(Stroke. 1999;30:556-566.)
© 1999 American Heart Association, Inc.
Original Contributions |
Neuropsychological Impairment Correlates With Hypoperfusion and Hypometabolism but Not With Severity of White Matter Lesions on MRI in Patients With Cerebral Microangiopathy
From the Departments of Nuclear Medicine (O.S., D.H., M.S., H-J.K., U.B.), Neurology (R.S.), and Neuroradiology (M.M.), Aachen University of Technology, and the Department of Neurology (E-B.R.), University of Münster (Germany).
Correspondence to Osama Sabri, MD, Department of Nuclear Medicine, University of Technology (RWTH), Pauwelsstrasse 30, D-52057 Aachen, Germany.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Background and Purpose—Cerebral microangiopathy, indicated on MRI by lacunar infarctions (LI) and deep white matter lesions (DWML), is said to lead to vascular dementia.
Methods—Fifty-seven patients with proven cerebral microangiopathy were assessed for changes in regional cerebral blood flow (rCBF) and glucose metabolism (rMRGlu) and compared with 19 age-matched controls. The findings were correlated with results of extensive neuropsychological testing, as well as with MRI findings. A special head holder ensured reproducibility of positioning during rCBF (single-photon emission CT [SPECT]), rMRGlu (positron emission tomography [PET]), and MR imaging. White matter and cortex were quantified with regions of interest defined on MRI and superimposed to corresponding PET/SPECT slices. LI and DWML were graded by number and extent.
Results—Even with severe DWML and multiple LI, rCBF and rMRGlu values were not reduced. ANOVAs identified brain atrophy and neuropsychological deficits as the main determinants for reduced rCBF and rMRGlu values in both cortex and white matter. Neuropsychological deficits correlated well with decreased rCBF and rMRGlu, whereas MRI patterns such as LI and DWML did not. Factor analysis revealed no correlation of LI and DWML with rCBF, rMRGlu, atrophy, and neuropsychological deficits, showing instead positive correlations between rCBF, rMRGlu, and neuropsychological performance and negative correlations of the latter 3 with brain atrophy.
Conclusions—From these data, we conclude that LI and DWML are epiphenomena that may morphologically characterize cerebral microangiopathy but do not in themselves indicate cognitive impairment. Dementia or neuropsychological deficits, by contrast, are reflected exclusively by functional imaging parameters (rCBF, rMRGlu) and cerebral atrophy.
Key Words: magnetic resonance imaging • microangiopathy • neuropsychological testing • tomography, emission computed • vascular dementia
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The current importance of vascular dementia is highlighted in a recent overview by Erkinjuntti and Hachinski,1 who emphasize identification and understanding of the interactive vascular factors that contribute to cognitive impairment. Vascular dementia supposedly results from a hypertensive occlusive disease of the small penetrating arteries, known as cerebral microangiopathy (CMA),1 2 leading to lacunar infarctions (LI) and deep white matter lesions (DWML) that can be visualized with MRI. The cause of so-called vascular dementia is attributed mainly to the white matter lesions.3 It is assumed that microangiopathy causes damage of the periventricular white matter due to an unidentified perfusion deficit mechanism. Postmortem histological examinations have shown that occlusion of the small cerebral arteries (eg, of the lenticulostriate arteries and other long penetrating vessels) is the probable cause of LI.4 We are not aware of any anatomic proof that this is also true for the diffuse white matter lesions frequently seen in chronic hypertensives.
Several authors reported a decrease of regional cerebral blood flow (rCBF) or regional cerebral glucose metabolism (rMRGlu) in the cortical and subcortical gray matter, as well as in the white matter in patients with CMA.5 6 7 8 9
The aim of the present study was to ascertain whether CMA is accompanied by changes in rCBF and rMRGlu in either the white matter or the cortex or both and to determine how closely these changes correlate with the presence and type of neuropsychological deficits.
According to the results of the aforementioned studies, our hypotheses were as follows: (1) CMA patients with severe DWML and LI on MRI show significantly lower rCBF and rMRGlu values than do age-matched controls or patients with only minor white matter findings; and (2) neuropsychological deficits in those patients are correlated with severe DWML and decreased rCBF and rMRGlu.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This study was performed according to a protocol approved by the local ethics committee.
