The Clinical Neuropsychologist KD Psychology Press

2004, Vol. 18, No. 1, pp. 4 1 ^ 9 ^ T^i.F™nci.c™,p

The Pattern of Neuropsychological Deficits

in Vascular Cognitive Impairment-No
Dementia (Vascular CIND)
David L. Nyenhuis', Philip B. Gorelick', Emily J. Geenen^, Clifford A. Smith",
Eugenia Gencheva', Sally Ereels^, and Leyla deToledo-Morrell'*'^
'Department of Neurology and Rehabilitation, University of Illinois at Chicago, IL, USA,
Psychology Department, Wayne State University, USA, 'School of Public Health,
University of Illinois at Chicago, IL, USA, and Rush University Medical Center, Chicago, IL, USA,
•"Psychology Department and 'Department of Neurological Sciences

ABSTRACT

We examined the pattern of neuropsychological deficits in Vascular Cognitive Impairment-No Dementia

(Vascular CIND) by comparing the cognitive and behavioral performance of 41 post-stroke Vascular CIND
patients to that of 62 post-stroke patients with no cognitive impairment (NCI). Neuropsychological test
scores were grouped into seven cognitive and four behavioral domains, then converted to standardized,
weighted principle component scores (PCS) for each domain. Multivariate logistic regression models built
on cognitive domains found the immediate recall and psychomotor domains to best predict diagnostic group
membership. In a separate model limited to behavioral data, the depressed mood domain best predicted
group membership. The combination of immediate memory deficits, psychomotor slowness and depression
have also been found in Vascular Dementia (VaD), suggesting that the pattern of deficits in Vascular CIND
and VaD neuropsychological deficits are similar. This cognitive and behavioral pattern similarity supports
the hypothesis that Vascular CIND lies on a continuum between NCI and VaD.

Vascular Dementia (VaD) may be the second lead-
ing cause of dementia in the United States and
Western Europe (Gorelick, Roman, & Mangone,
1994). However, VaD is only the most severe
subgroup of cerebrovascular-related cognitive
impairment (Babikian, Wolfe, Linn, Knoefel, &
Albert, 1990; Rockwood, Howard, MacKnight, &
Darvesh, 1999), and it's exclusive use may focus
treatment on persons too impaired to benefit
(Hachinski, 1994; Rockwood et al., 2000).
Vascular Cognitive Impairment (VCI) is a more
inclusive construct than VaD. It is subdivided into
three categories: Vascular Cognitive Impairment,
No Dementia (Vascular CIND), VaD, and Mixed
VCI with other dementing illness, most often
Alzheimer's disease (Rockwood et al., 1999).
This study is the first that we are aware of to
examine the neuropsychological impairment pat-
tern of Vascular CIND.
Vascular CIND impairments vary widely and
can include deficits in language, praxis, orienta-
tion, memory, executive functioning and visuo-
spatial functioning. The critical factors that
separate Vascular CIND from VaD are whether
there are enough deficient cognitive domains (all
dementia nosologies require at least two impaired
domains) and whether the cognitive deficits are
linked to functional (activities of daily living)

Address correspondence to: David L. Nyenhuis, Ph.D., University of Illinois Center for Stroke Research, 1645 W.
Jackson Blvd, Suite 400, Chicago, IL 60612, USA. E-mail: nyenhuis@uic.edu
Accepted for publication: January 29, 2004.

1385-4046/04/1801-041$!6.00 © Taylor & Francis Ltd.

