6 Month Cognitive Intervention Amnestic Mci and Mild Ad

Journal of Alzheimer’s Disease 26 (2011) 337–348DOI 10.3233/JAD-2011-0025IOS Press

337

Effects of a 6-Month Cognitive Interventionon Brain Metabolism in Patients withAmnestic MCI and Mild Alzheimer’s Disease

Stefan Forstera,i,∗, Verena C. Buschertb, Stefan J. Teipelc,d, Uwe Friesec,g, Hans-Georg Buchholzf ,Alexander Drzezgai, Harald Hampele, Peter Bartensteina and Katharina Buergerb,h

aDepartment of Nuclear Medicine, Ludwig-Maximilian University, Munich, GermanybDementia Research Section and Memory Clinic, Alzheimer Memorial Center and Geriatric PsychiatryBranch, Department of Psychiatry Ludwig-Maximilian University, Munich, GermanycDepartment of Psychiatry, University Rostock, Rostock, GermanydDZNE, German Center for Neurodegenerative Disorders, Rostock, GermanyeDiscipline of Psychiatry, School of Medicine & Trinity College Institute of Neuroscience (TCIN),Laboratory of Neuroimaging and Biomarker Research, Trinity College, University of Dublin, Dublin;The Adelaide and Meath Hospital Incorporating the National Children’s Hospital (AMiNCH), Irelandf Department of Nuclear Medicine, University Mainz, Mainz, GermanygInstitute of Psychology, University of Osnabrueck, Osnabrueck, GermanyhInstitute for Stroke and Dementia Research, Klinikum Großhadern, Ludwig-Maximilian University,Munich, GermanyiDepartment of Nuclear Medicine, Technische Universitat Munchen, Munich, Germany

Abstract. The effect of cognitive intervention on brain metabolism in AD is largely unexplored. Therefore, we aimed toinvestigate cognitive parameters and 18FDG PET to test for effects of a cognitive intervention in patients with aMCI or mild AD.

Patients with aMCI (N = 24) or mild AD (N = 15) were randomly assigned either to cognitive intervention groups (IGs),receiving weekly sessions of group-based multicomponent cognitive intervention, or active control groups (CGs), receivingpencil-paper exercises for self-study. We obtained resting-state FDG-PET scans and neuropsychological testing at baseline andafter six-months. Normalized FDG-PET images were analyzed using voxel-based SPM5 approaches to determine longitudinalchanges, group-by-time interactions and correlations with neuropsychological outcome parameters. Primary global cognitiveoutcome was determined by analyses of covariance with MMSE and ADAS-cog scores as dependent measures.

Both, aMCI and AD subgroups of CGs showed widespread bilateral cortical declines in FDG uptake, while the AD subgroupof IGs showed discrete decline or rather no decline in case of the aMCI subgroup. Group by time analyses revealed strongestattenuation of metabolic decline in the aMCI subgroup of the IGs, involving left anterior temporal pole and anterior cingulategyrus. However, correlation analyses revealed only weak non-significant associations between increased FDG uptake andimprovement in primary or secondary outcome parameters. Concurrently, there was significant improvement in global cognitivestatus in the aMCI subgroup of the IGs.

∗Correspondence to: Stefan Forster, MD; Klinik und Polik-linik fur Nuklearmedizin, TU Munchen; Ismaninger Str. 22,81675 Munchen, Germany. Tel.: +49 89 4140 2965; Fax: +49 894140 4950; E-mail: [email protected].

ISSN 1387-2877/11/$27.50 © 2011 – IOS Press and the authors. All rights reserved

mailto:[email protected]

338 S. Forster et al. / Effects of a 6-Month Cognitive Intervention on Brain Metabolism

A six-month cognitive intervention imparted cognitive benefits in patients with aMCI, which were concurrent with an attenuateddecline of glucose metabolism in cortical regions affected by neurodegenerative AD.

Keywords: FDG PET, cognitive intervention, cognitive training, cognitive stimulation, Alzheimer’s disease, mild cognitiveimpairment

INTRODUCTION

Cognitive intervention for patients with Alzheimer’sDisease (AD) is considered an important contributionto the treatment of AD [1]. Specifically designed cog-nitive interventions can improve memory performanceand can even attenuate the risk of future cognitivedecline in non-demented elderly subjects [2, 3]. Ben-eficial effects of interventions on cognitive decline arereported in subjects with preclinical cognitive impair-ment (see review [4]) or mild-to-moderate stages ofdementia (see meta-analyses [5, 6]).

Effective cognitive interventions might conceiv-ably impart their effects through altered cerebralmetabolism. The rate of cerebral glucose consump-tion can be assessed in positron emission tomography(PET) studies recording the uptake of the stable glu-cose analogue [18F]fluorodeoxyglucose (FDG); thePET-FDG technique has been extensively used forstudying the pathophysiology of neuropsychiatric andneurodegenerative disorders, including AD (see review[7]). FDG-PET recordings obtained at rest are a sensi-tive indicator of perturbed brain metabolism, and mayreveal very early cerebrometabolic changes precedingthe onset of clinical AD symptoms [8]. Voxel-wisemapping of FDG uptake is amenable for the lon-gitudinal assessment of brain functional changes indementia, and may more sensitively detect treatment-response than do commonly-used instruments forcognitive testing, such as the Mini Mental StateExamination (MMSE) and the Alzheimer’s DiseaseAssessment Scale - cognitive subscale, total score(ADAS-cog) [9]. However, the effect of cognitiveintervention on cerebral metabolism in AD is largelyunexplored. There are only two published studiesinvestigating the effects of cognitive intervention onbrain metabolism. In an early study, FDG-PET wasobtained in a visual stimulus activation paradigm withAD patients, and the effects of an unspecified cognitivetraining of six months duration and a pharmacologicalintervention were tested [10]; the combined treatmentproved to be superior to cognitive training alone. Morerecently, participation in 14-day long healthy lifestyleprogram consisting of a combination of mental and

physical exercise, stress reduction, and healthy dietwas associated with significant short-term benefits incognitive function and cerebral FDG uptake in elderlynon-demented subjects [11].

