Technetium HMPAO SPECT Study in Dementia with Lewy Bodies, Alzheimer's Disease and Idiopathic Parkinson's Disease

rent visual hallucinations, systematized delusions and spontaneous parkinsonian symptoms. Repeated falls, syncope,transient loss of consciousness and neuroleptic sensitivityare also useful clinical characteristics. In AD, the progressive decline ofmemory is prominent whereas neuropsychiattic features usually occur in the later stages (6). Moderateparkinsonian signs can be observed during AD evolution, asin DLB evolution (7). In some cases, the clinical distinctionof patients with DLB and those with AD may be difficultbecause of overlapping symptoms such as cognitive decline,psychiatric signs and parkinsonian symptoms. When theinitial presentation of DLB is characterized by impairedcognition, it can mimic AD (2,8). Furthermore, neuropsychological evaluation can disclose severe but similar degrees ofimpaired performances concerning attention, frontal lobefunction and motor sequencing in both DLB and AD (9).

Pathological study remains the only way to confirm thediagnosis of DLB, when evidence of Lewy bodies is foundin the cortex, the subcortical regions (nucleus basalis ofMeynert) and the brain stem (substantia nigra and locuscoeruleus) (2, 1012). The presence of Lewy bodies is theessential feature in the pathological diagnosis of DLB; otherfeatures (e.g., plaques and neuronal loss) can also bedescribed in most but not in all cases (3). Consensusguidelines for brain sampling, evaluation of Lewy bodies(distribution and frequency) and diagnostic rating protocolhave been proposed (3).

The precise nosological relationship of DLB and AD isstill a matter of debate. On the one hand, the links with ADare often discussed because of the common clinical characteristics of the dementia and the lesions of Alzheimer typefrequently observed in anatomopathological examination(2); on the other hand, other authors consider the twodiseases independent (13, 14), with DLB corresponding to amore extended form of idiopathic Parkinson's disease (IPD).DLB and IPD could then be the two extremes of acontinuum (15).

With the perspective of neuroprotective treatments, earlydiagnosis is becoming an important factor. The more extensive cholinergic deficit (8,16) observed in DLB comparedwith AD explains the beneficial effect of cholinergic therapy

The aim of this studywas to comparethe regionalcerebralbloodflow measurementsstudied by SPECT in dementia with Lewybodies(OLB)andAlzheimer'sdisease(AD)to determinethecontributionof SPECT to the differentialdiagnosis of these twodiseases. Methods: SPECT analysis with @Tc-hexamethyIpropyleneamineoxime (HMPAO)was performed in 20 patientswith probable DLB, 20 patients with probable AD and 20 patientswith idiopathicParkinson'sdisease (IPD).Tenpairs of regionsofinterest were analyzed. Tracer uptake was expressed as acorticocerebellaractivity ratio. Results: Comparedwith IPD, inthe DLBgroupthere was a global decreaseof HMPAOuptake incortical regions of interest except in the posterior frontal andoccipitalregions;in theADgrouptherewaslimitedlefttemporaland parietal hypopertusion. In the DLB group, frontal HMPAOuptake was significantly lower than in the AD group. Twopredictive scores were established by a factorial discriminantanalysis from six left cortical indices (medial frontal, lateralfrontal, posteriorfrontal, temporopanetal,panetaland panetooccipital) and the Mini-Mental State Examination,which correctlyclassified 53 of 60 patients (88%) (DLB, 18 of 20; AD, 16 of 20;IPD, 19 of 20). ConclusIon: These findings indicate the presenceof diffusecorticalabnormalitiesin DLBand suggestthatSPECT may be useful in discriminating in vivo DLB from AD,revealing mainly frontal hypoperfusionin the former group. Weestimate that SPECT study increasesthe possibilityof separating DLB andAD becauseboth disordersshare differentpatternsofcerebral bloodflow abnormality.Key Words: dementiawithLewybodies;Alzheimer'sdisease;Parkinson'sdisease;SPECT; @Tc-hexamethyIpropyleneamineoximeJ NucI Med 1999;40:956-962

ementia with Lewy bodies (DLB) has been establishedas the most common senile degenerative dementia afterAlzheimer's disease (AD) (13).Clinical diagnostic criteriafor DLB have been defined (3), refining previous criteria(4,5). The central feature required for a diagnosis of DLB isa progressive and fluctuating cognitive decline with recur

ReceivedApr.14,1998;revisionacceptedJan.4, 1999.For correspondence or reprints contact: Luc J.P. Defebvre, MD, PhD,

Servicede Neurologleet PathologieduMouvement,HpltalRogerSalengro,CHRU,59037Ulle,France.