42 DAVID L. NYENHUIS ET AL.
impairment. A sufficient number of impaired
domains and a link between cognitive and func-
tional impairments lead to a VaD diagnosis.
The prevalence and incidence of Vascular
CIND depend on which VaD criteria are used.
VaD criteria that are more stringent will result in
fewer patients diagnosed with dementia and a
correspondingly larger number of patients diag-
nosed with Vascular CIND. The National Institute
of Neurological Disorders and Stroke and the
Association Internationale pour la Recherche et
l'Enseignement en Neurosciences (NINDS-
AIREN) require memory plus two other impaired
cognitive domains (Roman et al., 1993). The State
of California Alzheimer's Disease Diagnostic and
Treatment Centers (ADDTC) criteria for
Ischemic Vascular Dementia (IVD) are relatively
less stringent. ADDTC criteria do not require
memory impairment, only that the cognitive
impairment is not isolated to a single narrow
category of intellectual performance (Chui et al.,
1992). A comparison of NINDS-AIREN and
ADDTC dementia criteria in our sample of post-
stroke patients showed VaD to be diagnosed in
13.2% by ADDTC and 7.7% by NINDS-AIREN
criteria (Geenen, Nyenhuis, Gencheva, &
Gorelick, 2003). Thus, use of NINDS-AIREN
criteria may lead to more Vascular CIND cases
than when ADDTC criteria are used.
Few studies have examined neuropathology
associated with Vascular CIND. As with VaD,
focal and diffuse white matter lesions are thought
to play a key role (Desmond, 2002; Englund,
2002). Global atrophic changes, and vascular
lesion location, size and number, have been impli-
cated in VaD (Nyenhuis & Gorelick, 1998). These
same variables may be related to Vascular CIND.
Similarly, stroke risk factors, such as hyperten-
sion, diabetes, obesity and smoking (Gorelick,
1995, 1997) are likely to be associated with
Vascular CIND. Rockwood, Ebly, Hachinski,
and Hogan (1997) found that 57% of their Vas-
cular CIND group had a history of stroke, 47.6%
had hypertension, 36.3% had cardiac symptoms
and 26.8% had a history of heavy smoking.
Few longitudinal studies have been completed
with Vascular CIND patients. Wentzel et al.
(2001) found that 46% of Vascular CIND patients
went on to develop dementia during a 5-year
period. By contrast, Nyenhuis, Gorelick, Freels,
and Garron (2002) found no cognitive decline in
their sample of African American ischemic stroke
patients who were not demented at baseline and
followed for up to 7 years. Differences in the two
study samples such as in the type of vascular
pathology and differences in inclusion criteria
(e.g., Nyenhuis et al. included all non-demented
patients, such as those with no cognitive impair-
ment) may explain the differences in results.
Neuropsychological research in VaD has at-
tempted to define the pattern of cognitive impair-
ments in VaD and compare it with the pattern in
other dementia syndromes, most often Alzhei-
mer's disease (AD) (Benthem, Jones, & Hodges,
1997; Crossley, D'Arcy, & Rawson, 1997;
Fuld, 1984; Kontiola, Laaksonen, Sulkava, &
Erkinjuntii, 1990; Mendez & Ashla-Mendez,
1991; Starkstein, Sabe, & Vazquez, 1996;
Villardita, 1993). A guiding hypothesis in this
research is that, relative to AD, VaD presents with
more deficits in attention, concentration and
executive function and less delayed memory
impairment (Kertesz & Clydesdale, 1994;
McPherson & Cummings, 1996). Libon et al.
and others (Bogdanoff, Gitlin, & Norman, 1997;
Lamar, Podell, & Carew, 1997; Libon, Bogdanoff,
& Bonavita, 1997; Libon, Bogdanoff, & Cloud
1998; Libon, Mattson, & Glosser, 1996) have
completed a number of projects that compare
deficits in Ischemic Vascular Dementia (IVD)
patients to those of patients with AD or Parkin-
son's disease. They conclude that neuropsycholog-
ical assessment can successfully differentiate
AD from IVD but that IVD is similar to other
subcortical dementing illnesses in the pattern of
neuropsychological results. Looi and Sachdev
(1999), in their review, conclude that VaD patients
showed better performances in verbal long-term
memory and poorer performance in frontal execu-
tive functioning than AD patients.
The present study examines the neuropsycho-
logical pattern of Vascular CIND patients by
comparing them with a group of patients without
cognitive impairment. Patients from both groups
had a history of ischemic stroke 3-6 months
prior to neuropsychological examination. Our
hypothesis is that Vascular CIND patients will
show a similar pattern of neuropsychological
impairment to that of patients with VaD when
compared with the NCI group.
 NEUROPSYCHOLOGICAL DEHCITS IN VASCULAR CIND
METHODS
Subjects
Subjects were consecutive ischemic stroke patients who
presented to the Stroke and Critical Care Service at
Rush-Presbyterian-St. Luke's Medical Center. Inclu-
sion criteria were age 50 or over; sufficient English
language skills for psychological tests; availability of a
friend or relative knowledgeable about the subject's
recent and past medical history who has known the
subject at least 10 years and sees them at least three
times per week; high likelihood of being available for