In the present study, we tested the hypothesis thata newly-developed multicomponent cognitive inter-vention [12] applied in a randomized controlled trialwould lead to cognitive and non-cognitive bene-fits and mitigate against declining brain metabolismin patients with amnestic MCI or mild AD. Totest this hypothesis, we examined global cognitive(MMSE- and ADAS-cog. scores) and additionallynon-cognitive parameters, such as mood (MontgomeryAsberg Depression Rating Scale, MADRS), which aredeemed to be important for cognitive performance[13]. We used FDG-PET to map the pattern of declin-ing cerebral glucose metabolism during six months(between baseline and follow-up scans) in a well-characterized sample of aMCI and mild AD patients,participating in a parallel group randomized controlledtrial with two treatment arms, (i) specific cognitiveintervention and (ii) an active control condition. Priorto the start of the study, a multicomponent cogni-tive intervention for patients with aMCI and patientswith mild AD was conceptualized aiming at capa-bilities and needs of patients in different stages ofAD [12]. In order to gain more direct insights intothe functional-anatomical substrate of neuronal func-tion along with cognitive improvement, voxel-basedbidirectional SPM patient-group-by-time interactionanalyses (later referred to as “difference of differ-ences” analyses), measuring baseline to follow-up PETmetabolic differences between the patient groups andtreatment arms, were applied.

MATERIALS AND METHODS

Subjects

We screened 43 patients at the Dementia ResearchSection and Memory Clinic of the AlzheimerMemorial Center and Geriatric Psychiatry Branch,Department of Psychiatry, Ludwig-Maximilian Uni-versity, Munich, from March to August 2007. In the

S. Forster et al. / Effects of a 6-Month Cognitive Intervention on Brain Metabolism 339

screening, we conducted a comprehensive clinical andneuropsychological assessment, in order to supportthe research diagnosis of either aMCI or mild AD.Participants with aMCI had a memory complaint andperformed at least 1.5 SD below the average level ofpersons of a similar age and education on at least oneof three memory tests of the Consortium to Estab-lish a Registry for Alzheimer’s Disease (CERAD)neuropsychological test-battery (Morris, et al., 1989)(immediate and/or delayed recall and/or recognition).Their cognitive difficulties had no significant reper-cussions on their functional independence, as assessedthrough clinical interviews with the patients and care-givers. Hence, they did not fulfil criteria for mild stagesof clinically probable AD [14]. Neither of the patient(groups) had major physical illness, other mental dis-order (i.e. major depression) or disability which couldhave affected participation. Thirty-nine patients, meet-ing stringent inclusion criteria, were included. Fordetails please see [12].

Routine laboratory testing of all included patientsconsisted of full blood cell count, blood glucose, thy-roid function tests, serum Vitamin B12 and folic acidlevels as well as cerebrospinal fluid protein and Apo-Egenotype testing; there were no major abnormalities

at baseline. Each patient underwent structural mag-netic resonance imaging (MRI) for detection of brainanatomical abnormalities; according to the Schelten´sscale [15] patients with structural lesions visible inthe T2-weighted MR scans exceeding 10 mm in diam-eter, including white matter hyperintensities, wereexcluded from the study. According to the declarationof Helsinki approval of the local ethics commission(of the Ludwig-Maximilian University, Munich), andthe German Radiation Safety Committee (BfS) wasobtained prior to starting the study.

Design

Effects of a six-month cognitive intervention pro-gram on brain glucose metabolism were assessed in 36patients with either aMCI (N = 21) or mild AD (N = 15)syndrome. The trial design is shown in Fig. 1.

Briefly, patients were randomly assigned to eithera cognitive intervention treatment arm (IG) or activecontrol condition treatment arm (CG), resulting in 9aMCI patients and 8 AD patients for the mixed inter-vention groups (IGs), as well as 12 aMCI patients and 7AD patients for the mixed control groups (CGs). Dur-ing a six month period members of the IGs received

Fig. 1. Trial design. aMCI = amnestic mild cognitive impairment; AD = Alzheimer’s Disease; IG = intervention treatment arm; CG = controlcondition treatment arm.


weekly sessions of 120-minutes group-based cogni-tive intervention, whereas the CGs participants metmonthly and received pencil and paper exercises forself-study. The cognitive intervention was based onthe theory of cognitive reserve [16] and was tailoredto the cognitive and functional requirements of aMCIand mild AD patients according to the theory of Ret-rogenesis, as a theoretical basis for the selection ofspecific training tasks [17]. It focussed particularly onglobal cognitive functioning but addressed also non-cognitive domains, e.g. mood and quality of life, thatare thought to be impaired in aMCI and frank AD[13]. On the supposition that MCI resides within thecognitive continuum from normal aging to AD [18],differences between the intervention for MCI and formild AD corresponded rather to a quantitative than aqualitative distinction [12]. In contrast to the tailoredcognitive intervention, the CGs sessions focussed onexercises of isolated, sustained attention, which is sup-posed to be largely unimpaired, at least in mild AD[19].

The colleagues who conducted neuropsychologicaltesting sessions and PET-scanning were blinded tosubject classification and treatment plan; an instruc-tor uninvolved in these steps administered the IGs andCGs sessions.

