1-AG 06432 and ROl-HD 32100. TheauthorsthankEva M. Gilliam for expertsecretarialassistance and Anthony Zagar for the illustrations.

REFERENCESI . Mesulam MM, Johnson N, Grujic Z, Weintraub S. Apolipoprotein genotypes in

primary progressive aphasia. Neurology. 1997;49:51—55.2. WestburyC, Bub D. Primaryprogressiveaphasia:a reviewof 112cases.Brain

Lang. 1997:60:381—406.3. Pogacar S. Williams RS. Alzheimer's disease presenting as slowly progressive

aphasia. Rhode Island Med J. l984;67: I81—185.4. Karbe H, Kertesz A, Polk M. Profiles of language impairment in primary

progressive aphasia. Arch Neumi. 1993;50:193—201.5. Chawluk lB. Alavi A. Neuroimaging of normal brain aging and dementia. In:

Greenberg JO, ed. Neumimaging: A Companion to Adonis and Victor c PrinciplesofNeumlogy. New York, NY: McGraw-Hill; 1995:235—382.

by on February 25, 2018. For personal use only. jnm.snmjournals.org Downloaded from

2000;41:228-233.J Nucl Med.   Elmer C. San Pedro, Georg Deutsch, Hong Gang Liu and James M. Mountz  Progressive AphasiaFrontotemporal Decreases in rCBF Correlate with Degree of Dysnomia in Primary

http://jnm.snmjournals.org/content/41/2/228This article and updated information are available at:

  http://jnm.snmjournals.org/site/subscriptions/online.xhtml

Information about subscriptions to JNM can be found at:  

http://jnm.snmjournals.org/site/misc/permission.xhtmlInformation about reproducing figures, tables, or other portions of this article can be found online at:

(Print ISSN: 0161-5505, Online ISSN: 2159-662X)1850 Samuel Morse Drive, Reston, VA 20190.SNMMI | Society of Nuclear Medicine and Molecular Imaging

is published monthly.The Journal of Nuclear Medicine

© Copyright 2000 SNMMI; all rights reserved.

by on February 25, 2018. For personal use only. jnm.snmjournals.org Downloaded from