956 THEJOuRNALOFNUCLEARMEDIcm@Vol. 40 No. 6 June 1999

Technetium HMPAO SPECT Study in Dementiawith Lewy Bodies, Alzheimer's Disease andIdiopathic Parkinson's DiseaseLuc J.P. Defebvre, Valerie Leduc, Alan Duhamel, Pascal Lecouffe, Florence Pasquier, Chantale Lamy-Lhullier,Marc Steinling and Alan Destde

Departments ofNeumlogy A, Medical Infor,natics, Nuclear Medicine and Memory Center, Lille, France

by on March 15, 2015. For personal use only. jnm.snmjournals.org Downloaded from

Group Leftmed-fr Rightmed-fr Leftlat-fr Rightlat-fr Leftpost-fr Rightpost-fr

med-fr = medial frontal; lat-fr = lateral frontal; post-fr = posterior frontal; DLB = dementia with Lewy bodies; AD = Alzheimer's disease;IPD= idiopathicParkinson'sdisease.

Statisticaldifference:Themultiplecomparisonof meanswasperformedusingBonferroni'scorrection(a/3witha 0.05).Foreachregionof interest, the groups that are significantly different are specified. DLB/AD indicates a significant difference between the DLB and AD groups.DLB/IPDindicatesa significantdifferencebetweentheDLBandIPDgroups.

However, heterogeneous patterns of ICBF deficits have beenseen with SPECT. This heterogeneity may reflect differentstages of the disease or cognitive subtypes (21). In DLB, atemporoparietal hypoperfusion has recently been shownassociated with an occipital hypoperfusion, which couldexplain the visual hallucinations (22).

The aim of this study was to compare the rCBF measurements studied by SPECT with 99mTc@hexamethyl propyleneamine oxime (HMPAO) in DLB, AD and IPD todetermine the contribution of SPECT to the differentialdiagnosis of DLB and AD.

MATERIALS AND METHODSPatients

SPECT data were collected from 20 patients with probable DLBaccording to the consensus guidelines for the clinical diagnosis ofDLB (3) (mean age 67.6 9.8 y, mean disease duration 6.0 3.4y); 20 patients with probable AD according to the National Instituteof Neurological and Communicative Disorders and Stroke andAlzheimer's Disease and Related Disorders Association (NINCDSADRDA) criteria (6) (mean age 71.4 5.2 y, mean diseaseduration 6.2 2.4 y); and 20 patients with IPD based on the UnitedKingdom Parkinson's Disease Society brain bank criteria (23)(mean age 66.1 6.0 y, mean disease duration 4.8 3.0 y). With aone-way analysis ofvariance (ANOVA), the duration ofthe diseasedid not differ significantly between the three groups (at level 5%).The mean age was significantly higher in the AD group than in theIPD group.

Patients with a history ofpsychiatric disorders or other neurological disorders (including dysthyroidism, vitamin B12 or folatedeficiency, cerebrovascular disease) were excluded. Patients werefree from psychotropic medications (anxiolytic, antidepressant,neuroleptic). CT scans and MR images were negative for stroke.Each patient's cognitive impairment was graded by the MiniMental State Examination (MMSE) (24). In DLB patients, theMMSE score (mean SD) was 16 6; in AD patients, it was 18 7; and in IPD patients, it was 25 3. MMSE scores weresignificantly lower in the DLB and AD groups compared with theIPD group; the difference was not significant between the DLB andAD groups.

SPECT MethodologySPECT was performed with a brain-dedicated fast-rotating

SPECT system, the Tomomatic 564 (Medimatic, Copenhagen,

FIGURE1. Tenpairsofregionsofinterest,rightandleft:medialfrontal(1),lateralfrontal(2),posteriorfrontal(3),temporoinsular(4), temporoparietal (5), parietal (6), parietooccipital (7), occipital(8), lenticularnucleus(9)andthalamus(10).

in DLB (cholinesterase inhibitor), which improves impairedcognitive functions (1 7). An early marker of the diagnosiswould be of the greatest interest, because the reliability ofclinical and pathological criteria of DLB are undergoingevolution without complete specificity (18). Thus, noninvasive imaging with PET or a more widely available techniquesuch as SPECT may be useful methods for the diagnosis ofdementing disorders.

The abnormal pattern of regional cerebral blood flow(rCBF) characteristic of AD is a bilaterally decreasedperfusion to the temporal and parietal regions (19,20).