PET imaging

Baseline and six-months follow-up FDG-PET scansof 36 patients were performed successfully. Allpatients had fasted for at least six hours prior to FDG-PET. Recordings were made using a Philips AllegroPET scanner with a 128 × 128 pixel matrix (pixelsize 2 mm), an axial field of view of 180 mm and afull width at half maximum (FWHM) resolution of5.5 mm. Patients were asked to recline on the scan-ner bed, resting quietly with their eyes covered andtheir ears occluded. Each patient’s head was positionedwithin the aperture of the tomograph, and comfortablyimmobilized using a foam cushion. At 20 minutes afterinjection of FDG (mean 200 MBq, i.v.), an attenua-tion scan was obtained with a rotating [137Cs] pointsource. A dynamic emission recording consisting offive six-minute frames was obtained in the interval30 to 60 minutes following the FDG injection. Aftervisual inspection to exclude frames with unaccept-able head motion, attenuation-corrected frames weresummed into a single frame, and final images werereconstructed iteratively using a three-dimensional rowaction maximum likelihood algorithm (3D-RAMLA).

Cognitive data analysis

Prior to the screening-phase, the study was regis-tered in a public trials registry (www.clinicaltrials.gov;ID: NCT00544856). Primary outcome was evaluatedseparately for aMCI and AD patients by comparingchange in the global cognitive parameters (MMSEscore [20] and Alzheimer’s Disease Assessment Scale– total score of the cognitive subscale (ADAS-cog)[21]) in the cognitive intervention treatment arm ver-sus change in the control condition treatment armtested by the time (progression) by arm interaction termin a two-factor ANOVA. Further information regard-ing secondary outcome measures, not relevant for thepresent study, and the neuropsychological data analy-ses is described elsewhere [12].

PET data analysis

SPM5 routines (Wellcome Department of CognitiveNeurology, London, UK) implemented in MATLAB(version 7.1) were used to perform basic image pro-cessing and voxel-based statistical analysis. All PETscans were spatially normalized using default trans-formation parameters to the SPM5 standard PETbrain template in the Montreal Neurological Institute(MNI) space. Normalized images were composed ofa 79 × 95 × 69 matrix, with 2 × 2 ×2 mm voxel size.These images were then smoothed using an isotropicGaussian kernel (12 mm FWHM), which accommo-dates inter-individual anatomical variability, and thusimproves the sensitivity of the statistical analysis [22].FDG uptake scaling was performed using the referencecluster approach as described in detail by Yakushevet al. [23]. Briefly, with this approach regional FDGuptake is normalized to that measured in brain regionsdefined a posteriori to be unaffected, relative to theresults in a non-demented control group. This a pos-teriori method was shown to be more sensitive thanthe global mean approach in detecting disease-relatedmetabolic disturbances in mild-to-moderate stages ofneurodegenerative disorders [24, 25], especially instudies on dementia [23, 26, 27], [28]. For thesecomparisons we made use of previously obtainedFDG-PET data from 11 healthy elderly subjects (5female; mean age 59.0 ± 10.9 years), who were non-complainers, had an MMSE score above 28, hadno signs of microangiopathy on MRI, no history ofneurological- or major psychiatric diseases and did notreceive any psychoactive medication [27]. The latterelderly control subjects had been examined in a resting-


state on the same scanner and with image processingaccording to the identical protocol as for the patients.

First of all, in order to characterize our patients neu-robiological status at baseline, we compared baselineregional FDG uptake between all patients and non-demented controls, using an unpaired t-test.

Then for evaluation of longitudinal changes in FDGuptake we performed within-group comparisons (inboth the AD and aMCI subgroups of the IGs and theCGs) between baseline and end-of-study PET scansusing paired t-tests. For a more specific evaluationof neurobiological intervention effects we elected touse a voxel-based bidirectional patient-group-by-timeinteractions design, referred to as “difference of dif-ferences” analysis, which assessed the differences inextent of metabolic changes between both the ADand aMCI subgroups of the IGs and the CGs betweenthe initial scan and the end-of-study scan. We con-sidered a statistical threshold of p < 0.001 uncorrectedand a threshold for minimum spatial extent of 30 con-tiguous voxels for anatomical reporting of significantchanges in FDG uptake. Foci of significant changeswere automatically assigned by the SPM software tothe coordinate system of the MNI space, and then con-verted to Talairach and Tournoux coordinates [29] foridentification of the Brodmann areas and anatomic des-ignations.

In order to link the clinical neuropsychologicaloutcome with the neurobiological PET findings weperformed additional correlation analyses betweenchanges in FDG uptake in brain regions, which hadshowed the highest metabolic effects, and changes inprimary and secondary outcome parameters in aMCIpatients. Using a ROI-based approach (MARSBARtoolbox implemented in SPM5), we extracted normal-ized baseline and follow-up FDG uptake values andperformed correlation analyses between FDG uptakechanges and the respective changes in neuropsycho-logical test scores.