Patients and Controls
A total of 61 patients with typical neurological symptomatology
of CMA (motor stroke, sensory stroke, atactic hemiparesis,
dysarthria–clumsy hand syndrome [see below]) and LI and hypodensity
of the periventricular white matter as shown by CT were
screened for the study. These patients had already been under
examination by the Department of Neurology at the Aachen University of
Technology10 for several years, where they had been
subjected to radiological, epidemiological, rheological,
microcirculatory, and therapeutic follow-up.11 In all
patients, an occlusion of the large cerebral arteries (macroangiopathy)
and sources of cardiac embolism were excluded by extracranial and
transcranial Doppler sonography, as well as
transthoracic and transesophageal
echocardiography. Four patients were excluded
because of hemodynamically relevant stenoses of
the cerebral arteries. Further exclusion criteria were severe brain
injury and wedge-shaped hypodense lesions of the cortex (indicating
embolic infarction10 ) or white matter defects on CT
indicating infarction or multiple sclerosis. The final patient group
consisted of 25 women and 32 men aged 42 to 91 years (mean, 69±10
years). All patients showed signs of CMA on CT, ie, bilateral white
matter hypodensity and LI. None showed territorial or watershed
infarctions according to recently published
criteria.10 12
After onset of symptoms, clinical follow-up of the patients had already been performed for 6 weeks up to 10.6 years (mean, 3.0±2.6 years). At the time of first treatment, 27 patients exhibited a pure motor stroke, 7 had had a pure sensory stroke, 10 had suffered atactic hemiparesis, 5 exhibited a dysarthria–clumsy hand syndrome, and 8 showed mixed forms. At the time of this investigation, patients only exhibited residual symptoms of their previous neurological complaints. None of the patients received psychotropic medication, nor were any of them undergoing psychotherapy. Therefore, at the time of examination, patients were free of medication effects that might have influenced the study data. With respect to vascular risk factors, 52 patients (91%) exhibited chronic hypertension, 19 (33%) suffered from diabetes mellitus, 40 (70%) showed a disturbed fat metabolism (hyperlipidemia), and 35 (61%) used nicotine in one form or another.
Positron emission tomography (PET) (18-fluorodeoxyglucose [18-FDG] PET) and single-photon emission CT (SPECT) (99mTc–hexamethylpropyleneamine oxime [HMPAO] SPECT) were performed on the same day, and MRI was performed within the same week. To ensure reproducibility, for each run the patient's head was kept in the same position with a special head holder and a thermoplastic head mask, which have a very high accuracy and excellent reliability.13 In addition, each patient underwent a comprehensive neuropsychological test battery for evaluation of attention, memory, and cognition.
We also studied 19 age-matched controls (mean age, 65±13 years) with PET/SPECT/MRI without neurological or psychiatric abnormalities or morphological alterations of the brain, especially without any LI, DWML, or brain atrophy on MRI.
Imaging Procedure and Data Analysis
PET Protocol
Between 30 to 60 minutes after administration of 141 to 302 MBq
(mean, 240±40 MBq) 18-FDG, PET examination was done with an ECAT
953/15 scanner (Siemens/CTI). Patients were prepared by fasting
for 12 hours previously. To correct for photon attenuation,
transmission scans using 8 68Ge ring sources were
done. Input function was calculated by determining the plasma activity
of arterialized venous blood samples over time (2, 4, 6, 8,
10, 15, 20, 30, 45, and 60 minutes following injection of
18-FDG). Reconstruction of 45 transverse attenuation-corrected
slices of 3.375-mm thickness each in a 128x128 matrix was done with a
Hanning filter (cutoff frequency 0.5). Resolution was 6 mm full
width at half maximum (FWHM). Absolute glucose metabolism
was calculated for each pixel (after Sokoloff-Phelps14 )
and expressed in micromoles per 100 g brain tissue per minute with
the use of measured input function, tissue radioactivity concentration,
and blood glucose concentration. For quantification of gray and white
matter, 2 sets of rate constants
K1–k4
and a lumped constant of 0.52 were used, according to Reivich et
al.15
As a result of the influence of excessively high blood glucose levels on PET examination of cerebral metabolism, data from 5 patients who had not fasted as required were excluded from further analysis, since the required conditions for quantification had not been met.14
SPECT Protocol
Fifteen minutes after injection of 450 to 788 MBq (mean, 730±25
MBq) 99mTc-HMPAO (with the patients lying with
eyes closed in a darkened room), measurements were done with a
double-head Rota gamma camera (Siemens-Gammasonics) fitted with LEAP
(low-energy all-purpose) collimators. With a rotation of 2x180° in
3° steps, image acquisition took 30 seconds per projection.
