Thalamic atrophy and cognition in multiple sclerosis

DOI 10121201wnl000027699217011b52007691213-1223 Neurology

M K Houtchens RHB Benedict R Killiany et al Thalamic atrophy and cognition in multiple sclerosis

This information is current as of September 17 2007

httpwwwneurologyorgcontent69121213fullhtmllocated on the World Wide Web at

The online version of this article along with updated information and services is

0028-3878 Online ISSN 1526-632Xsince 1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN

reg is the official journal of the American Academy of Neurology Published continuouslyNeurology

Thalamic atrophy and cognition inmultiple sclerosis

MK Houtchens MDRHB Benedict PhDR Killiany PhDJ Sharma MDZ Jaisani MDB Singh MDB Weinstock-Guttman MD

CRG GuttmannMD

R Bakshi MD

ABSTRACT

Objectives Recent studies have indicated that brain atrophy is more closely associated with cog-nitive impairment in multiple sclerosis (MS) than are conventional MRI lesion measures Enlarge-ment of the third ventricle shows a particularly strong correlation with cognitive impairmentsuggesting clinical relevance of damage to surrounding structures such as the thalamus Previ-ous imaging and pathology studies have demonstrated thalamic involvement in MS In this studywe tested the hypothesis that thalamic volume is lower in MS than in normal subjects and thatthalamic atrophy in MS correlates with cognitive function

MethodsWe studied 79 patients with MS and 16 normal subjects A subgroup of 31 MS subjectsunderwent cognitive testing The thalamus was segmented in whole from three-dimensional MRIscans We also determined whole brain atrophy (brain parenchymal fraction) third ventricularwidth and whole brain T2-weighted (fluid-attenuated inversion recovery) hyperintense T1hypointense and gadolinium-enhanced lesion volumes

Results Normalized thalamic volume was 168 lower in the MS group (p 00001) vs controlsCognitive performance in all domains was moderately to strongly related to thalamic volume in theMS group (r 0506 to 0724 p 0005) and thalamic volume entered and remained in allregression models predicting cognitive performance Thalamic volume showed a weak relation-ship to physical disability score (r 0316 p 0005)

Conclusion These findings suggest that thalamic atrophy is a clinically relevant biomarker of theneurodegenerative disease process in multiple sclerosis Neurologyreg 2007691213ndash1223

GLOSSARYBDI-FS Beck Depression InventoryndashFast Screen for Medical Patients BPF brain parenchymal fraction BVMT-R-D

Brief Visuospatial Memory Test (Delayed Recall) BVMT-R-TR Brief Visuospatial Memory Test (Total Recall) COWAT

Controlled Oral Word Association Test CVLT-II California Verbal Learning Test second edition CVLT-II-D CaliforniaVerbal Learning Test (Delayed Recall) CVLT-II-TR California Verbal Learning Test (Total Recall) EDSS Expanded Dis-ability Status Scale FLAIR fluid-attenuated inversion recovery FOV field-of-view JLO Judgment of Line OrientationTest MS multiple sclerosis NC normal controls NSA number of signal averages COWAT Controlled Oral WordAssociation Test PASAT Paced Auditory Serial Addition Test SPGR spoiled gradient recall SDMT Symbol DigitModalities Test TE echo time TR repetition time

Cognitive dysfunction is a disabling manifestation of multiple sclerosis (MS) affectingapproximately 50 of patients1 MS-related cognitive impairment mainly affects atten-tion information processing speed and episodic memory23 The severity of these deficitsmay reflect the extent of lesions and the degree of tissue loss and disorganization outsidelesions4

MS is now recognized as a destructive disease in that CNS atrophy is common occursearly in the disease course and is typically progressive56 The underlying substrate ofatrophy includes axonal and neuronal loss78 Recent data indicate that brain atrophy

From the Departments of Neurology (MKH JS ZJ CRGG RB) and Radiology (RK CRGG RB) Brigham and WomenrsquosHospital Harvard Medical School Partners MS Center Boston MA and Jacobs Neurological Institute (RHBB BS BW-G) Universityat Buffalo NY

Supported in part by research grants from the NIH (NS42379-01mdashDr Bakshi) and National Multiple Sclerosis Society (RG 3574A1 and RG3705A1mdashDrs Bakshi and Guttmann) and a Clinical Investigator Training Program from HarvardMassachusetts Institutes of TechnologyHealth Sciences and Technology Beth Israel Deaconess Medical Center Pfizer and Merck amp Co (Dr Houtchens)

Disclosure The authors report no conflicts of interest

Address correspondence andreprint requests to Dr RohitBakshi Brigham and WomenrsquosHospital Harvard MedicalSchool 77 Avenue LouisPasteurndashHIM 730 Boston MA02115rbakshibwhharvardedu

Copyright copy 2007 by AAN Enterprises Inc 1213

involves both gray and white matter9-13

Both the cortical and subcortical gray mat-ter structures develop atrophy8914-18

A growing body of data indicates thatcognitive impairment in MS is related inpart to damage of the subcortical gray mat-ter areas4 This includes studies showingstrong associations between impaired in-formation processing speed and the bicau-date ratio19 T2 hypointensity of deep graymatter nuclei20 and enlargement of the lat-eral ventricles21 Enlargement of the thirdventricle shows a particularly strong rela-tionship to cognitive impairment even afteraccounting for the influence of general le-sion measures and whole brain atrophy22

The close proximity of the thalamus to thethird ventricle and the correlation betweenthalamic volume and third ventricle widthsuggest a role for the thalamus in MS-associated cognitive impairment8

The thalamus is involved in limbic cir-cuitry and mediates or regulates many cog-nitive functions2324 Numerous studieshave shown damage to the thalamus inMS such as T2 hypointensity2526 hypome-tabolism27 increased diffusivity28 de-creased magnetization transfer ratio2930

decreased neuronal integrity and loss ofneurons8 and macroscopic atrophy831

Furthermore dysfunction of the thalamushas been linked to cognitive impairment27

and fatigue32 in patients with MSWe focused on thalamic volume and its

relationship to cognitive function in MSWe examined three hypotheses 1) tha-lamic volume is lower in patients com-pared with healthy controls 2) thalamic

atrophy is related to cognitive impairmentand 3) thalamic volume accounts for moreof the variance in cognitive function thando other brain MRI measures in the MSgroup including T1 hypointense and T2hyperintense lesion load whole brain atro-phy and third ventricle width

METHODS Patients Demographic characteristics aresummarized in table 1 We studied 79 patients who met diag-nostic criteria for MS33 and 16 normal subjects Age rangeswere 21 to 63 years in theMS cohort and 21 to 58 years in thenormal cohort The participants were enrolled consecutivelyfrom a community-based university-affiliated MS clinicThe mean ( SD) Expanded Disability Status Scale (EDSS)score34 was 34 20 (median 30 range 0 to 8) and thetimed 25-ft walk35 was 74 50 seconds Disease durationwas 97 71 years (range up to 28 years) The diseasecourse was relapsingndashremitting in 62 patients (784) sec-ondary progressive in 16 patients (203) and primary pro-gressive in 1 patient (13) Exclusion criteria were 1) acurrent or past major medical neurologic or psychiatric dis-order (other than MS related) 2) previous or current sub-stance abuse and 3) MS relapse or corticosteroid use within4 weeks preceding MRI or cognitive testing All MS patientsand controls underwent MRI