RESULTS

Of the 43 patients meeting initial inclusion criteria,39 were randomly assigned to IG- or CG treatmentarms, with exclusion of one patient due to concur-rent physical illness, and three patients due to lackof compliance (less than 50% presence at the inter-vention sessions). Of these 39 patients, two aMCIpatients dropped out from the IG arm because ofnew concurrent physical illness, or lack of compli-

Table 1Demographic data of IGs and CGs subjects (n = 36)

Igs CGs t-Test

Number 17 19AD 8 7aMCI 9 12Age (SD) 74.5 (8.6) 72.0 (7.1) n.s.

in yrs. 53.2–88.6 62.4–83.7min–max

Gender f / m 7/10 9/10Apo E4 12 11

allele (atleast one)

MMSE min–max 26.4 (2.4) 26.1 (1.6) n.s.22–30 23–29

Education 12.2 (3.4) 13.4 (4.1) n.s.(SD) in yrs. 8–17 11–21.5min–max

aMCI = amnestic mild cognitive impairment; AD = mildAlzheimer’s Disease; IGs = cognitive intervention groups; CGs =active control groups; min–max = ranges of scores; SD = standarddeviation.∗t-Test results refer to post-hoc comparisons testing the differencesof means in the adjacent columns

ance. Thirty-seven participants completed the study,with elimination of one aMCI patient from the IGarm due to poor PET image quality. Concurrent withthe randomization process there was good matchingbetween IG- and CG patient groups (IGs and CGs) fordegree of cognitive impairment, education, age, gen-der, and for Apo-E genotype. All mild AD patients andone aMCI patient in the IG arm were on antidemen-tia medication (acetylcholinesterase inhibitor and/orMemantine) at stable doses since at least since threemonths prior to study start and throughout the interven-tion/control condition until study end. Table 1 showsbaseline demographic and clinical characteristics of allpatients included in the PET data analysis.

Neuropsychological cognitive outcome

To detect intervention-related effects of the stage-specific cognitive intervention in the respective patientgroups, we conducted a series of ANCOVAs with treat-ment (intervention treatment arm, IG, versus controltreatment arm, CG), and progression (baseline versusend-of-study) as independent variables separately forAD and aMCI patients. The primary outcome measuresMMSE and ADAS-cog scores served as dependentmeasures. Furthermore, we entered educational level(years of schooling) and patients‘ age as covariates intothe analysis.


Table 2Change from baseline in measures of efficacy in the cognitive intervention subgroups (IGAD; IGMCI) and in the active control subgroups (CGAD;

CGMCI) regarding MMSE and ADAS-cog; standard deviation in ()

Target-variate Timepoint aMCI AD

IGMCI CGMCI ANCOVA** IGAD CGAD ANCOVA**

ADAS-cog Baseline 8.8 (3.1) 9.8 (4.3) Interaction Treatment 12.1 (5.3) 16.4 (4.8) Treatmentmin–max 3–13 5–19 and Progression 6–21 13–23 (F1,11 = 4.4,

(F1,17 = 4.7, p = .045, η2 = .22) p = .06, η2 = .282)Study end 7.4 (3.2) 11.7 (5.6) 11.4 (6.0) 16.4 (4.9)

t-Test* n.s. t(11) = 2.8, p = .02 n.s. n.s.MMSE Baseline 28.1 (1.6) 26.8 (1.5) Interaction Treatment 24.5 (1.6) 25.3 (1.5) n.s.

min–max 25–30 24–29 and Progression 22–27 23–27(F1,17 = 4.3, p = .05, η2 = .21)Main Effect Treatment(F1,17 = 6.8, p = 0.02, η2 = .29)

Study end 28.3 (1.2) 26.0 (1.3) 25.0 (2.7) 24.4 (2.4)t-Test* n.s t(11) = 2.8, p = .02 n.s. n.s.

aMCI = amnestic mild cognitive impairment; AD = mild Alzheimer’s Disease; ADAS-cog = Alzheimer’s Disease Assessment Scale;MMSE = Mini Mental State Examination; IGs = cognitive intervention groups; CGs = active control groups.;∗t-Test results refer to post-hoc comparisons testing the differences of means in the adjacent rows and columns.; **ANCOVA: univariate analysisof covariance; reported effects refer to the interaction between treatment and progression which reflects whether a change of the dependentvariables from baseline to follow-up is quantitatively different in the two treatment arms.

For AD patients, only the treatment factor app-roached significance for ADAS-cog scores (F1,11 =4.4, p = .06, η2 = .282). For aMCI patients, we deter-mined a marginally significant interaction effectbetween treatment and progression for ADAS-cog(F1,17 = 4.7, p = .045) and for MMSE (F1,17 = 4.3,p = .05). Furthermore, the analysis revealed a maineffect for treatment regarding MMSE (F1,17 = 6.8,p < .01). Post-hoc t-tests suggest that interaction effectsbetween treatment and progression occurred mainlydue to performance decline in the CGMCI. See Table 2for details.

FDG-PET data

All patients relative to elderly non-demented con-trols showed reduced brain FDG uptake at baseline,mainly in the bilateral temporo-occipital associationcortex (left worse than right), left temporal cortex,bilateral posterior cingulate cortices and precuneus, aswell as in left prefrontal cortex (Fig. 2).

MCI subgroups of the CGs and of the IGs (CGMCIand IGMCI) showed similar baseline patterns ofreduced brain FDG uptake, involving parieto-temporalcortex, posterior cingulate cortex, precuneus and pre-frontal cortex (Fig. 3). We omit reporting anatomicalcoordinates, since this finding matches the typicalanatomical distribution of AD-associated changes [30-32].

Fig. 2. Baseline pattern of reduced FDG uptake in all patients(n = 36) as compared to non-demented elderly controls (n = 11);(P < 0.005 uncorrected).


Fig. 3. Baseline patterns of reduced FDG uptake in a) IGMCI (n = 9) and in b) CGMCI (n = 12), as compared to non-demented elderly controls(n = 11); (P < 0.005 uncorrected).

There were no significant differences in baselinebrain FDG uptake between the respective CGs- andIGs.

After six months, the AD subgroup of CGs (CGAD)showed widespread bilateral declines in FDG uptake inparieto-temporal and parieto-occipital cortices as wellas in left prefrontal cortex, while the AD subgroup ofIGs (IGAD) showed minor decline in FDG uptake intwo single clusters located in the lingual gyrus and theleft inferior temporal gyrus.