Reconstruction of rCBF images was done in a 128x128 matrix using a
filtered back projection algorithm, a third-order Butterworth
filter with a cutoff frequency of 0.48, and an interslice factor of 2
with a slice thickness of 3.125 mm, with attenuation correction
according to Chang.16 Resolution was 15 mm
FWHM. rCBF was determined by normalization to the cerebellum. For
normalization to the cerebellum, a region of interest (ROI) of equal
size was generated in both cerebellar hemispheres, and the average ROI
counts were calculated as the reference value. Since no patient showed
LI or DWML in the cerebellum and since there were no significant
differences in cerebellar count rates between patients and healthy
controls, it can be assumed that no direct impairment of cerebellar
perfusion existed. Therefore, this reference region seems to be more
appropriate than normalization to the whole slice (which also contains
those areas with changed perfusion16 ).
MRI/CT Protocols
MRI was done with a circular polarized Helmholtz head coil in a
Magnetom 1.5-T apparatus (Siemens). With the use of
spin-echo technique, the brain was imaged in canthomeatal slices of
6 mm thickness (T1 weighting: echo time [TE] 19 ms, repetition
time [TR] 0.8 s; T2 weighting: TE 90 ms, TR 2.2 s; proton
weighting: TE 15 ms, TR 2.2 s). As part of the follow-up routine,
within up to 10 months before the beginning of this study, CT
transverse images were performed natively on a Somatom DR (Siemens)
with standard parameters.
Data Analysis
A special head holder coupled with a thermoplastic head mask
ensured that patients had exactly the same head position in all 3
examinations.13 After conversion of the data from MRI,
PET, and SPECT to a uniform data file structure, data were transferred
onto a computerized system for image analysis (Unix system;
SUN-SPARC). PET and SPECT were adapted transaxially to MRI (pixel size
in transaxial slicing: 1.802 mm), and layer thickness (MRI,
6.0 mm; PET, 3.375 mm; SPECT, 3.125 mm) was transformed
uniformly to 6.0 mm. With the use of the anatomic atlas by
Talairach,17 white matter (periventricular and
centrum semiovale), cortex (frontal, parietal, temporal, and
occipital), and subcortical gray matter (basal ganglia and thalamus)
were irregularly defined on T2-weighted MRI with ROIs in all slices
(114 ROIs per patient), which were superimposed on the respective
PET/SPECT slices (overlay method). All regions were defined for both
the right and the left hemispheres. Since morphological changes in CMA
(LI, DWML) and brain atrophy occur bilaterally, the average of the
corresponding left and right regions was used for further evaluation.
To account for possible lateralization effects in the
neuropsychological test findings, left and right regional values, as
well as their averages, were used for comparison with these
neuropsychological test findings. Since no significant differences
between contralateral ROIs were found, only average values are
given.
By means of quantifying the respective ROIs in several consecutive transaxial slices, volume-weighted rMRGlu values and rCBF ratios were determined for the quantified volume, thus minimizing influence of the partial volume effect as well as of statistical error. Volumes of evaluated regions were obtained as the product of all the pixels within each ROI and the volume of each individual pixel (temporal, 27±3 mL; frontal, 82±8 mL; parietal, 56±6 mL; occipital, 48±6 mL; centrum semiovale, 12±3 mL; periventricular, 32±6 mL; basal ganglia, 10±2 mL; thalamus, 8±1 mL; cerebellum, 8 mL). For the periventricular white matter, we first defined different ROIs around the frontal horns, in the central periventricular white matter, and around the posterior horns. Since there were no significant differences between the various locations in the periventricular white matter for any value in any of the patients, we could summarize them as one periventricular white matter region to increase the quantified volume of interest, thus minimizing influence of partial volume effects as well as of statistical error. SPECT evaluation of the centrum semiovale was not done since the FWHM of the gamma camera was 15 mm, so that the neighboring cortex caused an artificial increase in activity. Thus, only the periventricular white matter was evaluated in the rCBF SPECT examination. For SPECT, ROIs of the basal ganglia (ie, the caudate and lentiform nuclei) and of the thalamus were summarized as "subcortical gray matter" region (18±2 mL).