A subgroup of 31 MS patients (aged 22 to 55 years) un-derwent neuropsychological evaluation within 6 months ofMRI The EDSS score for this cohort was 37 22 (median25 range 1 to 8) the disease duration was 11 67 yearsand the timed 25-ft walk was 81 68 seconds The diseasecourse was relapsingndashremitting in 26 patients (839) andsecondary progressive in 5 patients (161) The patientswere selected for neuropsychological testing with the follow-ing additional selection criteria 1) willing to undergo neuro-psychological testing and 2) availability of a caregiver Thusthese patients were identified consecutively not based on thepresence of cognitive symptoms By two-way analysis therewere no differences between the MS patients not undergoingcognitive testing and those in the cognitive MS cohort aspertains to age (p 0765) sex (p 0793) disease duration(p 0165) EDSS score (p 0275) disease course (p

0532) or timed 25-ft walk (p 0820) Patients in the cogni-tiveMS cohort had demographics not different from the nor-mal controls (p 005) by two-way analysis (table 1) exceptthat they had fewer years of education (145 22 vs 165

Table 1 Demographic characteristics

All MS patients Cognitive subgroup Normal controls

No of subjects 79 31 16

Age mean SD y 420 96 424 84 465 93

Women n () 60 (76) 23 (74) 12 (75)

Men n () 19 (24) 8 (26) 4 (25)

The demographic characteristics were not different (p 005) among the three groups There were several comparisonsmade three-way analyses comparing normal controls vs multiple sclerosis (MS) patients in the cognitive subgroup vs MSpatients not in the cognitive subgroup and two-way analyses comparing normal controls vs MS patients in the cognitivesubgroup and MS patients in the cognitive subgroup vs MS patients not in the subgroup Also a two-way comparison wasmade between overall MS patients and normal controls

1214 Neurology 69 September 18 2007

24 years p 001) Disease course in the MS patients re-flected the expected distribution of a typical MS sample ofpatients with this age range and disease duration36

Nine patients (114) had not received immunomodu-lating therapy within 6 months of enrollment Fifty-five pa-tients (696) were receiving IM interferon beta-1a weeklyThe remaining patients were treated with either glatirameracetate (four patients) combined interferon with oral immu-nosuppressant therapy (one patient) IV immunoglobulin(two patients) or monthly IV methylprednisolone (three pa-tients) or were switched between different injectable immu-nomodulating agents (five patients) None of the patientsreceived IV chemotherapeutic agents

MRI Brain MRI was performed on each subject using thesame scanning protocol on a General Electric 4xLx 15-Tscanner (Milwaukee WI) The MRI protocol relevant forquantitative analysis consisted of a coronal three-dimensional T1-weighted spoiled gradient recall (SPGR) ax-ial fluid-attenuated inversion recovery (FLAIR) and axialT1-weighted conventional spin-echo pre-postgadoliniumT1 IV injection of 01 mmolkg gadolinium preceded post-contrast imaging by 5 minutes The details of the pulse se-quences were as follows for SPGR field-of-view (FOV) 24 18 cm matrix 256 256 70 slices 25 mm thicknessrepetition time (TR) 24 msec echo time (TE) 7 msec flipangle 30 deg number of signal averages (NSA) 1 scan time545 for FLAIR FOV 24 24 matrix 192 256 28 slices 5mm thickness TR 8002 TE 128 inversion time 2000 msececho train length 22 NSA 1 scan time 416 and for T1 spin-echo FOV 24 18 matrix 192 256 24 slices 5 mm thick-ness TR 600 TE 9 NSA 2 scan time 256 There were nointerslice gaps in any of the sequences

Image analysis Analysis was performed using the soft-ware package Jim (version 30 Xinapse Systems LtdNorthants UK httpwwwxinapsecom) by trained techni-cians who were blind to the clinical data As previously de-scribed37 hypointense lesions on T1-weighted images weresegmented using an edge-finding tool FLAIR lesion volumewas determined using a thresholding procedure37 Third ven-

tricle width was measured from T1-weighted axial images aspreviously described22 To assess whole brain atrophy a nor-malized measure of whole brain volume brain parenchymalfraction (BPF) was obtained from the axial T1-weightedspin-echo images3738 BPF was the ratio of brain parenchymalvolume (tissue compartment) to the total intracranialvolume

The whole thalamus was traced from the coronal three-dimensional images by trained technicians with verificationby an experienced observer (MH) none of whom wereaware of clinical information The thalamic boundaries weredetermined using an edge-finding tool with manual adjust-ments as necessary (figure 1) Raw thalamic volumes werenormalized within each subject as a ratio to the intracranialvolume The resulting normalized thalamic volume was re-ferred to as the thalamic fraction Only three subjects hadovert lesions in the thalamus (hypointensities) on the SPGRimages Only three subjects had gadolinium-enhancing le-sions thus this measure was excluded from subsequentanalysis

Reproducibility of MRI data Ten randomly chosensubjects (5 patients with MS and 5 normal controls) had tha-lamic segmentation repeated by the experienced observer(MH) to determine intrarater reliability (expressed as thecoefficient of variation [COV]) The mean COV for the 10subjects was 54 (89 in the MS group and 20 in thecontrol group) We have already established intrarater COVfor the other MRI measures (17 for T1 hypointense lesionvolume 12 for FLAIR hyperintense lesion volume 52for third ventricle width and 031 for BPF)2237

Cognitive testing Neuropsychological testing was ac-cording to consensus panel recommendations3 The neuro-psychological tests are reliable and valid3940 This batteryknown as the Minimal Assessment of Cognitive Function inMultiple Sclerosis includes the Controlled Oral Word Asso-ciation Test (COWAT)41 Judgment of Line Orientation Test(JLO)41 California Verbal Learning Test second edition(CVLT-II)42 Brief Visuospatial Memory TestndashRevised(BVMT-R)43 Paced Auditory Serial Addition Test(PASAT)44 and Symbol Digit Modalities Test (SDMT)45

The COWAT was administered by the Benton method41

In successive 1-minute trials participants generated as manywords as possible beginning with each of three designatedletters The dependent measure was the total number of cor-rect words over the three trials

The JLO required participants to identify the angle de-fined by two stimulus lines from among those defined by avisual array of lines covering 180 deg Both oral and pointingresponses were allowed The dependent variable was the to-tal number of correct responses over 30 items

The CVLT-II and BVMT-R are both learning and mem-ory tests Both require discrete exposures to new materialfollowed by unaided recall immediately after presentationThere is a 25-minute interval after the final learning trialafter which participants recall the information again withoutfurther exposure to the to-be-learned material Delayed re-call is followed by a yesno forced-choice recognition taskFor the CVLT-II there were five learning trials Examinersread 16 words and asked participants to repeat as manywords as possible The entire word list was repeated eachtime For the BVMT-R the stimulus material was a matrixof six visual designs held before the participant for 10 sec-onds Participants were asked to render the designs using

Figure 1 The thalamus traced in whole fromcoronal three-dimensional MRIscans

A representative slice isshown with the segmentedborders of the thalamus

Neurology 69 September 18 2007 1215

paper and pencil taking as much time as needed Each de-sign received a score of 0 1 or 2 based on accuracy andlocation scoring criteria There were three free-recall trialsfollowed by 25-minute delayed recall and yesno recognitiontrials In this study we considered the following measuresfor each memory test total recall over all learning trials(CVLT-II-TR BVMT-R-TR) and recall after the delay in-terval (CVLT-II-D BVMT-R-D)

In accordance with published guidelines3 we used Raorsquosadaptations1 of the PASAT and SDMT The PASAT in-cluded 60 trials presented at interstimulus intervals of 3 and2 seconds The 3-second version is part of the Multiple Scle-rosis Functional Composite a clinical outcome measurecomposed of quantitative measures of leg upper extremityand cognitive function3546 The dependent measure was thenumber of correct responses from each of two trials Weused only the oral response version of the SDMT Partici-pants were presented with a series of nine symbols eachpaired with a single digit in a key at the top of an 85 11-insheet The remainder of the page presented a pseudo-randomized sequence of symbols Participants responded byvoicing the digit associated with each symbol as quickly aspossible