TABLE 1Mean Values SD of Different Cerebral Blood Flow Indexes Obtained in Three Groups

DLB0.79 0.050.79 0.040.74 0.040.76 0.060.79 0.050.80 0.05AD0.840.070.85 0.070.83 0.070.84 0.070.87 0.070.88 0.06IPD0.860.070.87 0.080.80 0.060.81 0.050.83 0.060.85 0.06Statistical

resultsDLB/AD DLB/IPDDLB/AD DLB/IPDDLB/AD DLB/IPDDLB/AD DLB/IPDDLB/ADDLB/AD

SPECT STUDY IN DEMENTIAWITH LEWY BODY Defebvre et al. 957


Group Lefttemp-ins Righttemp-ins Lefttemp-par Righttemp-par Leftpar Rightpar

Group Leftpar-occ Rightpar-occ Leftccc Rightocc Leftlent Rightlent Leftthai Rightthai


DLB0.81 0.070.85 0.040.81 0.060.84 0.050.76 0.050.76 0.06AD0.880.070.88 0.060.84 0.110.89 0.080.77 0.120.82 0.10IPD0.870.040.90 0.060.90 0.040.92 0.040.84 0.050.85 0.06Statistical

resultsDLB/AD DLB/IPDDLB/IPDDLB/IPD AD/IPDDLB/AD DLB/IPDDLB/IPD AD/IPDDLB/AD DLB/IPD

temp-ins= temporoinsular;temp-par= temporoparietal;par = parietal;DLB= dementiawithLewybodies;AD = Alzheimer'sdisease;IPD= idiopathicParkinson'sdisease.

Statisticaldifference:Themultiplecomparisonof meanswasperformedusingBonferroni'scorrection(a/3witha 0.05).Foreachregionof interest,thegroupsthataresignificantlydifferentarespecified.DLB/ADindicatesa significantdifferencebetweenthe DLBandADgroups.DLB/IPDindicatesasignificantdifferencebetweentheDLBandIPDgroups.AD/IPDindicatesasignificantdifferencebetweentheADandIPDgroups.

Denmark), about 10 mm after intravenous administration of afreshly prepared saline bolus of 550 MBq @[email protected] with their eyes closed in a quiet dark room. The head wascarefully placed along the canthomeatal line using a three-laserlight-positioning system. A first test acquisition (60 s) was performed to control and, if necessary, to adjust the position of thehead of the patient. Two series of five 15-mm-thick slices with acenter-to-center distance of 10mm and parallel to the canthomeatalplane were obtained. Each series required a duration of 10 mm and2 million counts could be obtained in the medium slice. Reconstruction was performed using a filtered backprojection. We used anautogauss filter (DC amplification 80%; power of truncation 3;accepted noise level 3%). A correction for attenuation was performed using a value of0.l6 cm@.

Ten pairs of regions of interest (ROIs) were drawn in conformitywith the Cabanis Atlas (25) on a slice located 50 mm above theorbitomeatal plane. Size of the ROl depends on the region: Rangeswere between 1.9 cm2 for the frontal ROI and 5.5 cm2 for theoccipital ROl. These right and left ROIs were specified in Figure 1.Tracer uptake was expressed as a cotticocerebellar activity ratio, sothat cerebellar hemispheres were used (slice located on theorbitomeatal plane) as the particular reference of the patient, usinga method previously described (26).

Statistical AnalysisAll statistical analyses were performed using SAS software

(SAS Institute, Inc., Cary, NC). Data are expressed as mean andSD. Unless otherwise stated, the statistical tests were made at the0.05 level. For the SPECT data, the assumption of normality was

assessed using the Shapiro-Wilk test (27). The individual predictive value of each parameter related to the 10 pairs of ROIs. TheMMSE score was first examined by a one-way ANOVA andmultiple comparisons of means were performed using the Bonferroni's correction. A factorial discriminant analysis (FDA) wasperformed with the parameters having a significance level less than15%. FDA is a multivariate statistical procedure that uses a set ofexplanatory variables to classify patients into different subgroups(in this study, the AD, IPD and DLB subgroups) by specific rules.FDA allows two new variables, the predictive scores, which arelinear combinations of the explanatory variables. These scoresmaximize the ratio of the variability between the groups to thevariability within the groups, and therefore patients of differentgroups have scores as different as possible. The scores are used todetermine the classification rules. Subsequently, to select the bestsubset of predictor variables, we performed a stepwise discriminantanalysis (at level 15% with stepwise option, which is a forwardselection allowing elimination). The results of the rules wereassessed by the classification table.