follow-up; verbal and written consent; and ischemic
stroke within the last 6 months. All ischemic strokes
met criteria for diagnosis of cerebral infarction by the
National Institute of Neurological and Communicative
Disorders and Stroke (NINDS) Stroke Data Bank
(Foulkes, Wolf, Price, Mohr, & Hier, 1988) and have
at least one focal symptom and/or sign of stroke at the
time of stroke. Exclusion criteria included patients with
aphasia who score less than 50% on the Boston
Diagnostic Aphasia Examination Commands Subtest
or who otherwise could not complete psychological
testing because of a language disorder (two potential
subjects were excluded for this reason); chronic or
degenerative disease prior to stroke (e.g., Alzheimer's
disease) affecting the central nervous system; any
metabolic, traumatic, endocrine, infectious, prion-
related, nutritional or toxic disorder that could affect
the central nervous system; substance abuse disorders
as defined by DSM-IV; intracranial hemorrhage;
difficult to control epilepsy that could present with
cognitive impairment; or temporal lobe epilepsy.
Potential subjects were contacted by the study team
and informed of the purpose and methods of the study.
Those who consented to participate in the study were
scheduled for the baseline epidemiological, neurological,
neuropsychological and MRI examination. The Institu-
tional Review Board (IRB) at Rush-Presbyterian-St.
Luke's Medical Center approved all study procedures.
Study Iustrumeuts and Examination
Procedures
All study subjects completed formal neuropsycho-
logical assessment, administered by a trained neuro-
psychological technician under the supervision of
the study neuropsychologist. The examination took
approximately 90-120 min to administer and included
the following tests: Boston Diagnostic Aphasia Exami-
nation, Commands subtest (Goodglass & Kaplan,
1989); Controlled Learning and Enhanced Recall,
Immediate and Delayed (Buschke & Grober, 1986);
Self-Ordered Pointing Task (Shimamura & Jurica,
1994); Wechsler Memory Scale, Third Edition: "Anna
Thompson" paragraph and Digit Span subtests
(Wechsler, 1997); Wechsler Memory Scale, Form I
43
Mental Control (Wechsler, 1945); Symbol Digit (Oral
Version; Smith, 1991); portions of the Behavioral
Dyscontrol Scale (Grisby & Kaye, 1996); Mini-Mental
State Examination (MMSE; Folstein, Folstein, &
McHugh, 1975); Figural Recognition Test (Scherr
et al., 1988); CERAD Boston Naming Test and Verbal
Fluency (Morris et aL, 1989); Brief version of the
Ravens Progressive Matrices (Raven, 1995); Grooved
Pegboard Test (Russell & Starkey, 1993); and the
Chicago Multiscale Depression Inventory (Nyenhuis
et al., 1998). The subjects were also administered the
Blessed Information, Memory and Concentration Test
(Blessed IMC; Thai, Grundman, & Golden, 1986) by a
study physician as part of the neurological examination
and the patient's designated informant completed the
Frontal Systems Behavior Scale (Grace, Stout, &
Malloy, 1999).
While the study subject was completing the neuro-
psychological examination, the informant, usually the
patient's spouse or child, completed an epidemiologic
interview. A brief, 12-question background information
and mental status test was completed and passed by the
informant prior to the interview. The epidemiologic
interview consisted of pretested questionnaires about
background, demographic, medical history and risk
factor information. The information contained in the
questionnaires was adapted primarily from cardiovas-
cular disease risk factor questionnaires used previously
by the study team (Gorelick, Brody, & Cohen, 1993)
and from the East Boston Senior Health Project (Evans
et al., 1989). The epidemiologic interviewer, under the
supervision of the study neurologist, also completed an
unstructured neurologic interview to determine the
patient's history of, number of, timing, symptoms and
recovery from stroke; a history of memory loss and
other symptoms of dementia; and a structured neuro-
logic interview to determine the history of dementia
onset and symptoms of cognitive dysfunction.
Diagnostic Procedures
The study neuropsychologist classified cognitive tests
into the following domains based on test content: basic
attention, working memory/executive function, lan-
guage, visuospatial skills, psychomotor skills, immediate
memory, delayed memory and recognition memory.
Behavioral tests were classified into tests of depression,
apathy, dysexecutive function, and disinhibition. Table 1
lists the neuropsychological tests by domain. To confirm
the relationship of intra-domain tests, pairwise correla-
tional analyses were completed with all cognitive tests,
using the entire subject sample. With the exception of
the delayed recognition memory domain (r=—.113,
p> .10) all tests correlated significantly (p< .01) with
each other test in its assigned domain. The average
intra-domain correlation was .443 {SD = .092) while the
average inter-domain correlation was .364 (5D = .148).
The intra-domain correlations for each domain were as
44 DAVID L. NYENHUIS ET AL.