Depression was assessed using the Beck Depression In-ventoryndashFast Screen for Medical Patients (BDI-FS)47 TheBDI-FS emphasizes thought content (eg negative self-evaluation or guilt) and mood states (eg dysphoria) andavoids assessment of vegetative signs that can occur in medi-cal illness without depression This test has been validated inan MS sample48

Statistical analysis Group comparisons used analysis ofvariance or analysis of covariance All between-group neuro-psychological comparisons controlled for the influence ofeducation (in years) The significance of correlations wastested using the Pearson r statistic Regression modeling wasperformed with a forward selection procedure with p to en-ter 005 and p to exit 010 The linear regression modelscontrolled for the effects of age sex and depression as mea-sured by BDI-FS Specifically all models included age andsex entered and maintained in Block 1 followed by the MRIvariables in Block 2 Including depression in the models didnot alter the results Using this method models were testedin which all candidate MRI variables (thalamic fraction

third ventricular width BPF FLAIR hyperintense lesion vol-ume and T1 hypointense lesion volume) were used to pre-dict neuropsychological tests Models were then repeatedafter controlling for the effects of disease duration Effectssizes were the difference between means divided by thepooled SD49 Because of the exploratory nature of the studyand limitations in statistical power the threshold for signifi-cance among univariate tests was p 005 All of the depen-dent measures were examined for approximation tonormality and found to be normal by the KolmogorovndashSmirnov test (all p values 005) Two of the MRI mea-sures T1 hypointense and FLAIR hyperintense lesionvolumes were positively skewed Because these variableswere independent variables in the regression analyses weelected to not transform these particular distributions Re-garding multicollinearity the independent variables were in-tercorrelated but not inordinately so ie no correlationsbetween independent variables exceed 085 Regarding lin-earity residuals were examined for lack of fit and therewere no clear patterns indicating higher order nonlinearityRegarding homoscedasticity we examined normal probabil-ity plots and z-residual histograms to assess the assumptionof normally distributed residual error at the values of theindependent variables and there were no regression modelswith marked deviation from a normal distribution of resid-ual error

RESULTS MRI variables Because the right andleft thalamic fractions and raw volumes werehighly correlated at r 091 (p 00001) wecalculated a mean value for statistical analysis inorder to control for Type 1 error Absolute tha-lamic volumes were markedly decreased in MSpatients (mean 10324 22523 mm3) vs controls(mean 125819 11387 mm3 p 0001) Thisrepresented a 178 mean difference betweengroups and an effect size49 (d) of 13 This rela-tionship persisted after adjusting thalamic vol-ume for intracranial volume in each subjectthalamic fractions were lower in MS patients(mean 00079 00017) vs healthy subjects (mean00095 000068 p 00001 figure 2 and table2) This represented a 168 mean difference be-tween groups and an effect size of 12 This rela-tionship persisted after adjusting for age and sexand remained significant when right and left thal-ami were analyzed separately Thalamic fractionswere larger in women than in men in the MSgroup (women 00081 00016 men 00070 00019 p 001) This sex difference was notpresent in normal subjects (p 012) There wereno other significant sex-related differences inMRI measures Age was not related to thalamicfractions in the MS (p 044) and control groups(p 028) The difference in mean BPF betweenthe patient and control groups was 26 (p 0001 effect size 06 table 2)

The thalamic fractions and other MRI mea-sures were moderately to strongly intercorrelated

Figure 2 Thalamic atrophy in MSBar heights represent meanand error bars represent SDof thalamic fraction(bilateral thalamic volumenormalized to intracranialvolume) in the healthycontrols (n 16 open bar)and multiple sclerosis (MS)group (n 79 hatched bar)(p 00001)


(table 3) The strongest Pearson correlation wasbetween thalamic fraction and BPF (r 0718p 00001 table 3 and figure 3) Thalamic frac-tion was moderately to strongly correlated withlesion measures and third ventricle width (table3) When comparing all lesion and atrophy-related MRI variables to EDSS score thalamicatrophy showed the strongest correlation(r 0316 p 0005) albeit modest (table 3)When comparing all MRI variables including at-rophy variables with EDSS score third ventriclewidth was selected as the only variable remainingin the most parsimonious model identified by theforward selection procedure predicting EDSSscore (R2 0125 p 0008)

Cognitive performance As expected normal con-trols performed better than MS patients on allneuropsychological tests although this differencewas only significant for BVMT-R-TR and SDMT(table 4) In the MS group thalamic atrophy wasassociated with impairment on tests of processingspeedworking memory and visuospatial memory(figure 4)

Pearson correlation coefficients between allMRI and cognitive tests are shown in table 5Thalamic fraction correlated strongest with allcognitive tests as compared with all other MRIvariables However the other MRI variables alsoshowed moderate to strong correlations with cog-nitive data We could not definitively demon-strate that thalamic atrophy had better

correlations than the other MRI measures iewhen the magnitude of the top three Pearson cor-relation coefficients for each cognitive test werecompared by t test using the method of Blalock50

none of the comparisons were significantRegression modeling results are shown in table

6 Thalamic fraction was the only MRI measurethat entered and remained in regression modelspredicting COWAT (R2 0266 p 005)CVLT-II-TR (R2 0451 p 001) CVLT-II-D(R2 0497 p 0001) BVMT-R-TR (R2

0549 p 0001) BVMT-R-D (R2 0526 p

0001) PASAT (R2 0504 p 0001) andSDMT (R2 0514 p 0001) The model pre-dicting JLO included both thalamic fraction andthird ventricle width (R2 0581 p 0001) Re-peating the analyses controlling for depression(BDI-FS score) age sex and disease duration didnot change any of the results Thus thalamicfraction accounted for the most variance in allmodels predicting neuropsychological testperformance

DISCUSSION Our study showed a significant de-crease in thalamic size in MS patients relative tohealthy controls This relationship was seen forraw (187 difference) and normalized thalamicvolume (thalamic fraction) (168 difference)both of which showed large effect sizes Thalamicsize correlated strongly with brain parenchymalfraction FLAIR and T1 lesion volume and third

Table 2 MRI data on MS patients and control subjects

MRI variable MSndashall subjects (n 79) MSndash cognitive cohort (n 31) Normal controls (n 16)

Thalamic fraction total 00079 00017 00078 00020 00095 000068

Third ventricular width 422 23 459 26 Not assessed

Brain parenchymal fraction 0860 0049 0856 0056 0883 0025

T1 lesion volume mL 12379 24745 18000 32393 Not assessed

FLAIR lesion volume mL 128569 186019 139693 215265 Not assessed

There was no significant group difference between the patients with multiple sclerosis (MS) in the cognitive subgroup and thepatients with MS not in the cognitive subgroup MRI variables are expressed as mean SD Large differences were observedin thalamic volume (p 00001 figure 2) and BPF (p 0001) between the overall MS cohort and normal controlsFLAIR fluid-attenuated inversion recovery

Table 3 Relationship between thalamic atrophy and other MRI data and clinical data in MS patients (n 79)