RESULTSThe descriptive analysis showed that the distribution of

each CBF index was compatible with the normality assumption. The mean CBF indices SD obtained in the threegroups are reported in Tables 13.Six parameters wereexcluded after the one-way ANOVA: right and left occipital,lenticular nucleus and thalamus (because there was nostatistical difference at the 0.15 level among the three


DLBADIPDStatistical results

0.780.060.79 0.110.82 0.04

DLB/IPD

0.77 0.060.820.100.83 0.04

DLB/IPD

par-occ= panetooccipital;occ= occipital;lent= lenticuiarnucleus;thai = thalamus;DLB= dementiawithLewybodies;AD = Alzheimer'sdisease;IPD= idiopathicParkinson'sdisease;NS= notsignificant.

Statisticaldifference:Themultiplecomparisonof meanswasperformedusingBonferroni'scorrection(a/3witha 0.05).Foreachregionof interest, the groupsthat are significantly differentare specified. DLB/IPD indicates a significantdifference between the DLB and IPD groups.

0.94 0.07 0.95 0.07 0.91 0.06 0.91 0.05 0.86 0.06 0.85 0.070.97 0.07 0.99 0.07 0.95 0.07 0.96 0.06 0.88 0.06 0.88 0.070.980.04 0.990.04 0.920.06 0.930.06 0.880.05 0.890.05

NS NS NS NS NS NS

958 Tiii@JouRN@iOFNUCLEARMEDICINEVol. 40 No. 6 June 1999


DiagnosisPrediction DLB AD IPD

DLB 18

DLB = dementiawith Lewybodies;AD = Alzheimer'sdisease;IPD = idiopathic Parkinson's disease.

Alltogether, 88.3% were correctlyclassified.

be relevant for predicting diagnosis. Then, from Figure 2,three decision rules were derived: (a) if SI < 0, thendiagnosis = IPD; (b) if @l> 0 and @2@ 0, then diagnosisAD; (c) if S@> 0 and if 52 < 0@then diagnosis = DLB.Consider, for example, a patient with the following SPECTcharacteristics: left medial-frontal = 0.88, left lateralfrontal = 0.75, left posterior-frontal 0.79, left temporoparietal = 0.73, left parietal = 0.7, left parietal-occipital = 0.7,MMSE = 11. By computing the two score values and byusing the two rules, this patient would be classified in theDLBgroup(S1 +1,S2 1.35).

Table 5 gives the result of the classification obtained usingthe two decision rules. Of the 60 patients, 88.3% werecorrectly classified: 90% for the DLB group, 80% for the ADgroup and 95% for the IPD group.

DISCUSSIONIn this study, we found a diffuse cortex reduction of

HMPAO uptake in DLB patients compared with parkinsonian and AD patients for the same grade of cognitiveimpairment in the DLB and AD groups, whereas there wasno significant difference for duration of the disease. Moreover, a clear distinction among the three groups wasestablished with the FDA. The diffuse decrease of HMPAOcerebral uptake observed in the DLB group compared withthe IPD group suggests widespread lesions in the cortex, ashas been demonstrated in pathological studies (corticalLewy bodies are observed in many cortical structuresespecially in the frontal, anterior cingulate, insular andtemporal cortex) (8,11,29).

In IPD, cortical metabolism abnormality depends ondisease duration and presence or absence of dementia: In thefirst stage of the disease, there is no change of glucosemetabolism (30), whereas a significant global hypometabolism may appear a few years later (31). Piert et al. (32)showed a global reduction of glucose utilization in Parkinson's disease (mean reduction 22%) with a greater deficitacross the brain with progression of the disease and development of dementia, especially in the parietal and occipitalcortex. A marked parietal glucose hypometabolism, as

ParametersScore 1Score2Mini-MentalStateExamination0.10Left

medial-frontalindex12.51.2Leftlateral-frontalindex0+12.2Leftposterior-frontalindex+18.7+7.3Lefttemporoparietalindex4.2+5.3Leftpanetalindex8.46.9Leftpanetooccipitalindex+103.5Constancy+0.311.8

groups). In the IPD group, the ROl indices were normal: allthe mean values of each ROl were superior to 0.80 (26,28).Compared with IPD, in the AD group there was a limited lefttemporoparietal and parietal reduction of the HMPAOuptake, and in the DLB group there was a global andbilateral decrease of uptake in all cortical ROIs except in theposterior frontal and occipital regions. Comparison of thetwo demented groups revealed in the DLB group a hypoperfusion in all frontal regions and in the left temporoinsular,right temporoparietal and parietal regions.