The CGMCI showed widespread bilateral occipito-temporal (right more then left), parietal and prefrontaldecline in FDG uptake, while the IGMCI showed nodecline in FDG uptake (Fig. 4).

“Difference of differences” analyses revealedstrongest attenuated decline in FDG uptake in thesubgroup of aMCI patients in the IGs (IGMCI) in bi-hemispheric cortical areas, including bilateral tempo-ral, prefrontal- and anterior cingulate cortex (Fig. 5 a).Clusters with main peaks surviving the p < 0.001 (unc.)threshold were located in the left anterior temporal poleand the left anterior cingulate gyrus (Table 4, Fig. 6).

The subgroup of AD patients in the IGs (IGAD)revealed a more restricted attenuated decline in FDGuptake in the right temporal- and posterior cingulatecortex (Fig. 5 b), while none of these clusters survivedthe p < 0.001 (unc.) threshold.

Correlation analyses between changes in normal-ized FDG uptake (in those brain regions whichhad showed the highest metabolic attenuation) and

changes in neuropsychological outcome parameters,revealed a single significant correlation for totalMADRS score in the IGMCI (Pearsons correlation coef-ficient: r = −0.61; p = 0.039 one-tailed), such that anincrease in FDG uptake in the left anterior cingulate-and anterior temporal pole was associated with adecrease in total MADRS score. However, aftereliminating the cognition-related MADRS item (con-centration), which had primarily contributed to thetotal MADRS score, only a weak non-significant cor-relation remained (Pearsons correlation coefficientr = −0.398; p = 0.144 one-tailed). Detailed longi-tudinal outcome of secondary neuropsychologicaloutcome parameters (i.e. MADRS, Trail Making Test,etc.) is described elsewhere [12].

DISCUSSION

We used FDG-PET to map effects of a newly devel-oped multicomponent cognitive intervention on brainenergy metabolism in patients with aMCI and mildAD. Upon entering the study, participants were ran-domly assigned to intervention (IGs) or active controlgroups (CGs). In both CGs, significant decline of FDGuptake in AD-typical brain areas during a relativelybrief period of six months, confirmed that resting-stateFDG-PET is a sensitive marker of disease progres-sion [33]. Participation in the cognitive interventionprogram imparted cognitive benefits in the aMCI sub-


Fig. 4. Relative decline of brain FDG uptake in aMCI and AD- CGs and IGs after six months; (p < 0.005 uncorrected).

R R

a b

Fig. 5. Attenuated decline of brain FDG uptake after six months a) in the IGMCI relative to the CGMCI, b) in the IGAD relative to the CGAD;(p < 0.005 uncorrected).

group, which were reflected by an attenuated declinein cerebral FDG uptake relative to that seen in therespective active control group.

Upon entry in the study, the whole sample ofpatients had significantly reduced FDG uptake relativeto a group of non-demented elderly control subjects.


Table 3Regions of significantly attenuated decline in brain FDG uptake after

six months in the IGMCI relative to the CGMCI

Location Talairach Peak P Clusterand Tournoux z-value uncorr. extension

BA x y z

L, Ant. 38 −32 10 −42 3.39 0.000 152TemporalPole

L, Ant. 32 −10 28 20 3.37 0.000 1036CingulateGyrus

Cluster extension represents the number of contiguous voxels sur-passing the threshold of p < 0.001. Bold markings delineate a clusterand the peak z-value within the cluster. Associated anatomic struc-tures are indicated, along with designations of Brodman area (BA);R = right, L = left.

Main differences were seen in brain regions typicallyimpaired in AD, including the bilateral temporo-occipital association cortices, left temporal cortex,bilateral posterior cingulate cortices and precuneus,as well as in left prefrontal cortex (Fig. 2). This pat-tern confirms previous FDG-PET results in aMCI ormild to moderately diseased AD patients [31, 34, 35],suggesting that our stringent clinical inclusion criteriawere effective in selecting a representative patient sam-ple. Demographics indicated that the patient groupshad similar education background, Apolipoprotein E4allele status and similar baseline cognitive function, asassessed by MMSE and ADAS-cog scores at onset ofthe intervention.

Longitudinal FDG-PET evaluation in both CGs(AD and aMCI) after six-months revealed widespreadbilateral decline in metabolism throughout AD-typical

cortical areas (Fig. 4), consistent with the progressionof the clinical scores of cognitive function (MMSE andADAS-cog). Peaks in the declining FDG uptake over-lapped with the pattern of hypometabolism at baselinerelative to FDG uptake in the non-demented controlgroup, consistent with an on-going disease processduring only six months, as reported in previous PETstudies with one-year follow-up [9, 35]. In those one-year follow-up studies, FDG uptake was normalized tothe global mean value, which we have shown to resultin spurious detection of metabolic changes, whicharise from bias due to undetected but real declinesin metabolism [36]. Instead, we normalized regionalFDG uptake to that measured in brain regions defineda posteriori to be unaffected, relative to the resultsin the non-demented elderly control group [23]. Thea posteriori method is more sensitive than the globalmean approach in detecting disease-related metabolicdisturbances in mild-to-moderate stages of neurode-generative disorders [24, 25], especially in studies ondementia [23, 26–28].