Thalamic fraction total T1V FLV BPF TvW Age DD EDSSTimed25-ft walk

r Value 0624 0707 0718 0617 0088 0219 0316 0218

p Value 0001 00001 00001 00001 044 0053 0005 0070

Pearson correlation coefficients are shownMS multiple sclerosis T1V T1 hypointense lesion volume FLV fluid-attenuated inversion recovery hyperintense lesionvolume BPF brain parenchymal fraction TvW third ventricular width DD disease duration EDSS Expanded Disabil-ity Status Scale score


ventricular width in MS patients Modest but sig-nificant correlations were seen between thalamicvolume and EDSS scores The most interestingobservation was that thalamic atrophy accountedfor a large amount of variance in predicting cog-nitive performance in patients with MS and en-tered and remained in regression models morefrequently than all other MRI variables includingconventional T1 and T2 lesion measures wholebrain atrophy and third ventricle width How-ever the other MRI measures also showed mod-erate to strong correlations with cognitiveperformance We could not definitively demon-strate that thalamic atrophy had better correla-tions than the other MRI measures Nonethelessthese results highlight the significance of thalamicvolume loss in MS patients

Our findings of 168 decrease in thalamicvolume support previously published results of17 to 25 lower thalamic volumes in MS pa-tients851 The degree of thalamic atrophy is simi-lar to previously reported substantial andselective atrophy of other deep gray matter struc-

tures in MS patients such as caudate nucleus inwhich we reported a 19 lower normalized bi-caudate volume in MS patients compared withcontrols16

Thalamic fraction correlated strongly withwhole brain atrophy (BPF) in our study (r 0718 p 00001) However the absolute differ-ence in BPF between the patient and the controlgroups was less than 3 with only a moderateeffect size suggesting that thalamus may be dis-proportionately vulnerable to the destructive pro-cesses in MS These results agree with datashowing selective atrophy of the caudate nucleusin MS16 and progressive loss of gray matter inparticular deep gray structures in patients withrelapsingndashremitting and secondary progressiveMS18 MRI-histologic postmortem correlationshowed a 22 reduction of whole thalamic vol-umes and a similar reduction in mean neuronaldensity in MS patients compared with controls8

There are several potential explanations forpreferential loss of thalamic volume comparedwith whole brain volume ranging from biologicto technical factors The thalamus has rich recip-rocal connectivity with much of the brain andmight be particularly susceptible to hypometabo-lism and wallerian degeneration due to demyeli-nation and axonal loss in cerebral white matterThis is supported by an observation that hypome-tabolism in the thalamus measured by PETshowed a significant association with white mat-ter lesion burden in patients with MS27 In addi-tion reduction of N-acetylaspartate in thethalamus correlated with reduction ofN-acetylaspartate in the normal-appearing fron-tal white matter51 Consistent with these observa-tions we found a moderate relationship betweenthalamic atrophy and white matter lesion volumein the present study

Figure 3 Total thalamic fraction correlates withbrain parenchymal fraction in theMS group (n 79)

Table 4 Cognitive data in patients with MS vs controls

COWAT JLO CVLT-II-TR CVLT-II-D BVMT-R-TR BVMT-R-D PASAT SDMT BDI-FS

MS(n 31)

342 101 223 69 507 132 108 38 22 91 91 37 351 14 482 19 38 43

NC(n 16)

426 151 268 25 562 90 123 22 281 49 116 11 417 112 634 90 063 12

p value 0184 0102 0201 0281 0023 0083 0344 0015 0007

Effectsize

065 086 048 049 084 089 052 101 098

Calculated effect sizes49 between the cognitive cohort of multiple sclerosis patients (MS) and normal controls (NC) weremedium to large for most cognitive testsCOWAT Controlled Oral Word Association Test JLO Judgment of Line Orientation Test CVLT-II-TR California VerbalLearning Test (Total Recall) CVLT-II-D California Verbal Learning Test (Delayed Recall) BVMT-R-TR Brief VisuospatialMemory Test (Total Recall) BVMT-R-D Brief Visuospatial Memory Test (Delayed Recall) PASAT Paced Auditory SerialAddition Test mean of 30-second interval and 20-second interval trials SDMT Symbol Digit Modalities Test BDI-FS

Beck Depression InventoryndashFast Screen for Medical Patients

Pearson r 0718 p

00001 MS multiplesclerosis


The thalamus might also suffer direct damagesuch as iron deposition or MS plaque formationOne study showed that thalamic T2 hypointen-sity a proposed marker of iron deposition pre-dicts subsequent whole brain atrophy early in thedisease course in patients with relapsingndashremit-ting MS26 Thus one putative mechanism for tha-lamic damage is free radicals and lipidperoxidation related to high levels of iron Demy-elinating plaques may be found in the deep graymatter including the thalamus95253 These lesionsmay be focal and discrete or may affect up to onethird of the thalamus Demyelinating lesions inthe gray matter as opposed to white matter arethought to be relatively devoid of lymphocytic in-

flammation but show prominent neuronalloss95354 making them potentially difficult to de-tect on conventional MRI scans5556

One needs also to consider the potential effectsof measurement error affecting our results Oursemiautomated measure of thalamic volumeshowed much lower reproducibility than oursemiautomated measure of whole brain volumeIt is likely that the relatively poor reproducibilityof the thalamic segmentation was related to diffi-culty identifying the borders of the thalamussuch as the delineation from the capsula internaand anterior and posterior edges This wouldprobably be even more problematic in the MSgroup presumably because of disease-relatedchanges in the thalamus and adjacent tissues asreflected in a higher intrarater COV than in thenormal control group (leading to a segmentationbias) However the differences in thalamic vol-ume in MS vs controls exceeded the variabilityand the effect sizes were larger than for BPFThus our method likely detected truly increasedsensitivity of the thalamic vs whole brain atrophymeasure despite the technical limitations Futurestudies using automated segmentation of the thal-amus and other individual gray matter structuresare warranted to confirm and extend ourfindings

Our findings agree with previous work851

showing that thalamic volume is significantlyinversely correlated with third ventricularwidth in MS patients Previous studies indicate

Figure 4 Scatter plot of thalamic fraction andneuropsychological test score in 31patients with MS

Table 5 Correlation between MRI and cognitive variables in MS patients

COWAT JLO CVLT-II-TR CVLT-II-D BVMT-R-TR BVMT-R-D PASAT SDMT

Thalamic fraction 0506 0652 0625 0701 0723 0724 0714 0658

p 0002 p 0001 p 00001 p 00001 p 00001 p 00001 p 0001 p 00001

T1 lesion volume 0333 0411 0485 0452 0573 0610 0498 0444

p 0002 p 0011 p 0003 p 0005 p 00001 p 00001 p 0002 p 0007

FLAIR lesionvolume

0373 0515 0578 0537 0666 0706 0501 0524

p 0019 p 0002 p 00001 p 0001 p 00001 p 00001 p 0002 p 0001

BPF 0394 0584 0527 0542 0662 0644 0585 0570

p 0014 p 00001 p 0001 p 0001 p 00001 p 00001 p 00001 p 0001

Third ventricularwidth

0354 0586 0609 0507 0612 0629 0590 0443

p 0025 p 00001 p 00001 p 0002 p 00001 p 00001 p 00001 p 0007

Pearson correlation coefficients are shown Total thalamic fraction showed the highest correlations with impairment on testsof processing speed attention and special learning and memory Central and global cerebral atrophy were also strong predic-tors of cognitive impairment but were inferior to thalamic fractionMS multiple sclerosis COWAT Controlled Word Association Test JLO Judgment of Line Orientation TestCVLT-II-TR California Verbal Learning Test (Total Recall) CVLT-II-D California Verbal Learning Test (Delayed Recall)BVMT-R-TR Brief Visuospatial Memory Test (Total Recall) BVMT-R-D Brief Visuospatial Memory Test (Delayed Recall)PASAT Paced Auditory Serial Addition Test mean of 30-second interval and 20-second interval trials SDMT SymbolDigit Modalities Test FLAIR fluid-attenuated inversion recovery BPF brain parenchymal fraction