The FDA performed on 15 parameters (MMSE score and14 indices: all except for the right and left occipital lobes,lenticular nucleus and thalamus) showed a good separationamong the three groups by means of the two discriminantscores (S@and 52). The squared correlation ratio (CR) ofthese scores (ratio of the interclassclassvariance to the totalvariance) was CR1 0.70 (SI) and CR2 0.5 (52),respectively. Using Mahalanobis distance, we correctlyclassified 93% (56/60) of the patients. The IPD group wasdistinguished from the DLB and AD groups by means of @.The most discriminant parameters associated with SI wereMMSE (r 0.68), left temporoparietal (r = 0.46) and leftparietal (r = 0.46) (r denotes the Pearson correlationcoefficient between the variable and the score 51). 52distinguished DLB and AD groups and the correspondingdiscriminant variables were left lateral-frontal (r = 0.76),right lateral-frontal (r = 0.73), left posterior-frontal (r0.67) and right posterior (r = 0.68).

The following predictor variables were definitively retamed after the stepwise discriminant analysis: MMSE, leftmedial-frontal, left lateral-frontal, left posterior-frontal, lefttemporoparietal, left parietal and left parietooccipital. Asecond FDA was performed on these seven remainingparameters. The results are presented in Tables 4 and 5 andFigure 2. The correlation ratios of the two scores derivedfrom the FDA were CR1 = 0.6 and CR2 = 0.4. Table 4shows how these two scores can be computed from the sevenpredictors. Figure 2 represents the projection of the 60patients in the 52 X SI plane. On this plot, the first coordinateis defined by the@ value and the second coordinate isdefined by the 52 value, both computed from Table 4. Figure2 demonstrates that the subset of seven variables seemedto

TABLE 4CoefficientValuesObtainedbyFactorialDiscriminant

Analysis Computing Scores 1 and 2 Accordingto the Seven Predicting Factors

TABLE 5PredictiveResultsforWholePopulationAccording

toPredictingFactorsObtainedwithFactorialDiscriminant Analysis

22 160 2

90% 80%

01995%

ADlPD

Correctlyclassified

SPECT STUDYIN DEMENTIAWITHLEw@BODYDefebvre et al. 959


?CORE2U U

0. U U

20 U

U U U00 00@ C

U U U0

U U +U

UU00

@.__________l___Q_,@ 0 p-3 -2 .1 0

0

.10 0

-2

.3U

@ P , t-@@ 1@@ 2 3 4 5

+ ++ + + +

+

++

+ U

+

U++

FIGURE2. Separationof threegroupsofpatients by discnminant scores S@and 52.+ = DLB; = AD; 0 = IPD.

observed in AD, has been noted as the primary differencebetween Parkinson's disease and Parkinson's disease withdementia (31). However, in such studies, recent clinicalcriteria were not used to distinguish DLB. In our IPD group,no patients were demented, which is consistent with thenormality of the ROI indices (superior to 0.8). In AD, themost marked rCBF defect using the radionucleotide HMPAOis localized in posterior cortical regions bilaterally (19);however, left hemispheric uptake decrease has been found tobe more severe (33), which explains why there was only aleft temporal and parietal uptake reduction in the AD groupcompared with the IPD group in our study.

A recent SPECT study compared the brain perfusion(visual analysis) of six patients with probable AD to that ofseven patients with DLB (22). Compared with three controls, there was a bilateral temporal or parietal hypoperfusion for all AD patients with additional frontal hypoperfusion in two patients and an occipital defect in one patient. Inthe DLB group, there was, in addition to a temporoparietalHMPAO uptake decrease, an occipital hypoperfusion in sixpatients. Varma et al. (34) observed that the mean corticalCBF indices were lower in DLB (20 patients;meandurationof disease 4.3 y) and AD (57 patients; mean duration ofdisease 3.6 y) than in controls. The decrease was similar andsymmetrical in the two groups and was most pronounced inposterior cortical ROl. However, there was a significantlysmaller degree of variance in the left-right CBF asymmetryparameter for the DLB group. Taking into account theclinical onset (parkinsonian or cognitive-psychiatric symptoms), no difference for the CBF results was observed withinthe DLB group. The more diffuse decrease of regionalcerebral perfusion observed in our DLB group in comparison with the study of Donnemiller et al. (22) can beexplained by the fact that the dementia was more severe in

our patients (mean MMSE score 16 6) than in theirpatients (mean MMSE score 20 5).