In contrast to the widespread six month declines inFDG uptake seen in both CGAD and CDMCI (Fig. 4),we saw in the IGAD discrete decline with two singlesignificant clusters located in the lingual gyrus and theleft inferior temporal gyrus, while the IGMCI showedno decline at all (Fig. 4). The latter finding stronglysuggests positive effects of the cognitive interventionprogram on brain energy metabolism in the MCI sub-group. The relative preservation of normalized FDGuptake during six months especially in the IGMCI con-curs with our clinical findings showing a significantchange in global cognitive status, which seemed to be

Fig. 6. Overlay image on transaxial slices of an averaged MRI data set of 152 healthy subjects, showing peak clusters of attenuated decline ofbrain FDG uptake after six months in the left anterior cingulate cortex and the left anterior temporal pole in the IGMCI relative to the CGMCI.Radiological convention (Left is Right); set at >30 contiguous voxels passing the significance threshold p < 0.005 uncorrected.


driven by performance stabilisation in the IGMCI, con-trasted with declining performance in the CGMCI (seeTable 2). Given the progressive nature of AD and itsprodromal stage aMCI, the present suggestion of cog-nitive stabilization may emerge as a beneficial effectof the cognitive intervention.

In order to test more objectively for time-dependent effects of the cognitive intervention onbrain metabolism, we made bidirectionally contrasted“difference of differences” analyses of the FDG-PETresults in the aMCI and AD subgroups of the IGs andCGs. These SPM-based approaches revealed attenu-ated metabolic decline during six months of cognitiveintervention mainly in the IGMCI subgroup. Instead,at the same statistical threshold the opposite contrastsrevealed no attenuation of declining metabolism in theCGMCI relative to the IGMCI (or the respective ADsubgroups), supporting the real presence of an atten-uating effect on FDG decline in aMCI patients whichhad received the cognitive interventions.

The effect of attenuated metabolic decline was morewidespread and pronounced in the IGMCI than in theIGAD (Fig. 5 a,b), which is in line with the results fromthe clinical cognitive parameters (MMSE, ADAS-cog.), showing significant changes only in the aMCIsubgroup (Table 2). As mentioned before, changes inclinical cognitive parameters seemed to be attributedprimarily to cognitive decline in the CGMCI, raisingthe question about whether there were a greater per-centage of aMCI subjects with prodromal AD in thecontrol group who may have deteriorated over the 6month follow-up. This possibility can be denied. Groupcomparisons of IGMCI and CGMCI relative to elderlynon-demented control subjects at baseline, revealed forboth MCI subgroups a similar distributional patternof FDG hypometabolism, involving AD-typical brainregions (see Fig. 3). The latter finding strongly sug-gests an inclusion of aMCI patients in a preclinical ADstage and is in line with our stringent MCI-diagnosisinclusion criteria. Furthermore, a number-needed-totreat (NNT) analysis in the aMCI subpopulation indi-cated effectiveness of the cognitive intervention, asimprovement by means of 4-point change on ADAS-cog occurred only in the IGMCI, whereas cognitivedecline appeared in both IG- and CGMCI [12].

In the aMCI subgroup we were able to map thepeak areas of metabolic attenuation to the left ante-rior temporal pole and anterior cingulate cortex (ACC).The latter structure belongs to the fronto-limbic net-work and is well-known for its role in emotional andmotivational control [37] as well as in attentional pro-

cessing [38], [39], [40]. To explore if mood-relatedeffects might have lead to the observed PET effects, weperformed an additional correlation analysis betweenthe changes in normalized FDG uptake in the leftanterior cingulate- as well as left anterior tempo-ral pole, and the changes in MADRS score in theIGMCI. This analysis revealed a single significant cor-relation for total MADRS score in the IGMCI, suchthat an increase in FDG uptake in the left anteriorcingulate- and superior temporal gyrus was associ-ated with a decrease in total MADRS score. However,after eliminating the cognition-related MADRS item(concentration), which had primarily contributed to thetotal MADRS score [12], only a weak, non-significantcorrelation remained. Therefore, mood-specific effectsmight not have led to the observed attenuating effectson brain metabolism. However, we cannot rule outthat cognition-related mood stimulating effects (i.e.increased concentration) of the intervention might haveto some extent contributed to the observed PET effects.

Probably due to the limited patient number and therestricted variance of clinical scores in this relativelyshort time period of six months, correlation analy-ses between FDG uptake changes and performancechanges in the global cognitive domains (MMSE,ADAS cog) did not show any significant correlation(data not shown).

We feel that the specific linkage between neuropsy-chological and neurobiological outcome remains to beestablished in future studies in larger patient popula-tions.

Overall, our findings strongly suggest a positiveshort-term effect of the cognitive intervention programon brain energy metabolism in aMCI patients, whichmay manifest in cognitive benefits. We modelled ourstudy on an earlier FDG-PET follow-up study inves-tigating effects of a 14-day healthy longevity lifestyleprogram on cognition and cerebral energy metabolismin a group of non-demented subjects with mild age-related memory complaints [11]. In that study, theintervention group had after two weeks a 5% relativedecrease in normalized FDG uptake in the left dorso-lateral prefrontal cortex, which the authors interpretedto reveal an effect of training on cognitive efficiency ofa brain region involved in working memory. Unlike inthat study, we selected patients with objective cognitiveimpairment, suggesting that our findings of attenuateddeclines relate to moderation of a pathological pro-cess. In the only other FDG-PET study of cognitivetraining in AD patients [10], there was a six-monthdecline in resting-state cerebral glucose metabolism,


most notably in the temporo-parietal region, in a groupof AD patients who had only received social sup-port. In that study, FDG-PET changes evoked by avisual recognition task served as an indicator of ther-apeutic efficacy in AD. The AD patients who hadreceived cognitive training with and without pharma-cological treatment had a more distinct pattern of FDGactivations after six months, whereas the subgroupwith both interventions showed some improvement incognitive performance, as measured by the MMSE.Despite diverse conceptual and methodological differ-ences between that and the present studies, there isagreement that FDG-PET can detect effects of inter-vention on cerebral metabolism during six monthsprogression of impaired cognition.