Thalamic atrophy wasassociated with impairedperformance on tests ofprocessing speedworkingmemory (Symbol DigitModalities Test [SDMT]black squares Pearson r

0658 p 00001) andvisuospatial memory (BriefVisuospatial Memory TestndashTotal Recall [BVMT] opensquares Pearson r

0724 p 00001) MS

multiple sclerosis


that third ventricular width is related to cogni-tive impairment in MS422 However in thepresent study while both variables showedmoderate to strong correlations with cognitiveperformance regression modeling suggestedthat thalamic volume was even more closely re-lated to cognitive impairment than was thirdventricular width

Our sample size is small and the relationshipsfound should be considered preliminary and re-quire replication In particular future studiesshould test a larger sample of patients with MSand normal controls with MRI and cognitive test-ing to evaluate more completely the relationshipbetween thalamic atrophy and cognitive dysfunc-tion Furthermore although theMS patients weremildly to moderately impaired compared withnormal subjects on tests of visual memory(BVMT-R) and processing speed (SDMT) thedifference in performance on the remaining cog-nitive tests was not significant (although itshowed a trend toward impairment) A larger pa-tient sample or patients with more severe cogni-tive impairment would have allowed forstatistical power to detect medium-size effects(d 048 to 101)49 such as seen in our sampleThalamic volume accounted for the main vari-ance in predicting neuropsychological test perfor-mance indicating a specific relationship betweencognitive function and thalamic atrophy How-ever partial correlation coefficients for otherMRI variables particularly third ventricularwidth and BPF also indicate moderate to strong

correlations with neuropsychological functionThe observed significant association between tha-lamic atrophy and cognition explains only 50 ofvariance in cognitive impairment One must con-sider the possibility that the degree of atrophymay not exceed an individual patientrsquos brain re-serve capacity Adaptive mechanisms such as re-cruitment of secondary neural pathways wouldlimit the association between structural damageand clinical status early in the disease course It islikely that integrity of other circuits and struc-tures not explored in this study also contributeto cognitive function among our patients Furtherstudies are warranted to compare the correlationwith cognitive function of gray matter atrophy inindividual structures such as the thalamus to dif-fuse occult damage in the white matter or graymatter with techniques such as magnetizationtransfer imaging57 diffusion tensor imaging58

magnetic resonance spectroscopy59 or other newtechniques52

There are several plausible reasons for the linkbetween thalamic atrophy and cognitive dysfunc-tion in MS The thalamus is an integral compo-nent of the limbic system and Papez circuit Itconsists of five functional classes of nuclei thatsubserve memory emotion attention arousalmood motivation and language modulation60

Vascular and inflammatory lesions that involvethe thalamic nuclei in various combinations pro-duce unique sensorimotor and behavioral syn-dromes61 A wide range of cortical or subcorticalbehavioral syndromes may be mimicked by iso-

Table 6 Regression modeling examining relationships between MRI and cognitive variables in MS patients

Cognitive test

Variables remainingin model afteradjusting forage and sex

Partial r forvariable remainingin final model Multiple R2 R2 change p Value

COWAT Thalamic fraction 047 026 020 0037

JLO Thalamic fraction 074 049 048 0001

Third ventricle width 077 058 009 0001

CVLT-II-TR Thalamic fraction 055 045 042 0001

CVLT-II-D Thalamic fraction 068 050 027 0001

BVMT-R TR Thalamic fraction 067 055 041 0001

BVMT-R-D Thalamic fraction 074 053 049 0001

PASAT Thalamic fraction 074 050 049 0001

SDMT Thalamic fraction 069 051 043 0001

All regression models controlled for age and sex by entering these covariates and holding them in the model in Block 1 Partialr values are after controlling for age and sex The five MRI variables were then entered in Block 2 using a forward stepwiseprocedure Results did not change when depression scores and disease duration were forced into the modelsMS multiple sclerosis COWAT Controlled Word Association Test JLO Judgment of Line Orientation TestCVLT-II-TR California Verbal Learning Test (Total Recall) CVLT-II-D California Verbal Learning Test (Delayed Recall)BVMT-R-TR Brief Visuospatial Memory Test (Total Recall) BVMT-R-D Brief Visuospatial Memory Test (Delayed Recall)PASAT Paced Auditory Serial Addition Test mean of 30-second interval and 20-second interval trials SDMT SymbolDigit Modalities Test


lated strokes in various thalamic vascular terri-tories62 Dysexecutive syndrome and poorcognitive planning among other phenomenaare common features of cognitive impairmentassociated with injury to the thalamus A PETstudy showed a correlation between thalamichypometabolism and cognitive impairment inpatients with MS27

We report mild correlations between thalamicvolume and neurologic disease severity scoresThe EDSS is biased heavily toward motor perfor-mance whereas relatively little weight is given tosensory impairment or cognitive disability A re-cent study showed no correlation between tha-lamic magnetization transfer ratio and physicaldisability in MS patients30 Our previous workalso failed to show a relationship between tha-lamic damage (as assessed by diffusion imaging)and EDSS score or disease duration28 Anatomi-cally the thalamus is not involved in generatingor sustaining motor function Its role in motorcontrol is best described as functional modulatorIt is not surprising therefore that thalamic in-volvement in MS is only weakly related to physi-cal disability

Our study furthers our understanding ofmechanisms of MS-related cognitive dysfunctionand suggests that thalamic atrophy is a clinicallyrelevant biomarker of the neurodegenerative dis-ease process in MS These findings should con-tinue to fuel the growing interest in uncoveringthe mechanisms behind gray matter involvementin MS9

ACKNOWLEDGMENTThe authors thank Ms Sophie Tamm for assistance with manuscriptpreparation and Dr Ashish Arora and Dr Venkata Dandamudi fortechnical assistance The authors are also grateful to Gary CutterPhD and Diane L Cookfair PhD for statistical consultation

Received December 27 2006 Accepted in final form April16 2007

REFERENCES1 Rao SM Leo GJ Bernardin L Cognitive dysfunction

in multiple sclerosis frequency patterns and predic-tion Neurology 199141685ndash691

2 Deloire MS Salort E Bonnet M et al Cognitive im-pairment as marker of diffuse brain abnormalities inearly relapsing remitting multiple sclerosis J NeurolNeurosurg Psych 200576519ndash526

3 Benedict RHB Fischer JS Archibald CJ et al Mini-mal neuropsychological assessment of MS patients aconsensus approach Clin Neuropsychol 200216381ndash397

4 Benedict RH Carone DA Bakshi R Correlating brainatrophy with cognitive dysfunction mood distur-bances and personality disorder in multiple sclerosis JNeuroimaging 200414 (suppl 3)36Sndash45S

5 Bermel RA Bakshi R The measurement and clinicalrelevance of brain atrophy in multiple sclerosis LancetNeurol 20065158ndash170

6 Lin X Tench CR Evangelou N et al Measurement ofspinal cord atrophy in multiple sclerosis J Neuroimag-ing 200414 (suppl 3)20Sndash26S

7 Trapp BD Peterson J Ransohoff RM et al Axonaltransection in the lesions of multiple sclerosis N EnglJ Med 1998338278ndash285