In the same way, a fluorodeoxyglucose (FDG) PET studyprovided evidence ofdiffuse cerebral hypometabolism (frontal, temporal and parietal) in both pure DLB (3 patients) andcombined DLB-AD (3 patients) patients with a markeddecline in parietal cortex, sparing the primary somatomotorcortex and subcortical strutures (35). An occipital hypometabolism was also found in both occipital association andprimary visual cortex and might be related to the visualhallucinations classically observed in DLB, although theoccipital lobe may be relatively spared in pathologicalanalysis (3537). Minoshima et al. (38) also observedsimilar metabolic reduction of FDG in AD patients with(n = 10) and without (n = 7) cortical Lewy bodies inmultiple brain regionsincluding parietal, temporal andfrontal cortexwith a significant severe reduction in theoccipital cortex for the first group (25% of controls) ascompared with the other group (10%), suggesting that thisoccipital defect may serve as an in vivo indicator for thepresence of cortical Lewy bodies. However, this occipitalhypometabolism is observed not only in DLB but in ADpatients (39,40) and consequently does not seem to be usefulfor distinguishing the metabolic pattern of these two diseases.

The comparison between our two demented groups essentially revealed a hypoperfusion in all frontal regions in theDLB group and seems to confirm, on the contrary, thatfrontal CBF uptake is more sensitive for distinguishing invivo DLB and AD. This HMPAO uptake decrease could beexplained by the high density of cortical Lewy bodiessometimes observed in the frontal cortex (8). Finally, it canbe assumed that CBF abnormalities in DLB patients reflect acombination of direct cortical pathology and cortical deaffer

960 THE JOURNALOF NUCLEARMEDICINE Vol. 40 No. 6 June 1999


under the auspices of Department of Health and Human Services Task Force onAlzheimer's disease. Neurology. 1984:34:939944.

7. SoininenH,LaulumaaV.HelkalaEL,HartikainenP.RieckkinenPJ.Extrapyramidal signs in Alzheimer's disease: a 3-year follow-up study. I Neural Transm.1992:4:107119.

8. Frstl H, Burns A, Luther P. Cairns N, Levy R. The Lewy body variant ofAlzheimer's disease: clinical and pathological findings. Br) Psychiatry. 1993:162:385392.

9. Gnanalingham KK, Byrne El, Thornton A, Sambrook MA, Bannister P. Motor andcognitive function in Lewy body dementia: comparison with Alzheimer's andParkinson's diseases. I Neumi Neumsurg Psychiatry. 1997:62:243-252.

10. Kosaka K, Yoshimura M, Ikeda K, Budka H. Diffuse type of Lewy body disease:progressive dementia with abundant cortical Lewy bodies and senile changes ofvarying degree a new disease? Clin Neumpharmacol. 1984:3:185192.

11. Gibb WR, Luthert PJ, Janota I, Lantos PL. Cortical Lewy body dementia: clinicalfeatures and classification. J NeumlNeumsurg Psychiatry. 1989:52:185192.

12. Kosaka K. Diffuse Lewy body disease in Japan. JNeumL 1990:237:197204.13. Dickson DW, Crystal HA, Davies P. Hardy H. Cytoskeletal and Alzheimer-type

pathology in Lewy body disease. In: Perry R, McKeith I, Perry E, eds. Dementiawith Lewy Bodies. Clinical, Pathological and Treatment Issues. Cambridge, UK:Cambridge University Press; 1996:224237.

14. Kosaka K, Iseki E. Diffuse Lewy body disease within the spectrum of Lewy bodydisease. In: Perry R, McKeith I, Perry E, eds. Dementia with Lewy Bodies.Clinical, Pathologicaland TreatmentIssues. Cambridge, UK: Cambridge University Press; 1996:238247.

15. Filley CM. Neuropsychiatric features of Lewy body disease. Brain Cogn.1995:28:229239.

16. Perry EK, Irving D, Kerwin JM. Cholinergic transmitter and neurotrophicactivities in Lewy body dementia: similarity to Parkinson's and distinction toAlzheimer's disease.AlzheimerDisAssoc Disord. 1993:7:6979.

17. Lebert F, Souliez L, Pasquier F. Tacrine and symptomatic treatment in Lewy bodydementia. In: Perry R, McKeith I, Perry E, eds. Dementia with Lewy Bodies.Clinical, Pathologicaland Treatment Issues. Cambridge, UK: Cambridge University Press; 1996:439448.

18. Mega MS. Masterman DL, Benso DF, et al. Dementia with Lewy bodies:reliability and validity ofcinical and pathologiccritaria. Neurology. 1996:47:14031409.

19. Neary D, Snowden iS, Shields RA, Burjan AWl, Northen B. Single photonemission tomography using @Fc-HMPAOin the investigation of dementia. INeural Neumsurg Psychiatry. 1987:50:11011109.

20. Jagust Wi, Johnson KA, Holman BL. SPECT perfusion imaging in the diagnosisofdementia. INeuroimag. l995;5(suppl):545S52.