LIMITATIONS

The relatively small number of participants in thisstudy limits its statistical power. Furthermore, thisstudy design reveals only group differences in FDGuptake and cognitive scores, which may however beoffset by the benefits of the prospective setting, inwhich the groups were well-matched for demograph-ics and baseline metabolism. Direct result comparisonsbetween the AD and MCI subgroups might be con-founded by differences in the intervention design andshould be interpreted with caution.

We allowed a more permissive statistical thresholdof p < 0.005 (uncorrected) for the visualisation of PETresults. However, application of the same permissivethreshold for the respective negative SPM t-contrastsdid not bear any significant results, which would haveindicated an increased risk of false positive results.

ACKNOWLEDGMENTS

The authors thank Dr. Paul Cumming for criticalreview of the manuscript.

FINANCIAL DISCLOSURES

The authors report no biomedical financial interestsor potential conflicts of interest.

REFERENCES

[1] Buschert V, Bokde AL, Hampel H (2010) Cognitive interven-tion in Alzheimer disease. Nat Rev Neurol 6, 508-517.

[2] Valenzuela M, Sachdev P (2009) Can cognitive exercise pre-vent the onset of dementia? Systematic review of randomized

clinical trials with longitudinal follow-up. Am J Geriatr Psy-chiatry 17, 179-187.

[3] Valenzuela MJ, Sachdev P (2006) Brain reserve, dementia: asystematic review. Psychol Med 36, 441-454.

[4] Belleville S (2008) Cognitive training for persons with mildcognitive impairment. Int Psychogeriatr 20, 57-66.

[5] Sitzer DI, Twamley EW, Jeste DV (2006) Cognitive trainingin Alzheimer’s disease: a meta-analysis of the literature. ActaPsychiatr Scand 114, 75-90.

[6] Olazaran J, Reisberg B, Clare L, Cruz I, Pena-Casanova J,Del Ser T, Woods B, Beck C, Auer S, Lai C, Spector A, FazioS, Bond J, Kivipelto M, Brodaty H, Rojo JM, Collins H, TeriL, Mittelman M, Orrell M, Feldman HH, Muniz R (2010)Nonpharmacological therapies in Alzheimer’s disease: A sys-tematic review of efficacy. Dement Geriatr Cogn Disord 30,161-178.

[7] Herholz K (2003) PET studies in dementia. Ann Nucl Med17, 79-89.

[8] Reiman EM, Caselli RJ, Yun LS, Chen K, Bandy D,Minoshima S, Thibodeau SN, Osborne D (1996) Preclinicalevidence of Alzheimer’s disease in persons homozygous forthe epsilon 4 allele for apolipoprotein E. N Engl J Med 334,752-758.

[9] Alexander GE, Chen K, Pietrini P, Rapoport SI, Reiman EM(2002) Longitudinal PET Evaluation of Cerebral MetabolicDecline in Dementia: A Potential Outcome Measure inAlzheimer’s Disease Treatment Studies. Am J Psychiatry 159,738-745.

[10] Heiss WD, Kessler J, Slansky I, Mielke R, Szelies B, HerholzK (1993) Activation PET as an instrument to determine thera-peutic efficacy in Alzheimer’s disease. Ann N Y Acad Sci 695,327-331.

[11] Small GW, Silverman DH, Siddarth P, Ercoli LM, Miller KJ,Lavretsky H, Wright BC, Bookheimer SY, Barrio JR, PhelpsME (2006) Effects of a 14-day healthy longevity lifestyle pro-gram on cognition and brain function. Am J Geriatr Psychiatry14, 538-545.

[12] Buschert VC, Friese U, Teipel SJ, Schneider P, Merensky W,Rujescu D, Moller HJ, Hampel H, Buerger K (2011) Effectsof a newly developed cognitive intervention in amnestic mildcognitive impairment and mild Alzheimer’s disease:. A PilotStudy. J Alzheimers Dis, doi10.3233/JAD-2011-100999.

[13] Lachman ME, Weaver SL, Bandura M, Elliott E, LewkowiczCJ (1992) Improving memory and control belief through cog-nitive restructuring and self-generated strategies. J Gerontol47, P293-P299.

[14] McKhann G, Drachman D, Folstein M, Katzman R, PriceD, Stadlan EM (1984) Clinical diagnosis of Alzheimer’s dis-ease: report of the NINCDS-ADRDA Work Group underthe auspices of Department of Health and Human ServicesTask Force on Alzheimer’s Disease. Neurology 34, 939-944.

[15] Scheltens P, Barkhof F, Leys D, Pruvo JP, Nauta JJ, Verm-ersch P, Steinling M, Valk J (1993) A semiquantative ratingscale for the assessment of signal hyperintensities on magneticresonance imaging. J Neurol Sci 114, 7-12.

[16] Stern Y (2006) Cognitive reserve and Alzheimer disease.Alzheimer Dis Assoc Disord 20, 112-117.

[17] Reisberg B, Franssen EH, Hasan SM, Monteiro I, BoksayI, Souren LE, Kenowsky S, Auer SR, Elahi S, Kluger A(1999) Retrogenesis: clinical, physiologic, and pathologicmechanisms in brain aging, Alzheimer’s and other dement-ing processes. Eur Arch Psychiatry Clin Neurosci 249(3),28-36.


[18] Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG,Kokmen E (1999) Mild cognitive impairment: clinical char-acterization and outcome. Arch Neurol 56, 303-308.

[19] Perry RJ, Hodges JR (1999) Attention and executive deficits inAlzheimer’s disease. A critical review. Brain 122(3), 383-404.

[20] Folstein M, Folstein S, McHugh P (1975) “Mini Mental State”A practical method for grading the cognitive state of patientsfor the clinician. J Psychiatric Res 12, 189-198.