8 Cifelli A Arridge M Jezzard P et al Thalamic neuro-degeneration in multiple sclerosis Ann Neurol 200252650ndash653

9 Pirko I Lucchinetti CF Sriram S Bakshi R Gray mat-ter involvement in multiple sclerosis Neurology 200768634ndash642

10 Dalton CM Chard DT Davies GR et al Early devel-opment of multiple sclerosis is associated with progres-sive grey matter atrophy in patients presenting withclinically isolated syndromes Brain 20041271101ndash1107

11 Sanfilipo MP Benedict RH Sharma J et al The rela-tionship between whole brain volume and disability inmultiple sclerosis a comparison of normalized gray vswhite matter with misclassification correction Neuro-image 2005261068ndash1077

12 Sanfilipo MP Benedict RHB Weinstock-Guttman BBakshi R Gray and white matter brain atrophy andneuropsychological impairment in multiple sclerosisNeurology 200666685ndash692

13 Ge Y Grossman RI Udupa JK et al Brain atrophy inrelapsing-remitting multiple sclerosis fractional volu-metric analysis of gray matter and white matter Radi-ology 2001220606ndash610

14 De Stefano NMatthews PM Filippi M et al Evidenceof early cortical atrophy inMS relevance to white mat-ter changes and disability Neurology 2003601157ndash1162

15 Sailer M Fischl B Salat D et al Focal thinning of thecerebral cortex in multiple sclerosis Brain 20031261734ndash1744

16 Bermel RA Innus MD Tjoa CW Bakshi R Selectivecaudate atrophy in multiple sclerosis a 3DMRI parcel-lation study Neuroreport 200314335ndash339

17 Chen JT Narayanan S Collins DL et al Relatingneocortical pathology to disability progression inmultiple sclerosis using MRI Neuroimage 2004231168ndash1175

18 Pagani E Rocca MA Gallo A et al Regional brainatrophy evolves differently in patients with multiplesclerosis according to clinical phenotype AJNRAm J Neuroradiol 200526341ndash346

19 Bermel RA Bakshi R Tjoa C et al Bicaudate ratio asa magnetic resonance imaging marker of brain atrophyin multiple sclerosis Arch Neurol 200259275ndash280

20 Brass SD Benedict RHB Weinstock-Guttman B et alCognitive impairment is associated with subcorticalMRI gray matter T2 hypointensity in multiple sclero-sis Mult Scler 200612437ndash444

21 Christodoulou C Krupp LB Liang Z Cognitive per-formance and MR markers of cerebral injury in cogni-tively impaired MS patients Neurology 2003601793ndash1798

22 Benedict RH Weinstock-Guttman B Fishman I et alPrediction of neuropsychological impairment in multi-


ple sclerosis comparison of conventional magnetic res-onance imaging measures of atrophy and lesionburden Arch Neurol 200461226ndash230

23 Stein TMoritz C QuigleyM et al Functional connec-tivity in the thalamus and hippocampus studied withfunctional MR imaging AJNR Am J Neuroradiol2000211397ndash1401

24 Aggleton JP Brown MW Episodic Memory amnesiaand the hippocampal-anterior thalamic axis BehavBrain Sci 199922425ndash489

25 Bakshi R Benedict RHB Bermel RA et al T2 hypoin-tensity in the deep gray matter of patients with multiplesclerosis a quantitative magnetic resonance imagingstudy Arch Neurol 20025962ndash68

26 Bermel RA Puli SR Rudick RA et al Prediction oflongitudinal brain atrophy in multiple sclerosis by graymatter magnetic resonance imaging T2 hypointensityArch Neurol 2005621371ndash1376

27 Blinkenberg M Rune K Jensen CV et al Cortical ce-rebral metabolism correlates with MRI lesion load andcognitive dysfunction in MS Neurology 200054558ndash564

28 Fabiano AJ Sharma J Weinstock-Guttman B et alThalamic involvement in multiple sclerosis adiffusion-weighted magnetic resonance imaging studyJ Neuroimaging 200313307ndash314

29 Ranjeva JP Audoin B Au Duong MV et al Local tis-sue damage assessed with statistical mapping analysisof brain magnetization transfer ratio relationship withfunctional status of patients in the earliest stage of mul-tiple sclerosis AJNR Am J Neuroradiol 200526119ndash127

30 Davies GR Altmann DR Rashid W et al Emergenceof thalamic magnetization transfer ratio abnormalityin early relapsing-remitting multiple sclerosis MultScler 200511276ndash281

31 Taylor I Butzkueven H Litewka L et al Serial MRI inmultiple sclerosis a prospective pilot study of lesionload whole brain volume and thalamic atrophy J ClinNeurosci 200411153ndash158

32 Filippi M Rocca MA Colombo B et al Functionalmagnetic resonance imaging correlates of fatigue inmultiple sclerosis Neuroimage 200215559ndash566

33 McDonald WI Compston A Edan G et al Recom-mended diagnostic criteria for multiple sclerosisguidelines from the International Panel on the diag-nosis of multiple sclerosis Ann Neurol 200150121ndash127

34 Kurtzke JF Rating neurologic impairment in multiplesclerosis an expanded disability status scale (EDSS)Neurology 1983331444ndash1452

35 Fischer JS Rudick RA Cutter GR The Multiple Scle-rosis Functional Composite Measure (MSFC) an inte-grated approach to MS clinical outcome assessmentNational MS Society Clinical Outcomes AssessmentTask Force Mult Scler 19995244ndash250

36 Jacobs LD Wende KE Brownscheidle CM et al Aprofile of multiple sclerosis the New York State Multi-ple Sclerosis Consortium Mult Scler 19995369ndash376

37 Bermel RA Sharma J Tjoa CW et al A semiauto-mated measure of whole-brain atrophy in multiplesclerosis J Neurol Sci 200320857ndash65

38 Sharma J Sanfilipo MP Benedict RH et al Whole-brain atrophy in multiple sclerosis measured by auto-

mated versus semiautomated MR imagingsegmentation AJNR Am J Neuroradiol 200425985ndash996

39 Benedict RH Effects of using same vs alternate formmemory tests in short-interval repeated assessment inmultiple sclerosis J Int Neuropsychol Soc 200511727ndash736

40 Benedict RH Cookfair D Gavett R et al Validity ofthe minimal assessment of cognitive function in multi-ple sclerosis (MACFIMS) J Int Neuropsychol Soc200612549ndash558

41 Benton AL Sivan AB Hamsher KS et al Contribu-tions to neuropsychological assessment A clinicalmanual 2nd ed New York Oxford University Press1994

42 Delis DC Kramer JH Kaplan E et al California Ver-bal Learning Test manual 2nd ed Adult version SanAntonio TX Psychological Corp 2000

43 Benedict RH Brief Visuospatial Memory TestndashRe-vised Professional manual Odessa FL PsychologicalAssessment Resources Inc 1997

44 Gronwall DM Paced auditory serial addition task ameasure of recovery from concussion Percept MotSkills 197744367ndash373

45 Smith A Symbol Digit Modalities Test Manual LosAngeles Western Psychological Services 1982

46 Cutter GR Baier ML Rudick RA et al Developmentof a multiple sclerosis functional composite as a clinicaltrial outcome measure Brain 1999122871ndash882

47 Beck AT Steer RA Brown JK BDI-Fast Screen forMedical Patients Manual San Antonio TX Psycho-logical Corp 2000

48 Benedict RH Fishman I McClellan MM et al Valid-ity of the Beck Depression InventoryndashFast Screen inmultiple sclerosis Mult Scler 20039393ndash396

49 Cohen J Statistical power analysis for the behavioralsciences 2nd ed Hillsdale NJ Lawrence Erlbaum As-sociates 1988

50 Blalock H Social statistics New York McGraw-Hill1972

51 Wylezinska M Cifelli A Jezzard P et al Thalamicneurodegeneration in relapsingndashremitting multiplesclerosis Neurology 2003601949ndash1954

