Detecting Cognitive Disorders in MS

Ralph H. B. Benedict, Ph.D., ABPP-CN

Associate Professor of Neurology and Psychiatry

SUNY Buffalo School of Medicine

Jacobs Neurological Institute

Detecting Cognitive Disorders in MS

Outline

Brief History Update: Screening for NP dysfunction Update: Consensus NP Test Battery Clinical Application: Integrating Cognitive

Assessment into Routine Care

MS: MRI T2

Grade 1 Grade 2 Grade 3

Adapted from Rao et al. (1991)Adapted from Rao et al. (1991)

Language SpatialSpatialAbilities

AttentionAttentionSpanSpan

ProcessingProcessingSpeedSpeed

MemoryMemory ProblemProblemSolvingSolving

3030

2525

2020

1515

1010

55

00

8-9%

12-19%

7-8%

22-25%

22-31%

13-19%

SET A

SET B

SET C

“marked enfeeblement of the memory …. conceptions are formed slowly..”

Charcot (1877)

NP Profiles in MS

30

55

4540

60

50

0

10

20

30

40

50

60

language

spatial

mem

ory

atten/wm

executive

premor IQ

Normal (30%)

% ile

60

45

71

60

50

0

10

20

30

40

50

60

language

spatial

mem

ory

atten/wm

executive

premor IQ

Mild Cognitive Disorder (55%)

% ile

2

45

33

19

4550

0

10

20

30

40

50

60

language

spatial

mem

ory

atten/wm

executive

premor IQ

Focal Defect (<5%)

% ile25

2 2 1 1

50

0

10

20

30

40

50

60

language

spatial

mem

ory

atten/wm

executive

premor IQ

Dementia (10-15%)

% ile

A line is drawn through the long axis of the third ventricle, parallel to the inter-hemispheric fissure. The width is measured by drawing a second line perpendicular to the first at its midpoint, and then recording its length.

Third Ventricle Width (TVW)

Brain tissue volume (contour – CSF)

Outer contour of brain surface

Brain parenchymal fractionBrain parenchymal fraction

BPF =BPF =

Bicaudate Ratio (BCR)Bicaudate Ratio (BCR)

•BCR – the minimum BCR – the minimum intercaudate intercaudate distance divided by distance divided by brain width along brain width along same linesame line

•Higher BCR = Higher BCR = atrophyatrophy

Bermel et al., Arch Neurol 2002;59:275-280Bermel et al., Arch Neurol 2002;59:275-280

Brain Lesion Burden or Atrophy: Which MRI Measure Best Predicts Cognitive Impairment in Multiple Sclerosis?Benedict et al. Archives of Neurology. 2004; in press

• 37 MS patients volunteers, 28 controls matched age, educ, gender

• Conventional MRI: T1 LV; FLAIR LV; TVW; BCR; BPF

• Neuropsychological Testing: PASAT; SDMT; CVLT; BVMT; JLO; WCST

• Controls: Premorbid IQ = North American Adult Reading Test (NAART) Depression = Center for Epidemiologic Studies (CES-D)

T1 LV F LV BCR TVW

T1 LV

F LV .89

BCR .56 .49

TVW .50 .47 .80

BPF -.42 -.26 -.68 -.79

Retained MRI variable

Partial r Mult R2

Change in R2 from Block 1

P

CVLT Learning TVW -.44 .26 .15 <.05

CVLT Delay TVW -.45 .44 .14 <.01

BVMT Delay TVW -.47 .42 .17 <.01

PASAT TVW -.57 .44 .26 <.001

SDMT TVW -.71 .57 .43 <.001

Benedict et al, J Neuroimaging. 2004, in press

• 31 MS patients from Benedict et al (AN; 2004) also had informant report NPI profiles

Depression Euphoria* Disinhibition*

T1 .55 .41 --

FLV .47 .35 --

BCR -- -- --

TVW -- .46 .46

BPF -- -.43 -.34

*only NPI scales correlated with cognitive function (eg PASAT r = -.53).

Screening for Cognitive Disorders?