21. Waldemar G, Bruhn P, Kristensen M, Johnscn A, Paulson OB, Lassen NA.Heterogeneity of neocortical cerebral blood flow deficits in dementia of theAlzheimer type: a w@@Tc@d,l@HMPAOSPECT study. I Neuml Neumsurg Psychiatry. 1994:57:285295.

22. Donnemiller E, Heilmann J, Wenning GK, Ct al. Brain perfusion scintigraphy withw@@Tc@HMPAOor @Tc-ECDand @I-@-CITsingle-photon emission tomographyin dementia of the Alzheimer-type and diffuse Lewy body disease. Eur I NuclMed. l997;24:320325.

23. Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical diagnosis ofidiopathic Parkinson's disease: a clinico-pathological study of 100 cases. JNeurolNeumsurg Psychiatty. 1992:55:181184.

24. FolsteinM, FolsteinS, McHughPR. Mini-MentalState:practicalmethodforgrading the cognitive state of patients for the clinician. I Psychiatr Res.1975:12:911922.

25. Cabanis EA, Tamraz J, Iba-Zizen MT. Imagerie par resonance magntique(IRM)de la tte a 0,5 Tesla. Atlas de correlations anatomiques normales dans 3dimensions, scIon l'orientation du plan neuro-oculaire (PNO). Feuillets deRadiologie. 1986:26:309416.

26. Steinling M, Mazingue A, Kassiostis P. et al. Le HmPaO-Tc comme indicateur dudebit sanguin crbrallocal: tudequantifiecomparea Ia mthodeparinhalation du Xenon 133. Ann Radio! 1988:4:229237.

27. Shapiro 55, Wilk MB. An analysis of variance test for normality (completesamples). Biometrika. 1965:52:591611.

28. Steinling M, Leys D, Amegassi F, et al. Assessment ofdementia with HMPAO-Tc:which is the best reference for the quantification? In: Schmidt HAE, Chambron J,ala. NuclearMedicine. QuantitativeAnalysis in Imaging and Function. Strasbourg,France: Schattaueur; 1989:349351.

29. Kosaka K. Lewy bodies in cerebral cortex. Report of three cases. AdaNeumpathol. 1978:42:127134.

30. Rougemont D, Baron JC, Collard P. Bustany P. Comar D, Agid Y. Local cerebralglucose utilization in treated and untreated patients with Parkinson's disease. INeumlNeumsurg Psychiatry. 1984:47:824830.

SPECT STUDYIN DEMENTIAWITHLEw@BODYDefebvre et al. 961

entation phenomenon secondary to subcortical lesions. Thesubcortical dysfunction can also be estimated by using 12312@3carboxymethoxy-3@3-[4-iodophenyl]tropane (23I-@3-CIT):A lower striatal-to-cerebellar ratio of 23I-@3-CITbinding hasbeen demonstrated in DLB patients compared with ADpatients and controls (22), in accordance with the knownsevere nigrostriatal degeneration in DLB. Some of the limitsof our methodology must be specified. Use of HMPAO givesan instantaneous assessment of local cerebral perfusionlimited to a qualitative or semiquantitative approach. Indexvalues also depend on the choice of the reference; thefollowing references were used the most: cerebellar hemispheres, global hemisphere activity and occipital region.Criticisms can be made for all these references; however, wehave already established that the cerebellar reference wasthe best one (28). ROIs were determined on a single sliceonly (50 mm above the orbitomeatal plane) and they did notallow extensive estimation of the CBF activity of thedifferent cortical regions; however, a predictive score couldbe obtained rapidly and simply with this method. Our systemhad poor resolution and, at times, we encountered somedifficulties when measuring HMPAO uptake in subcorticalregions of small volume (thalamus and lenticular nucleus).This probably explains in part why no difference wasobserved among the three groups for these regions in ourstudy. It would be interesting to use other markers such as123I-@3-CIT(28) or 23I-iodobenzamide(41) to analyze thesubcortical dysfunction for distinguishing DLB and AD inpatients.

CONCLUSIONWe consider that SPECT study increases the possibility of

distinguishing between DLB and AD, because both disorders share different patterns of CBF abnormality. Our datasuggest that the rCBF cortical changes observed in patientswith parkinsonian syndrome and dementia could constitute apossible means of investigation in vivo for establishing adifferential diagnosis of DLB and AD. A prospective longitudinal study will be required to define the role of such SPECTstudies in obtaining a more precise diagnosis at the first stageof the evolution. The accurate clinical differentiation of DLBand AD is important to patient management, long-termprognosis and critical evaluation of future treatments.