Scores
Because of the lack of a generally accepted score for
quantifying the severity of CMA on brain images, a special score was
developed for the purpose of this investigation, after which 4 blinded
neuroradiologically expert investigators allocated the patients to 4
groups: group 1, patients with only 1 to 3 LI and without DWML on MRI;
group 2, patients with 1 to 3 LI and with slight to moderate DWML;
group 3, patients with 4 to 10 LI and slight to moderate DWML; and
group 4, patients with >10 LI and severe, confluent DWML on MRI
(Figure 1
). The interobserver agreement
was high; assessment did not coincide in only 4 patients. In these
cases, MRI examinations were reevaluated by the whole team for a
final verdict.
|
A semiquantitative score for the degree of brain atrophy was also used.
Again, 4 neuroradiologically expert investigators judged the extent of
atrophy by CT and T1-weighted MRI images: group A, patients with no
inner or outer atrophy; group B, patients with slight inner and/or
outer atrophy; group C, patients with moderate inner and outer atrophy;
and group D, patients with severe inner and outer atrophy (Figure 2). Assessment of atrophy did not
coincide in only 5 patients. In these cases, CT and MRI examinations
were reevaluated by the team for a final verdict.
|
Neuropsychological Testing
On the day of the PET and SPECT examinations, patients underwent
an extensive neuropsychological test battery, with allowance of
sufficient time to rest (
4 hours) between PET/SPECT and
neuropsychological examination. Various tests were used to assess
cognitive and mnemonic abilities, as well as attentiveness. For
cognitive evaluation, we used the 7 subtests (numbered 1, 2, 3, 5, 6,
7, and 10) from the Performance Evaluation System by Horn in a
version for older people (LPS-50plus-K18 ). These tests
assessed verbal intelligence, abstract thinking, spatial imagination,
recognition of forms and figures, general knowledge, and
spelling.18 The results of these tests were quantified as
T values for (1) the Verbal Subtest result (assessing verbal skills),
consisting of the results of subtests 1, 2, 5, and 6, and (2) the
Nonverbal Subtest result (nonverbal skills), consisting of the results
of subtests 3, 7, and 10.
For mnemonic evaluation, the Recurring Words Test, Recurring Figures Test, and Digit Span were used. The Recurring Words Test assesses short-time memory learning ability for verbal material using 120 test cards showing 2-syllable nonsense words of high or low associative character.19 The Recurring Figures Test assesses short-time memory learning ability for nonverbal material using 160 test cards showing geometric or irregular stick figures that are difficult to verbalize. We used the version described by Büenfeld, normalized by Häger on a random sample of 400 non–brain-damaged test subjects.20 In the Digit Span, the patient's short-time memory for numbers was assessed by reiterating standardized strings of numbers composed of 3 to 9 elements.21
For assessing attentiveness, computer-assisted tests for Alertness and for Divided and Selective Attentiveness22 were used. In principle, these tests measure the reaction time after certain stimuli or combinations thereof, so that these tests can also be used to evaluate the speed of information processing.23 For the Alertness Test, a visual reaction task (screen image of a cross) was used to measure the visual reaction speed with and without an acoustic cue signal. For the Divided Attentiveness Test, a visual and an acoustic stimulus must both be observed to assess selective reaction by measuring the time between presentation of stimulus and reaction. The Selective Attentiveness Test measures the ability to suppress irrelevant stimuli: a screen presentation of 2 similar stimuli serves to gauge the reaction time under stimulus selection conditions. These tests are also called "Go/No Go" tests.23
The patients' neuropsychological performance could be
quantified as a percentile rank (showing the percentage of healthy
persons who did worse than the patients or equally well) of a large
collection of healthy test subjects,18 19 20 21 22 23 24 according to
which each patient was allocated to 1 of 2 groups relative to each
neuropsychological examination (Verbal and Nonverbal Subtests of the
Performance Evaluation System, Recurring Words, Recurring
Figures, Digit Span, Alertness, and Divided and Selective Attentiveness
Tests): (1) patients with T values <43, the mean value of these
patients corresponding to a percentile rank <5 (neuropsychologically
abnormal), and (2) patients with T values 
43, the mean value of these
patients corresponding to a percentile rank >92 (neuropsychologically
normal) (see Results).