Table 1. Neuropsychological Tests Listed by Domain. tion Memory domain score was removed from subse-
quent analyses.
Basic attention Cognitive impairment was rated in the following
Digits Forward domains for each subject: basic attention, working
WMS Mental Control memory, language, spatial skills, psychomotor skills,
Working memory orientation and memory. This rating was a clinical
Self-Ordered Pointing Test Total Score decision made by an experienced, board-certified
Digit Span Backward neuropsychologist (ABPP/ABCN) and based on nor-
Behavioral Dyscontrol Scale mative data and subject demographic characteristics.
When normative data were applicable, test performance
Language was judged as impaired if it was below the 5th
BDAE Commands percentile for persons of similar age and educational
Boston Naming Test background. Occasionally, a subject differed signifi-
Animal Naming cantly from a test's standardization sample (e.g., there
Spatial was a large deviation between the subject's reported
Figural Recognition Test education and that of the normative sample or English
Ravens Matrices [while fluently spoken] was a second language for the
subject). When this occurred, a correction of up to one-
Psychomotor half of a standard deviation (based on the test's
Grooved Pegboard (pegs/sec) standardization sample) was applied to the score when
Symbol Digit Oral making the impairment decision. The overall perfor-
mance rating in a domain was based on the preponder-
Memory: Immediate
ance of scores within the domain, taking into account
Controlled Learning and Enhanced Recall Total
the degree of impairment severity on each test and the
WMS-III Paragraph I
number of impaired tests in the domain.
Memory: Delayed Because the impairment decision was, in part, based
Controlled Learning and Enhanced Recall on clinical judgment and was not entirely based on preset
Delayed Recall cut scores, we examined the inter-rater reliability and
WMS-III Paragraph I Delayed Recall the validity of the impairment decision. To assess the
reliability, a second neuropsychologist (CS) was trained
Memory: Recognition
in study methods and independently rated 25 subjects
Controlled Learning and Enhanced Recall
chosen at random from the study sample. He there-
Delayed Recognition
fore made 175 impairment/non-impairment ratings
WMS-III Paragraph I Delayed Recognition
(25 subjects x 7 cognitive domains). There was indepen-
Depression dent agreement on 166 of 175 ratings, yielding a Kappa
Chicago Multiscale Depression Inventory - Mood score of .79, suggestive of "substantial agreement"
(Landis & Koch, 1977) between raters and reliable
Apathy
placement of subjects into impairment categories.
Frontal Systems Behavior Scale
To examine the validity of the method, we compared
Dysexecutive the Vascular CIND group's (a group largely defined by
Frontal Systems Behavior Scale the impairment rating decision) performance to the NCI
group on two tests that were not considered when
Disinhibition making impairment decisions: the MMSE and the
Frontal Systems Behavior Scale Blessed IMC Test. The Vascular CIND group was more
Note. WMS = Wechsler Memory Scale. impaired than the NCI group on both tests, as well as on
BDAE = Boston Diagnostic Aphasia Exami- two measures of functional ability (Table 2).
nation. Diagnoses and Clinical Dementia Rating (CDR) were
completed at a diagnostic consensus case conference,
attended by the neurologist, neuropsychologist and the
follows: Basic attention: .405; Working memory: .388; epidemiologic interviewer. Based on cognitive, func-
Language: .401; Visuospatial: .403; Psychomotor: .653: tional (activities of daily living scale scores) and other
Itnmediate Memory: .497; Delayed Recall Memory: clinical data, patients were diagnosed with vascular de-
.546. These results demonstrate the inter-relationship of mentia (VaD), vascular cognitive impairment-no demen-
all cognitive tests and a higher degree of relationship tia (Vascular CIND), or no cognitive impairment (NCI).
within than across cognitive domains. Because of the VaD was diagnosed according to NINDS-AIREN Criteria
lack of intra-domain correlation, the Delayed Recogni- (Roman et al, 1993). Vascular CIND was diagnosed
 NEUROPSYCHOLOGICAL DEHCrrS IN VASCULAR CIND 45