REFERENCES1. Byrne El, Lennox 0, Lowe J, Godwin-Austen R. Diffuse Lewy body disease:

clinical features in 15 cases. JNeumlNeumsurg Psychiatry. 1989:52:709717.2. Hansen L, Salmon D, Galasko D, et al. The Lewy body variant of Alzheimer's

disease: a pathological and clinical entity. Neurology. 1990:40:1-8.3. McKeith 1G. Galasko D, Kosaka K, et al. Consensus guidelines for the clinical and

pathologic diagnosis of dementia with Lewy bodies (DLB): report of theconsortium ofDLB intemational workshop. Neumlogy. 1996:47:11131124.

4. Byrne El, Lennox GO, Godwin-Austen RB, et al. Dementia associated withcortical Lewy bodies: proposed clinical diagnostic criteria. Dementia. 1991:2:283284.

5. McKeith 1G. Fairbain A, Perry RH. Clinical diagnostic criteria for Lewy bodydementia. Dementia. 1992:3:251252.

6. McKhann G, Drachman D, Folstein M, Katzman R, Puce D, Stadlan EM. Clinicaldiagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group


31. KuhI DE, Metter El, Riege WH. Patterns of local cerebral glucose utilizationdetermination in Parkinson's disease by the [F]fiuorodeoxyglucosemethod. AnnNeurol. 1984:15:419-424.

32. Piert M, Koeppe RA, Giordani B, Minoshima 5, Kuhl DE. Determination ofregional rate constants from dynamic FDG PET studies in Parkinson's disease. INuclMed. 1996:37:11151122.

33. Pearlson GD, Harris GJ, Powers RE, et al. Quantitative changes in mesialtemporal volume, regional cerebral blood flow, and cognition in Alzheimer'sdisease. Arch Gen Psychiatry. 1992:49:402408.

34. Varma AR, Talbot PR, Snowden iS, Lloyd ii, Testa Hi, Neary D. A @mTc@HMPAOsingle-photonemissioncomputedtomographystudy of Lewybodydisease. J Neural. 1997:244:349359.

35. Albin RL, Minoshima S, D'Amato Ci, Frey KA, KuhI DA, Sima AAF.Fluoro-deoxyglucose positron emission tomography in diffuse Lewy bodydisease. Neurology. 1996:47:462466.

36. Gibb WR, Eseiri MM, Lee AJ. Clinical and pathological features of diffusecortical Lewy body disease (Lewy body dementia). Brain. 1985;lI0:ll3l1153.

37. McKeith 10, Fairbairn AF, Perry RH, Thompson P. The clinical diagnosis andmisdiagnosis of senile dementia of Lewy body type (SDLT). Br I Psychiatry.1994; 165:324332.

38. Minoshima 5, Foster NL, Frey KA, Ct al. Metabolic differences in Alzheimer'sdisease with and without cortical Lewy bodies as revealed by PET [abstract]. ICerebBloodFlowMetab.1997;17:5437.

39. Hof PR, Bouras C, Constantinidis J, Morrison JH. Selective disconnection ofspecific visual association pathways in cases of Alzheimer's disease presentingwith Balint's syndrome. I Neumpathol Exp Neuml. 1990:49:168184.

40. Berthier ML, Leiguarda R, Starkstein SE, Sevlever G, Taratuto AL. Alzheimer'sdisease in a patient with posterior cortical atrophy. I Neumi Neurosurg Psychiatry.1991;54:lllOllll.

41. Walker Z, Costa DC, Janssen AG, Walker RW, Livingstone G, Katona CL.Dementia with Lewy bodies: a study of post-synaptic dopaminergic receptors withiodine-123 iodobenzamide single-photon emission tomography. Eur I Nucl Med.1997:24:609614.

962 THEJoui@i. OFNUCLEARMEDICINEVol. 40 No. 6 June 1999


1999;40:956-962.J Nucl Med.

Steinling and Alain DesteLuc J.P. Defebvre, Valrie Leduc, Alain Duhamel, Pascal Lecouffe, Florence Pasquier, Chantale Lamy-Lhullier, Marc

and Idiopathic Parkinson's DiseaseTechnetium HMPAO SPECT Study in Dementia with Lewy Bodies, Alzheimer's Disease

http://jnm.snmjournals.org/content/40/6/956This article and updated information are available at:

http://jnm.snmjournals.org/site/subscriptions/online.xhtmlInformation 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 1999 SNMMI; all rights reserved.


Documents

Technetium HMPAO SPECT Study in Dementia with Lewy Bodies, Alzheimer's Disease and Idiopathic Parkinson's Disease