Screening for MS Cognitive Impairment using a Self-Administered 15-Item QuestionnaireBenedict et al. Multiple Sclerosis. 2003; 9:95-101

Phase I: Item Selection/Reduction

Participants: 100 consecutive clinic patientsMeasure: 68 Items (atten, mem, other, pers/beh)

Scale: 0 (never) to 4 (severe)Procedure: Reduce to 15 Items via Rasch analysis

Correlate with Cognitive FunctionsQuestionnaire

• easily distracted• lose focus when listening• slowed problem solving• trouble describing programs

recently watched• forgetting appointments• forgetting what is read• instructions repeated• reminded to do tasks

• forgetting errands

• difficulty answering questions

• difficulty tracking two things at once

• missing the point of conversations

• difficulty controlling impulses

• laugh/cry with little cause

• talk excessively

MS Neuropsychological Screening Questionnaire (MSNQ)

0 not at all 3 very often & disruptive1 rarely & no problem 4 very often & greatly interferes with

life2 occasionally, seldom a problem

Screening for MS Cognitive Impairment Using a Self-Administered 15-Item Questionnaire

Participants: 50 pts (40 RR), random recruitmentMean age 42, education 15 years

Mean SD Alpha Corr

BDI

Corr

CVLT

Corr

PASAT

Patient Report 22.5 10.2 .93 .53 ns ns

Informant Report 18.4 11.1 .94 ns -.53 -.47

Benedict, et al. Multiple Sclerosis. 2003; 9:95-101

Bayesian Probabilities, Informant Report MSNQ

Sensitivity # TP / # patientsSpecificity # TN / # normalsPositive PP # TP / # positive testsNegative PP # TN / # negative tests

Imp Nrm

MSNQ +

10 1 11

MSNQ -

2 37 39

12 38 50

Sens = .83 Spec = .97PPP = .91 NPP = .95

Processing Speed and Memory Tests

Problems?

small sample, one center

no controls

no test-retest reliability

MSNQ Cross Validation - Methods

85 MS pts, 40 age and education matched normal controls

80% pts with RR course

62% informants were spouses, known pt mean 23 yrs

MSNQ administered to all subjects, 40 MS retest at 1 week

MACFIMS

Depression: CESD-10; BDI-FS

Centers: Jacobs Neurological InstituteBernard Gimbel MS CenterUCSFUCHSC

MSNQ Cross Validation - Reliability

Cronbach’s Alpha Self-Report .94

Cronbach’s Alpha, Informant Report .93

Test-Retest Reliability (r) Self-Report .90

Test-Retest Reliability (r) Informant-Report .93

MSNQ Cross Validation - Validity

Self Inf

COWAT -.17 -.33*

JLO -.44± -.36±

CVLT-II Learning -.37± -.45±

CVLT-II Delay -.42± -.50±

BVMT-R Learning -.46± -.57±

BVMT-R Delay -.45± -.55±

PASAT -.38± -.59±

SDMT -.45± -.58±

D-KEFS Sorting -.30* -.38±

Composite Z -.49± -.64±

CES-D-10 Depression .61± .37±

* p<.05± p<.01

MSNQ Cross Validation – Predictive Validity

SENS SPEC PPV NPV

Self Report .83 .60 .53 .87

Self Report (dep) .80 .68 .73 .75

Informant .87 .84 .74 .92

Informant (2003) .83 .97 .91 .95

Screening is positive, now what?

Minimal Neuropsychological Assessment of MS Patients: A Consensus ApproachBenedict et al. Clinical Neuropsychologist, 2002.

• Ralph Benedict, Ph.D. Co-Chair

• Jill Fischer, Ph.D. Co-Chair

• Cate Archibald, Ph.D.

• Peter Arnett, Ph.D.

• William Beatty, Ph.D.

• Julie Bobholz, Ph.D.

• Gordon Chelune, Ph.D.

• John Fisk, Ph.D.

• Dawn Langdon, Ph.D.

• Lauren Caruso, Ph.D.

• John DeLuca, Ph.D.

• Fred Foley, Ph.D.

• Nicholas LaRocca, Ph.D.

• Frederick Munschauer, M.D.

• Steve Rao, Ph.D.

• Lindsey Vowels, Ph.D.

• Amy Weinstein, Ph.D.

Process:

(1) Review Literature (MS and Psychometric Data) (2) Consensus: Purpose of Minimal Exam Cognitive Domains Psychometric Criteria (3) Rank Order Candidate Tests (4) Propose Directions for Research

Consensus: Psychometric Criteria

Standardized Manual, commercially available

Normative Data Published, large sample, age 20-55

Adequate Range No ceiling or floor effects.

Reliability Moderate/high test-retest reliability.

Criterion Validity Discriminates MS and controls, good sensitivity & specificity.

Alternate Forms Alternate forms are equivalent.

Practical Brief, minimal equipment, minimal confound of neurol signs.