Table 2. Demographic and Summary Cognitive and above (Table 1). To allow for inter-domain comparison, a
Functional Scores by Diagnosis. principal component score was calculated for each
domain, based on all subjects' test scores. All domain
Vascular NCI scores were standardized to have a mean of zero and a
CIND in = 62) standard deviation of one. Estimated effects of domain
scores are then comparable as the effect of one standard
deviation increase. For domains with a single variable,
Male (%) 54 52 the score was simply that variable standardized. For
Age (years) 65.6 (9.7) 63.8 (8.4) domains with two variables, both variables were first
Years of education*** 12.71 (3.1) 15.1 (3.7) standardized and then weighted equally to form a mean
CDR*** 0.54 (0.52) 0.10 (0.22) value with a mean of one and a standard deviation of zero.
MMSE*** 26.5 (2.3) 28.5 (1.3) For domains with more than two variables, the first
Blessed** 3.3 (2.7) 1.7 (1.9) principle component score provided optimal weights
Barthel ADL* 95.2 (15.6) 99.8 (1.3) to produce a score which captured the most variation
Total IADL* 12.3 (3.3) 13.6(1.1) possible.
r-tests and Chi-square analyses were completed to
Note. ADL = Activities of Daily Living. compare groups on demographic variables. Correla-
CDR = Clinical Dementia Rating. tional analyses were used to test a priori assumptions
IADL = Instrumental Activities of Daily Living. about intra-domain tests. Stepwise logistic regression
MMSE = Mini-Mental State Examination. analyses were completed to test models of neuropsy-
*/7<.05;**/7<.01:***p<.001. chological patterns.

according to the method of Rockwood and his colleagues

in the Canadian Study of Health and Aging (Ebly, Hogan,
& Parhad, 1995), which requires that patients not meet
dementia criteria, but do present with cognitive impair-
ment in at least one domain. NCI was diagnosed for
patients who were not impaired in any cognitive domain.
Statistical Analyses
Neuropsychological tests were grouped a priori into
cognitive domains, based on test content as is explained
RESULTS
The Vascular CIND and NCI groups did not differ
in age or gender distribution (Table 2). The NCI
group completed more years of education than the
Vascular CIND group. As expected, the Vascular
CIND group showed greater cognitive and
functional deficits than the NCI group, as

Table 3. Principle Component Scores by Domain and Diagnosis.