52 Geurts JJ Bo L Pouwels PJ et al Cortical lesions inmultiple sclerosis combined postmortem MR imagingand histopathology AJNR Am J Neuroradiol 200526572ndash577

53 Vercellino M Plano F Votta B et al Grey matter pa-thology in multiple sclerosis J Neuropathol Exp Neu-rol 2005641101ndash1107

54 Peterson JW Bo L Mork S et al Transected neuritesapoptotic neurons and reduced inflammation in corti-cal multiple sclerosis lesions Ann Neurol 200150389ndash400

55 Bakshi R Ariyaratana S Benedict RH Jacobs LFluid-attenuated inversion recovery magnetic reso-nance imaging detects cortical and juxtacorticalmultiple sclerosis lesions Arch Neurol 200158742ndash748

56 Geurts JJ Reuling IE Vrenken H et al MR spectro-scopic evidence for thalamic and hippocampal but notcortical damage in MS Magn Reson Med 200655478ndash483


57 Horsfield MA Magnetization transfer imaging in multi-ple sclerosis J Neuroimaging 200515 (suppl)58Sndash67S

58 Goldberg-Zimring D Mewes AUJ Maddah M Warf-ield SK Diffusion tensor magnetic resonance imagingin multiple sclerosis J Neuroimaging 200515 (suppl)68Sndash81S

59 Narayana PA Magnetic resonance spectroscopy in themonitoring of multiple sclerosis J Neuroimaging200515 (suppl)46Sndash57S

60 Schmahmann J Vascular syndromes of the thalamusStroke 2003342264ndash2278

61 Ghika-Schmid F Bogousslavsky J The acute behav-ioral syndrome of anterior thalamic infarction a pro-spective study of 12 cases Ann Neurol 200048220ndash227

62 Carrera E Bogousslavsky J The thalamus and behav-ior effects of anatomically distinct strokes Neurology2006661817ndash1823

Experience the 2007 Annual Meeting from theComfort of Your Home

Much of this yearrsquos top programming is now available to you through the AAN VirtualAnnual Meeting Products

Audio MP3 Filesmdashstarting at $10Conveniently listen to some of the best talks on todayrsquos most significant neurological topics

bull Webcasts-on-Demandmdashseveral program hours FREEInstant online access to the slides audio and video of more than 240 hours of programs andpresentations including five Plenary Sessions Enjoy several hours of free programming inaddition to other programs available for purchase ($399 members $199 Junior members$499 non-members)

bull 2007 Syllabi on CD-ROMmdashonly $199Complete syllabi for more than 160 education programs ($199 members $299 non-members)

Check out the Virtual Annual Meeting Programming at wwwaancomvirtualAM today





ServicesUpdated Information amp

httpwwwneurologyorgcontent69121213fullhtmlincluding high resolution figures can be found at

References

1httpwwwneurologyorgcontent69121213fullhtmlref-list-at This article cites 55 articles 24 of which you can access for free

Citations

icleshttpwwwneurologyorgcontent69121213fullhtmlotherartThis article has been cited by 46 HighWire-hosted articles

Subspecialty Collections

httpwwwneurologyorgcgicollectionmultiple_sclerosisMultiple sclerosis

httpwwwneurologyorgcgicollectionmriMRI

e_disorders_dementiahttpwwwneurologyorgcgicollectionassessment_of_cognitivAssessment of cognitive disordersdementia

_dementiahttpwwwneurologyorgcgicollectionall_cognitive_disordersAll Cognitive DisordersDementiafollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwneurologyorgmiscaboutxhtmlpermissionsor in its entirety can be found online atInformation about reproducing this article in parts (figurestables)

Reprints

httpwwwneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

Thalamic atrophy and cognition inmultiple sclerosis

MK Houtchens MDRHB Benedict PhDR Killiany PhDJ Sharma MDZ Jaisani MDB Singh MDB Weinstock-Guttman MD

CRG GuttmannMD

R Bakshi MD

ABSTRACT

Objectives Recent studies have indicated that brain atrophy is more closely associated with cog-nitive impairment in multiple sclerosis (MS) than are conventional MRI lesion measures Enlarge-ment of the third ventricle shows a particularly strong correlation with cognitive impairmentsuggesting clinical relevance of damage to surrounding structures such as the thalamus Previ-ous imaging and pathology studies have demonstrated thalamic involvement in MS In this studywe tested the hypothesis that thalamic volume is lower in MS than in normal subjects and thatthalamic atrophy in MS correlates with cognitive function

MethodsWe studied 79 patients with MS and 16 normal subjects A subgroup of 31 MS subjectsunderwent cognitive testing The thalamus was segmented in whole from three-dimensional MRIscans We also determined whole brain atrophy (brain parenchymal fraction) third ventricularwidth and whole brain T2-weighted (fluid-attenuated inversion recovery) hyperintense T1hypointense and gadolinium-enhanced lesion volumes

Results Normalized thalamic volume was 168 lower in the MS group (p 00001) vs controlsCognitive performance in all domains was moderately to strongly related to thalamic volume in theMS group (r 0506 to 0724 p 0005) and thalamic volume entered and remained in allregression models predicting cognitive performance Thalamic volume showed a weak relation-ship to physical disability score (r 0316 p 0005)

Conclusion These findings suggest that thalamic atrophy is a clinically relevant biomarker of theneurodegenerative disease process in multiple sclerosis Neurologyreg 2007691213ndash1223

GLOSSARYBDI-FS Beck Depression InventoryndashFast Screen for Medical Patients BPF brain parenchymal fraction BVMT-R-D

Brief Visuospatial Memory Test (Delayed Recall) BVMT-R-TR Brief Visuospatial Memory Test (Total Recall) COWAT

Controlled Oral Word Association Test CVLT-II California Verbal Learning Test second edition CVLT-II-D CaliforniaVerbal Learning Test (Delayed Recall) CVLT-II-TR California Verbal Learning Test (Total Recall) EDSS Expanded Dis-ability Status Scale FLAIR fluid-attenuated inversion recovery FOV field-of-view JLO Judgment of Line OrientationTest MS multiple sclerosis NC normal controls NSA number of signal averages COWAT Controlled Oral WordAssociation Test PASAT Paced Auditory Serial Addition Test SPGR spoiled gradient recall SDMT Symbol DigitModalities Test TE echo time TR repetition time

Cognitive dysfunction is a disabling manifestation of multiple sclerosis (MS) affectingapproximately 50 of patients1 MS-related cognitive impairment mainly affects atten-tion information processing speed and episodic memory23 The severity of these deficitsmay reflect the extent of lesions and the degree of tissue loss and disorganization outsidelesions4

MS is now recognized as a destructive disease in that CNS atrophy is common occursearly in the disease course and is typically progressive56 The underlying substrate ofatrophy includes axonal and neuronal loss78 Recent data indicate that brain atrophy

From the Departments of Neurology (MKH JS ZJ CRGG RB) and Radiology (RK CRGG RB) Brigham and WomenrsquosHospital Harvard Medical School Partners MS Center Boston MA and Jacobs Neurological Institute (RHBB BS BW-G) Universityat Buffalo NY

Supported in part by research grants from the NIH (NS42379-01mdashDr Bakshi) and National Multiple Sclerosis Society (RG 3574A1 and RG3705A1mdashDrs Bakshi and Guttmann) and a Clinical Investigator Training Program from HarvardMassachusetts Institutes of TechnologyHealth Sciences and Technology Beth Israel Deaconess Medical Center Pfizer and Merck amp Co (Dr Houtchens)