Minimal Assessment of Cognitive Function in Multiple Sclerosis (MACFIMS)

Test Cognitive Domain Min

Paced Auditory Serial Addition Test Processing Speed/Working Memory

10

Symbol Digit Modalities Test Processing Speed/Working Memory

5

California Verbal Learning Test Learning and Memory 25

Brief Visuospatial Memory Test Learning and Memory 10

D-KEFS Sorting Test Executive Function 25

Judgement of Line Orientation Test Visual-Spatial Ability 10

Controlled Oral Word Association Test Language and Other Domains 5

Practice Effects, Alternate Forms

Benedict & Zgaljardic. (1998). J Clin Exp Neuropsychol Zgaljardic & Benedict. (2001). Appl Neuropsych, in press

Subjects: 32 healthy volunteers, mean age 63.3, 37% male

Design: Counterbalanced adm of 4 HVLT-R & BVMT-R formsTwo week Test/Retest IntervalRandom assignment to same- and alternate-form conditions

Brief Visuospatial Memory Test - Revised

4

5

6

7

8

9

10

11

12

T-1 T-2 T-3 T-4

Session 1Session 2Session 3Session 4

4

5

6

7

8

9

10

11

12

T-1 T-2 T-3 T-4

Session 1Session 2Session 3Session 4

MACFIMS Validation

373 MS pts (176 clinical referral) 59 age and education matched normal controls 73% pts with RR course

MACFIMS Timed Ambulation, 9-H Peg,Maximum Repetition Rate

CESD-10; BDI-FS Fatigue Severity Scale NEO – Five Factor Inventory Neuropsychiatric Inventory MSQOL-54

Test-Retest Reliability

Alternate Form Testing

r

CVLT-II Learning .77

CVLT-II Delayed Recall .87

CVLT-II Delayed Recognition .77

BVMT-R Learning .80

BVMT-R Delayed Recall .75

BVMT-R Delayed Recognition .60

Sample: MS = 39

Group Discrimination

TEST % impaired

all

% impaired exclude referrals

ANOVA MS v

Control p

Controlled Oral Word Association 17 13 <0.001

Judgement Of Line Orientation 41 35 <0.001

CVLT-II Learning 39 29 <0.001

CVLT-II Delayed Recall 40 27 <0.001

BVMT-R Learning 43 34 <0.001

BVMT-R Delayed Recall 50 39 <0.001

PASAT 43 28 <0.001

Symbol Digit Modalities Test 60 42 <0.001

D-KEFS Sorting Test 57 39 <0.001

Study Design

VocationalStatus

Retained Significant Predictors

FatigueAge,

Education, etc

DiseaseFeatures

PhysicalDisability

Cognitive Function

MoodDisorder

PersonalityBehavior Disorder

MSQOL-54Mental

Composite

MSQOL-54Physical

Composite

1st tier regression models, controlling for any significant demographics

2nd tier regression models, controlling for any significant demographics

Regression Analysis Results

MSQOL-54 PhysicalFatigue Severity Scale (R2=0.57)Center for Epidemiologic Studies Depression Scale (R2change=0.08)EDSS (R2change=0.02)Beck Depression Inventory – Fast Screen for Medical Patients (R2change=0.02)

MSQOL-54 MentalCenter for Epidemiologic Studies Depression Scale (R2=0.68)Fatigue Severity Scale (R2change=0.03)

Vocational StatusSymbol Digit Modalities Test (Wald=12.8)WCST Perseverations (Wald=7.8)NEO Personality Inventory, Informant Report Conscientiousness (Wald=6.7)Disease Duration (Wald=5.3)Judgment of Line Orientation Test (Wald=3.8)

Sample: MS = 120; Control = 44

Patient Care?

Neuropsychological Practice Guidelines

Routine ScreeningIn the Office

RoutineMinimal NP Exam

+MSNQ

MACFIMS

ComprehensiveNP Evaluation

Psychiatry OT, ST

-0.40 .25 0 .5 0 .8

-0 .75

0 .5

-4

-3

-2

-1

0

1

2

z score

Case HJ

0 0 0 0 0 00.3 0 .1 0 .2 0 .5 0 0.102468

1012

me an scoreraw score

Case HJ

Case HJ• age 23, EDSS 2.5, DD = 3.0 yr• CC is fatigue• BDI-FS borderline• TVW=3.0, BPF=0.87

Case BK• age 26, EDSS 4.0, DD = 7.0 yr• CC is situational, needs not met• BDI-FS elevated• TVW=6.4, BPF=0.79

-0.340.24

-1-1.5

-3.1

0.5

-4

-3

-2

-1

0

1

2

z score

Case BK

69

0

69

20.3 0 .1 0 .2 0 .5 0 0.10

2468

1012

me an scoreraw score

Case BK

Thank You

Jacobs Neurological InstituteLawrence Jacobs, MDFrederick Munschauer, MDBianca Weinstock Guttman, MD

Buffalo Neuroimaging Analysis CenterRohit Bakshi, MDRobert Zivadinov, MD

UB NeuropsychologyDavid Shucard, PhD Inna Fishman, MAJanet Shucard, PhD Elizabeth Wahlig, MADominic Carone, PhD Jane Braun, PhDDanielle McCabe, MA