Vascular CIND (n = 41) NCI (n = 62) Difference

Cognitive Domains
Basic attention -0.26 (1.42) 0.17 (0.53) 0.43
Working memory -0.36 (1.20) 0.24 (0.77) 0.60
Language -0.53 (1.16) 0.35 (0.70) 0.88
Visuospatial -0.42 (1.38) 0.28 (0.47) 0.70
Psychomotor" -0.58(1.02) 0.40 (0.77) 0.98
Immediate memory -0.66 (0.86) 0.44 (0.83) 1.10
Delayed recall memory -0.66 (0.89) 0.43 (0.82) 1.09
Behavioral Domains
Depression 0.25 (1.11) 0.16 (0.89) 0.41
Apathy'' 0.02 (0.98) 0.02 (1.02) 0.04
Executive*" 0.07 (1.03) 0.05 (0.97) 0.12
Disinhibition** 0.16 (1.27) 0.11 (0.77) 0.27
Note, "n = 35 for Vascular CIND, n = 51 for NCI.
"n = 40 for Vascular CIND, n = 61 for NCI.
CIND = cognitive impairment-no dementia.
NCI = no cognitive impairment.
46 DAVID L. NYENHUIS ET AL.

measured by more impaired CDR, MMSE, sion analyses were completed to examine the
Blessed, Barthel and IADL scores. ability of each individual principal component
Principal component scores are listed by score to predict group membership, controlling
domain in Table 3. Visual inspection shows the for education (Table 4). All cognitive domains
greatest cognitive group differences to be in successfully predicted group membership with
immediate memory, delayed recall memory and the exceptions of basic attention and working
psychomotor speed, while the greatest behavioral memory. Of the behavioral variables, depressed
difference is in depressed mood. Logistic regres- mood showed a trend (p = .056) toward predicting
group membership.
Two step-wise multiple logistic regression mod-
Table 4. Effects of Principle Components Predicting els were tested to see which variables indepen-
Vascular CIND (n = 41) Versus NCI (n = 62) dently explained the greatest group membership
Adjusted for Years of Education.
variance. In the first, cognitive domain variables
OR Confidence p value were eligible for inclusion after years of education
interval was forced into the equation. Two domains,
immediate memory and psychomotor speed, were
Cognitive Domains
selected to the model (Table 5). A second model
Basic attention 0.54 (0.22-2.13) .208
Working memory 0.52 (0.35-1.03) .062 tested only the behavioral domains. After years of
Language 0.38 (0.21-0.67) .001 education was forced into the equation, only the
Spatial 0.37 (0.17-0.78) .009 depressed mood domain was selected.
Psychomotor^ 0.32 (0.17-0.61) .001
Immediate memory 0.24 (0.13-0.47) .000
Delayed recall memory 0.23 (0.12-0.45) .000 DISCUSSION
Behavioral Domains
Depressed mood 1.55 (0.98-2.43) .056 In this study. Vascular CIND patients were most
Apathy'' 0.85 (0.55-1.31) .458 impaired on tests of immediate memory and
Disinhihition"' 1.20 (0.77-1.85) .422 psychomotor speed when compared to post-stroke
Executive*" 0.99 (0.65-1.53) .970
patients without cognitive impairment. The two
Note. "^ = 35 for Vascular CIND, n = 51 for NCI. groups also differed on tests of language function-
"n = 40 for Vascular CIND, n = 61 for NCI. ing, spatial functioning and delayed memory, but
CIND = cognitive impairment-no dementia. these domains did not add additional explained
NCI = no cognitive impairment.
variance in our multivariate model. Tests of basic
attention and working memory did not differenti-
Table 5. Multiple Logistic Regression Models ate the two diagnostic groups. A separate model,
Predicting Vascular CIND Versus NCI. limited to behavioral symptoms, found depressed
mood, as measured by the Mood subscale of the
OR Confidence p value CMDI, to successfully differentiate the two
interval groups, with Vascular CIND subjects reporting
Cognitive Domains^ more depressed mood than NCI subjects.
Years of education 0.946 (0.80-1.12) .509 Decreased immediate recall memory, psycho-
Psychomotor 0.367 (0.19-0.72) .004 motor slowing and depressed mood have also
Immediate memory 0.284 (0.14-0.59) .001 been noted in patients with VaD (e.g., Looi &
Behavioral Domains"' Sachdev, 1999). In this way, the present pattern of
Years of education 0.775 (0.66-0.91) .002 results in our non-demented post-stroke patients
Depression 1.618 (1.02-2.56) .040 is similar to what has been seen in cerebrovascular
Note. ''n = 35 for Vascular CIND, n = 51 for NCI.
patients with more severe cognitive impairment.
"n = 40 for Vascular CIND, n = 61 for NCI. The lack of predictive value of working memory
CIND = cognitive impairment-no dementia. tests in our study was unexpected, as tests of
NCI = no cognitive impairment. working memory and executive function have
 NEUROPSYCHOLOGICAL DEHCITS IN VASCULAR CIND
been shown to be sensitive to VaD. This may be
due to the specific tests of working memory that
we chose to include in the study. Perhaps other
tests of executive function, such as the Wisconsin
Card Sorting Test, the Stroop, or the Trailmaking
Test would have been more sensitive to deficits in
this domain. Previous research has shown that
some Vascular CIND patients develop dementia
(Wentzel et al., 2001). These data, combined with
the present findings, suggest that Vascular CIND
may be on a continuum between normal cognition
and VaD in post-ischemic stroke patients. Vascu-
lar CIND patients may therefore be well suited for
treatment studies designed to arrest the progres-
sion of cognitive and functional decline in
patients with cerebrovascular disease.
Limitations of this study include the fact that
the same neuropsychological tests used to classify
patients into the Vascular CIND and NCI groups
were also used to examine the neuropsychological
deficit pattern. However, the classification proce-