[21] Rosen WG, Mohs RC, Davis KL (1984) A new rating scalefor Alzheimer’s disease. Am J Psychiatry 141, 1356-1364.

[22] Friston KJ, Frith CD, Liddle PF, Frackowiak RS (1991) Com-paring functional (PET) images: the assessment of significantchange. J Cereb Blood Flow Metab 11, 690-699.

[23] Yakushev I, Hammers A, Fellgiebel A, Schmidtmann I,Scheurich A, Buchholz HG, Peters J, Bartenstein P, Lieb K,Schreckenberger M (2009) SPM-based count normalizationprovides excellent discrimination of mild Alzheimer’s diseaseand amnestic mild cognitive impairment from healthy aging.Neuroimage 44, 43-50.

[24] Borghammer P, Aanerud J, Gjedde A (2009) Data-drivenintensity normalization of PET group comparison studies issuperior to global mean normalization. Neuroimage 46, 981-988.

[25] Borghammer P, Cumming P, Aanerud J, Forster S, GjeddeA (2009) Subcortical elevation of metabolism in Parkinson’sdisease - A critical reappraisal in the context of global meannormalization. Neuroimage 47, 1514-1521.

[26] Yakushev I, Landvogt C, Buchholz HG, Fellgiebel A,Hammers A, Scheurich A, Schmidtmann I, Gerhard A,Schreckenberger M, Bartenstein P (2008) Choice of referencearea in studies of Alzheimer’s disease using positron emis-sion tomography with fluorodeoxyglucose-F18 . PsychiatryRes 164, 143-153.

[27] Forster S, Teipel S, Zach C, Rominger A, Cumming P, FougereCL, Yakushev I, Haslbeck M, Hampel H, Bartenstein P,Burger K (2010) FDG-PET mapping the brain substrates ofvisuo-constructive processing in Alzheimer s disease. J Psy-chiatr Res 44, 462-469.

[28] Forster S, Vaitl A, Teipel SJ, Yakushev I, Mustafa M, laFougere C, Rominger A, Cumming P, Bartenstein P, Ham-pel H, Hummel T, Buerger K, Hundt W, Steinbach S (2010)Functional representation of olfactory impairment in earlyAlzheimer’s disease. J Alzheimers Dis 22, 581-591.

[29] Lancaster JL, Woldorff MG, Parsons LM, Liotti M, FreitasCS, Rainey L, Kochunov PV, Nickerson D, Mikiten SA, FoxPT (2000) Automated Talairach atlas labels for functionalbrain mapping. Hum Brain Mapp 10, 120-131.

[30] Ishii K, Willoch F, Minoshima S, Drzezga A, Ficaro EP, CrossDJ, Kuhl DE, Schwaiger M (2001) Statistical brain map-ping of 18F-FDG PET in Alzheimer’s disease: validation ofanatomic standardization for atrophied brains. J Nucl Med 42,548-557.

[31] Minoshima S, Giordani B, Berent S, Frey KA, Foster NL,Kuhl DE (1997) Metabolic reduction in the posterior cingulatecortex in very early Alzheimer’s disease. Ann Neurol 42, 85-94.

[32] Drzezga A, Riemenschneider M, Strassner B, Grimmer T,Peller M, Knoll A, Wagenpfeil S, Minoshima S, Schwaiger M,Kurz A (2005) Cerebral glucose metabolism in patients withAD and different APOE genotypes. Neurology 64, 102-107.

[33] Forster S, Grimmer T, Miederer I, Henriksen G, YousefiBH, Graner P, Wester HJ, Forstl H, Kurz A, DickersonBC, Bartenstein P, Drzezga A (2011) Regional expansionof hypometabolism in AD follows amyloid-deposition withtemporal delay. Biological Psychiatry (in press).

[34] Minoshima S, Frey KA, Koeppe RA, Foster NL, Kuhl DE(1995) A diagnostic approach in Alzheimer’s disease usingthree-dimensional stereotactic surface projections of fluorine-18-FDG PET. J Nucl Med 36, 1238-1248.

[35] Drzezga A, Lautenschlager N, Siebner H, Riemenschnei-der M, Willoch F, Minoshima S, Schwaiger M, Kurz A(2003) Cerebral metabolic changes accompanying conversionof mild cognitive impairment into Alzheimer’s disease: a PETfollow-up study. Eur J Nucl Med Mol Imaging 30, 1104-1113.

[36] Borghammer P, Jonsdottir KY, Cumming P, Ostergaard K,Vang K, Ashkanian M, Vafaee M, Iversen P, Gjedde A (2008)Normalization in PET group comparison studies – the impor-tance of a valid reference region. Neuroimage 40, 529-540.

[37] Wagner G, Koch K, Schachtzabel C, Schultz CC, Sauer H,Schlosser RG (2011) Structural brain alterations in patientswith major depressive disorder and high risk for suicide: Evi-dence for a distinct neurobiological entity? Neuroimage 54,1607-1614.

[38] Haupt S, Axmacher N, Cohen MX, Elger CE, Fell J (2009)Activation of the caudal anterior cingulate cortex due to task-related interference in an auditory Stroop paradigm. HumBrain Mapp 30, 3043-3056.

[39] Crottaz-Herbette S, Menon V (2006) Where and when theanterior cingulate cortex modulates attentional response:combined fMRI and ERP evidence. J Cogn Neurosci 18,766-780.

[40] Margulies DS, Kelly AM, Uddin LQ, Biswal BB, CastellanosFX, Milham MP (2007) Mapping the functional connectivityof anterior cingulate cortex. Neuroimage 37, 579-588.

Documents

6 Month Cognitive Intervention Amnestic Mci and Mild Ad