Disclosure The authors report no conflicts of interest

Address correspondence andreprint requests to Dr RohitBakshi Brigham and WomenrsquosHospital Harvard MedicalSchool 77 Avenue LouisPasteurndashHIM 730 Boston MA02115rbakshibwhharvardedu

Copyright copy 2007 by AAN Enterprises Inc 1213

















Age mean SD y 420 96 424 84 465 93

Women n () 60 (76) 23 (74) 12 (75)

Men n () 19 (24) 8 (26) 4 (25)

































0549 p 0001) BVMT-R-D (R2 0526 p













r Value 0624 0707 0718 0617 0088 0219 0316 0218

p Value 0001 00001 00001 00001 044 0053 0005 0070











MS(n 31)

342 101 223 69 507 132 108 38 22 91 91 37 351 14 482 19 38 43

NC(n 16)

426 151 268 25 562 90 123 22 281 49 116 11 417 112 634 90 063 12

p value 0184 0102 0201 0281 0023 0083 0344 0015 0007

Effectsize

065 086 048 049 084 089 052 101 098



Pearson r 0718 p












p 0002 p 0001 p 00001 p 00001 p 00001 p 00001 p 0001 p 00001

T1 lesion volume 0333 0411 0485 0452 0573 0610 0498 0444

p 0002 p 0011 p 0003 p 0005 p 00001 p 00001 p 0002 p 0007

FLAIR lesionvolume

0373 0515 0578 0537 0666 0706 0501 0524

p 0019 p 0002 p 00001 p 0001 p 00001 p 00001 p 0002 p 0001

BPF 0394 0584 0527 0542 0662 0644 0585 0570

p 0014 p 00001 p 0001 p 0001 p 00001 p 00001 p 00001 p 0001


0354 0586 0609 0507 0612 0629 0590 0443

p 0025 p 00001 p 00001 p 0002 p 00001 p 00001 p 00001 p 0007




0724 p 00001) MS

multiple sclerosis









Cognitive test


































































































References


Citations









Reprints


















Age mean SD y 420 96 424 84 465 93

Women n () 60 (76) 23 (74) 12 (75)

Men n () 19 (24) 8 (26) 4 (25)

































0549 p 0001) BVMT-R-D (R2 0526 p













r Value 0624 0707 0718 0617 0088 0219 0316 0218

p Value 0001 00001 00001 00001 044 0053 0005 0070











MS(n 31)

342 101 223 69 507 132 108 38 22 91 91 37 351 14 482 19 38 43

NC(n 16)

426 151 268 25 562 90 123 22 281 49 116 11 417 112 634 90 063 12

p value 0184 0102 0201 0281 0023 0083 0344 0015 0007

Effectsize

065 086 048 049 084 089 052 101 098



Pearson r 0718 p












p 0002 p 0001 p 00001 p 00001 p 00001 p 00001 p 0001 p 00001

T1 lesion volume 0333 0411 0485 0452 0573 0610 0498 0444

p 0002 p 0011 p 0003 p 0005 p 00001 p 00001 p 0002 p 0007

FLAIR lesionvolume

0373 0515 0578 0537 0666 0706 0501 0524

p 0019 p 0002 p 00001 p 0001 p 00001 p 00001 p 0002 p 0001

BPF 0394 0584 0527 0542 0662 0644 0585 0570

p 0014 p 00001 p 0001 p 0001 p 00001 p 00001 p 00001 p 0001


0354 0586 0609 0507 0612 0629 0590 0443

p 0025 p 00001 p 00001 p 0002 p 00001 p 00001 p 00001 p 0007




0724 p 00001) MS

multiple sclerosis









Cognitive test


































































































References


Citations









Reprints
































0549 p 0001) BVMT-R-D (R2 0526 p













r Value 0624 0707 0718 0617 0088 0219 0316 0218

p Value 0001 00001 00001 00001 044 0053 0005 0070











MS(n 31)

342 101 223 69 507 132 108 38 22 91 91 37 351 14 482 19 38 43

NC(n 16)

426 151 268 25 562 90 123 22 281 49 116 11 417 112 634 90 063 12

p value 0184 0102 0201 0281 0023 0083 0344 0015 0007

Effectsize

065 086 048 049 084 089 052 101 098



Pearson r 0718 p












p 0002 p 0001 p 00001 p 00001 p 00001 p 00001 p 0001 p 00001

T1 lesion volume 0333 0411 0485 0452 0573 0610 0498 0444

p 0002 p 0011 p 0003 p 0005 p 00001 p 00001 p 0002 p 0007

FLAIR lesionvolume

0373 0515 0578 0537 0666 0706 0501 0524

p 0019 p 0002 p 00001 p 0001 p 00001 p 00001 p 0002 p 0001

BPF 0394 0584 0527 0542 0662 0644 0585 0570

p 0014 p 00001 p 0001 p 0001 p 00001 p 00001 p 00001 p 0001


0354 0586 0609 0507 0612 0629 0590 0443

p 0025 p 00001 p 00001 p 0002 p 00001 p 00001 p 00001 p 0007




0724 p 00001) MS

multiple sclerosis









Cognitive test


































































































References


Citations









Reprints










MS(n 31)

342 101 223 69 507 132 108 38 22 91 91 37 351 14 482 19 38 43

NC(n 16)

426 151 268 25 562 90 123 22 281 49 116 11 417 112 634 90 063 12

p value 0184 0102 0201 0281 0023 0083 0344 0015 0007

Effectsize

065 086 048 049 084 089 052 101 098



Pearson r 0718 p












p 0002 p 0001 p 00001 p 00001 p 00001 p 00001 p 0001 p 00001

T1 lesion volume 0333 0411 0485 0452 0573 0610 0498 0444

p 0002 p 0011 p 0003 p 0005 p 00001 p 00001 p 0002 p 0007

FLAIR lesionvolume

0373 0515 0578 0537 0666 0706 0501 0524

p 0019 p 0002 p 00001 p 0001 p 00001 p 00001 p 0002 p 0001

BPF 0394 0584 0527 0542 0662 0644 0585 0570

p 0014 p 00001 p 0001 p 0001 p 00001 p 00001 p 00001 p 0001


0354 0586 0609 0507 0612 0629 0590 0443

p 0025 p 00001 p 00001 p 0002 p 00001 p 00001 p 00001 p 0007




0724 p 00001) MS

multiple sclerosis









Cognitive test


































































































References


Citations









Reprints











p 0002 p 0001 p 00001 p 00001 p 00001 p 00001 p 0001 p 00001

T1 lesion volume 0333 0411 0485 0452 0573 0610 0498 0444

p 0002 p 0011 p 0003 p 0005 p 00001 p 00001 p 0002 p 0007

FLAIR lesionvolume

0373 0515 0578 0537 0666 0706 0501 0524

p 0019 p 0002 p 00001 p 0001 p 00001 p 00001 p 0002 p 0001

BPF 0394 0584 0527 0542 0662 0644 0585 0570

p 0014 p 00001 p 0001 p 0001 p 00001 p 00001 p 00001 p 0001


0354 0586 0609 0507 0612 0629 0590 0443

p 0025 p 00001 p 00001 p 0002 p 00001 p 00001 p 00001 p 0007




0724 p 00001) MS

multiple sclerosis









Cognitive test


































































































References


Citations









Reprints









Cognitive test


































































































References


Citations









Reprints






















































































References


Citations









Reprints


Thalamic atrophy and cognition in multiple sclerosis

Documents

Transcript of Thalamic atrophy and cognition in multiple sclerosis