dure was independent of the pattern analysis. It
did not matter which cognitive domain(s) was
impaired for diagnostic classification purposes.
Therefore, classification results did not bias pat-
tern analysis. Secondly, our patient sample con-
sisted entirely of ischemic stroke patients. It may
be that other vascular patient groups, such as
patients with extensive subcortical leukoencepha-
lopathy in the absence of stroke, may present with
a different deficit pattern. Indeed, one of the
challenges of vascular cognitive impairment is
the heterogeneity of patients' presenting clinical
pictures. Thirdly, the study classified cognitive
tests a priori into separate cognitive domains. The
bivariate correlational analyses supported the
classification. While a more comprehensive (and
satisfying) method would be to complete explora-
tory or confirmatory factor analysis, the study
lacked a sufficient number of subjects to complete
such analyses.
The present study defined dementia using the
NINDS-AIREN criteria. This affected who was
diagnosed with Vascular CIND. It is likely that a
greater percentage of patients would have been
diagnosed with VaD using ADDTC criteria,
resulting in fewer Vascular CIND patients. It is
not known how this might affect the neuropsy-
chological deficit pattern. Future Vascular CIND
47
research should focus on a comparison of diag-
nostic criteria with an effort to reach consensus on
this ongoing diagnostic problem.
We will continue to follow this cohort of post-
ischemic stroke patients to examine the rate of
cognitive and functional decline, risk factors for
decline, and the rate of "turnover" from Vascular
CIND to VaD. We will also examine whether the
pattern of neuropsychological impairment will
remain constant in patients who progress from
Vascular CIND to VaD.
ACKNOWLEDGMENT
This work was supported by NIA ROl AG17934 (PI:
PBG).
REFERENCES
Babikian, V, Wolfe, N., Linn, R., Knoefel, J., & Albert,
M. (1990). Cognitive changes in patients with
multiple cerebral infarcts. Stroke, 21, 1013-1018.
Benthem, PW., Jones, S., & Hodges, J.R. (1997). A
comparison of semantic memory in vascular
dementia and dementia of the Alzheimer type.
International Journal of Geriatric Psychiatry, 12,
575-580.
Bogdanoff, B., Gitlin, H., & Norman, L. (1997). Neuro-
psychological comparisons between subjects with
ischemic vascular dementia, Parkinson's disease, and
Alzheimer's disease. Neurology, 48, A231-A232.
Buschke, H., & Grober, E. (1986). Genuine memory
deficits in age-associated memory impairment.
Developmental Neuropsychology, 2, 287-307.
Chui, H., Victoroff, J., Margolin, D., Jagust, W.,
Shankle, R., & Katzman, R. (1992). Criteria for
the diagnosis of ischemic vascular dementia pro-
posed by the State of Califomia Alzheimer's Dis-
ease Diagnostic and Treatment Centers (ADDTC).
Neurology, 42, 473^80.
Crossley, M., D'Arcy, C , & Rawson, N.S.B. (1997).
Letter and category fiuency in community-dwelling
Canadian seniors: A comparison of normal partici-
pants to those with dementia of the Alzheimer or
vascular type. Journal of Clinical and Experimental
Neuropsychology, 19, 52-62.
Desmond, D.W. (2002). Cognition and white matter
lesions. Cerebrovascular Diseases, 13, 53-57.
Ebly, E.M., Hogan, D.B., & Parhad, I.M. (1995).
Cognitive impairment in the nondemented elderly:
Results from the Canadian Study of Health and
Aging. Archives of Neurology, 52, 612-619.
48 DAVID L. NYENHUIS ET AL.
Englund, E. (2002). Neuropathology of white matter
lesions in vascular cognitive impairment. Cerebro-
vascular Diseases, 13, 11-15.
Evans, D.A., Funkenstein, H.H., Albert, M.S., Scherr,
P.A., Cook, N.R., Chown, M.J., Herbert, L.E.,
Hennekens, C.H., & Taylor, J.O. (1989). Prevalence
of Alzheimer's disease in a community population
higher than previously reported. Journal of the
American Medical Association, 262, 2251-2256.
Folstein, M.F., Folstein, S., & McHugh, P (1975).
Mini-Mental state: A practical method for grading
the cognitive state of patients for the clinician.
Journal of Psychiatric Research, 12, 189-198.
Foulkes, M.A., Wolf, RA., Price, T.R., Mohr, J.R, &
Hier, D.B. (1988). The stroke data bank: Design,
methods, and characteristics. Stoke, 19, 547-554.
Fuld, PA. (1984). Test profile of cholinergic dysfunc-
tion and Alzheimer-type dementia. Journal of
Clinical Neuropsychology, 6, 380-392.
Geenen, E.J., Nyenhuis, D.L., Gencheva, E., &
Gorelick, PB. (2003). Comparison of VaD fre-
quency using NINDS-AIREN and ADDTC diag-
nostic criteria. Journal of the International
Neuropsychological Society, 9, 166.
Goodglass, H., & Kaplan, E. (1989). The assessment
of aphasia and related disorders (2nd ed.).
Philadelphia: Lea and Fehiger.
Gorelick, PB. (1995). Stroke prevention. Archives of
Neurology, 52, 347-354.
Gorelick, PB. (1997). Status of risk factors for
dementia associated with stroke. Stroke, 28,
459-463.
Gorelick, PB., Brody, J., & Cohen, D. (1993). Risk
factors for dementia associated with multiple
cerebral infarcts: A case-control analysis in pre-
dominantly African-American hospital-based
patients. Archives of Neurology, 50, 714-720.
Gorelick, PB., Roman, G., & Mangone, C.A. (1994).
Vascular dementia. In PB. Gorelick & M.A.
Alter (Eds.), Handbook of neuroepidemiology
(pp. 197-214). New York: Marcel Dekker.
Grace, J., Stout, J., & Malloy, PF. (1999). Assessing
frontal lobe behavioral syndromes with the Frontal
Lobe Personality Scale. Assessment, 6, 269-284.
Grisby, J., & Kaye, K. (1996). Behavioral Dyscontrol
Scale: Manual. Denver, CO: Author.
Hachinski, V.C. (1994). Vascular dementia: A radical
redefinition. Dementia, 5, 130-132.
Kertesz, A., & Clydesdale, S. (1994). Neuropsychologi-
cal deficits in vascular dementia versus Alzheimer's
disease: Frontal lobe deficits prominent in vascular
dementia. Archives of Neurology, 51, 1226-1231.
Kontiola, P., Laaksonen, R., Sulkava, R., & Erkinjuntii,
T. (1990). Pattern of language impairment is
different in Alzheimer's disease and multi-infarct
dementia. Brain and Language, 38, 364-383.
Lamar, M., Podell, K., & Carew, T.G. (1997).
Perseverative behavior in Alzheimer's disease and
subcortical ischemic vascular dementia. Neuro-
psychology, 11, 1-12.
Landis, J.R., & Koch, G.G. (1977). The measurement
of observer agreement for categorical data. Bio-
metrics, 33, 159-174.
Libon, D.J., Bogdanoff, B., & Bonavita, J. (1997).
Dementia associated with periventricular and deep
white matter alterations: A subtype of subcortical
dementia. Archives of Clinical Neuropsychology,
12, 239-250.
Libon, D.J., Bogdanoif, B., & Cloud, B.S. (1998).
Declarative and procedural learning, quantitative
measures of the hippocampus, and subcortical white
matter alterations in Alzheimer's disease and
ischemic vascular dementia. Journal of Clinical
and Experimental Neuropsychology, 20, 30-41.
Libon, D.J., Mattson, R.E., & Glosser, G. (1996). A
nine word dementia version of the California
Verbal Learning Test. Clinical Neuropsychology,
10, 237-244.
Looi, J., & Sachdev, P (1999). Differentiation of
vascular dementia from AD on neuropsychological
tests. Neurology, 53, 670-^78.
McPherson, S.E., & Cummings, J.L. (1996). Neuro-
psychological aspects of vascular dementia. Brain
and Cognition, 31, 269-282.
Mendez, M.F., & Ashla-Mendez, M.A. (1991). Differ-
ences between multi-infarct dementia and Alzhei-
mer's disease on unstructured neuropsychological
tasks. Journal of Clinical and Experimental Neu-
wpsychology, 13, 923-932.
Morris, J.C., Heyman, A., Mohrs, R.C., Hughes, J.P,
van Belle, G., Fillenbaum, G., Mellits, E.D., Clark,
C , & the CERAD Investigators. (1989). The
Consortium to Establish a Registry for Alzheimer's
Disease (CERAD). Part 1. Clinical and neuropsy-
chological assessment of Alzheimer's disease.
Neurology, 39, 1159-1165.
Nyenhuis, D.L., & Gorelick, PB. (1998). Vascular
dementia: A contemporary review of epidemiology,
diagnosis, prevention, and treatment. Journal of the
American Geriatric Society, 46, 1437-1448.
Nyenhuis, D.L., Gorelick, PB., Freels, S., & Garron,
D. (2002). Cognitive and functional decline in
African-Americans with VaD, AD, and stroke
without dementia. Neurology, 58, 56-61.
Nyenhuis, D.L., Luchetta, T, Yamamoto, C , Terrien,
A., Bemardin, L., Rao, S.M., & Garron, D.C.
(1998). The development, standardization and
initial validation of the Chicago Multiscale Depres-
sion Inventory. Journal of Personality Assessment,
70,386-401.
Raven, J.C. (1995). Colored Progressive Matrices Sets
A, Ab, B. Oxford, England: University Press Ltd.
 NEUROPSYCHOLOGICAL DEHCITS IN VASCULAR CIND
Rockwood, K., Ebly, E., Hachinski, V.C., & Hogan, D.
(1997). Presence and treatment of vascular risk
factors in patients with vascular cognitive impair-
ment. Archives of Neurology, 54, 33-39.
Rockwood, K., Howard, K., MacKnight, C , & Darvesh,
S. (1999). Spectrum of disease in vascular cognitive
impairment. Neuroepidemiology, 18, 248-254.
Rockwood, K., Wentzel, C , Hachinski, V.C, Hogan,
D., MacKnight, C , & McDowell, I. (2000). Pre-
valence and outcomes of vascular cognitive impair-
ment. Neurology, 54, 447-451.
Roman, G.C., Tatemichi, T.K., Erkinjuntti, T.,
Cummings, J., Masdeu, J., Garcia, J., et al. (1993).
Vascular dementia: Diagnostic criteria for research
studies. Neurology, 43, 250-260.
Russell, E., & Starkey, R. (1993). Halstead-Russell
Neuropsychological Evaluation System (HRNES).
Los Angeles: Western Psychological Services.
Scherr, PA., Albert, M.S., Funkenstein, H.H., Cook,
N.R., Hennekens, C.H., Branch, L.G., White, L.R.,
Taylor, J.O., & Evans, D.A. (1988). Correlates of cog-
nitive function in an elderly community population.
American Journal of Epidemiology, 128, 1084—1101.
Shimamura, A., & Jurica, P. (1994). Memory inter-
ference effects and aging: Finds from a test of
frontal lobe function. Neuropsychology, 1,408-412.
49
Smith, A. (1991). Symbol Digit Modalities Test. Los
Angeles: Westem Psychological Services.
Starkstein, S.E., Sabe, L., & Vazquez, S. (1996).
Neuropsychological, psychiatric, and cerebral blood
flow findings in vascular dementia and Alzheimer's
disease. Stroke, 27, 408^14.
Thai, L.J., Grundman, M., & Golden, R. (1986).
Alzheimer's disease: A correlational analysis of the
Information-Memory-Concentration Test and the
Mini-Mental State Exam. Neurology, 36, 262-264.
Villardita, C. (1993). Alzheimer's disease compared
with cerebrovascular dementia: Neuropsychological
similarities and differences. Acta Psychiatrica
Scandinavica, 87, 299-308.
Wechsler, D. (1945). A standardized memory scale
for clinical use. Journal of Psychology, 19,
87-95.
Wechsler, D. (1997). WMS-III administration and
scoring manual. San Antonio, TX: The Psycholog-
ical Corporation, Harcourt Brace Jovanovich.
Wentzel, C , Rockwood, K., MacKnight, C , Hachinski,
V.C, Hogan, D.B., Feldman, H., Ostbye, T,
Wolfson, C , Gauthier, H., Verreau, H.R., &
McDowell, I. (2001). Progression of impairment in
patients with vascular cognitive impairment without
dementia. Neurology, 57, 714-716.