Statistical Analysis
According to our hypotheses, we first performed a priori
single variable comparisons between the patient subgroups and
controls. The t tests for independent samples (which could
be used since all ROI values were normally distributed; Shapiro-Wilks
test and Lilliefors test; P>0.05) revealed any regional or
global 
CBF/MRGlu differences according to morphological criteria
(microangiopathy score, degree of atrophy) between the patient
subgroups and controls. The t tests for independent samples
also revealed rCBF/rMRGlu and CBF/MRGlu differences between the 2
groups divided according to neuropsychological test results
(neuropsychologically impaired versus normal CMA patients). Then
multiple ANOVAs were done for all regions of interest with the 3
factors—(1) neuropsychological grouping, (2) degree of atrophy, and
(3) microangiopathy score—to determine which factor or which
combination of the 3 factors caused the rCBF/rMRGlu and global
CBF/MRGlu differences. The F values are a measure of the explained
variance due to these factors. Calculation of significance levels
yielded the factors that had a significant influence on the observed
rCBF/rMRGlu changes. Two-way and higher interactions were not found.
The homogeneity of variances required for the ANOVAs was checked at the
5% level with the Bartlett-Box and Cochran C test.25
For testing hypotheses formulated a priori and followed by ANOVAs,
adjustment is not essential,25 but correction of the
error for multiple testing would not have changed the significance
of most of the results in this study.
According to results of principal-component factor analysis
with varimax rotation for reduced factor solution (Table 3) (see
Results), neuropsychological performance on all tests (n=9) was
summarized as (1) NPSsum1 (summarizing results of Alertness, Digit
Span, and Verbal/Nonverbal Subtests of the Performance
Evaluation System) and (2) NPSsum2 (summarizing results of Recurring
Figures, Recurring Words, and Divided and Selective Attentiveness).
Next, Spearman correlation coefficients were calculated between degree
of atrophy and microangiopathy, global CBF/MRGlu, NPSsum1, and
NPSsum2. Finally, we did principal-component factor analysis
with these 6 variables to investigate multivariate
correlations between white matter lesions, atrophy, neuropsychological
performance, and CBF/MRGlu (the ratio of number of
variables [6]/number of patients [57] <1/3 is still acceptable
for factor analysis26 ).
|
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Cerebral Microangiopathy Score and Regional/Global CBF and MRGlu Values
There were no significant differences in rCBF/rMRGlu and global
CBF/MRGlu between patients with morphological microangiopathy
severity degrees 1 to 4 and controls (Table 1). Not even patients with the most
extreme morphological microangiopathic changes (severity degree 4: >10
LI and severe, confluent DWML) showed any significant reductions in
either global CBF/MRGlu or in rCBF/rMRGlu in any ROI compared with
controls or with patients with degrees of severity 1 to 3.
|
Degree of Atrophy and Regional/Global CBF and MRGlu Values
There were significant reductions in both global CBF/MRGlu and
rCBF/rMRGlu in every ROI for patients with cerebral atrophy degrees C
(moderate inner and outer atrophy) and D (severe inner and outer
atrophy) compared with controls (Table 2). In contrast, there were no
significant reductions for either global CBF/MRGlu or for
rCBF/rMRGlu in any ROI between patients with degrees A (no
atrophy) and B (slight inner and/or outer atrophy) and controls
except for a marginal reduction of rMRGlu in the basal ganglia
of patients with degree B compared with controls
(P=0.065).
|
Factor Analysis for Variable Reduction
In the factorization of all 9 neuropsychological test results,
Scree test preceding-factor analysis revealed 2 factors with
eigen values >1 (Table 3
). The Kaiser-Meyer-Olkin Measure of Sampling
Adequacy for all variables was 0.80, which according to Kaiser et
al27 shows a good variable selection for factor
analysis. The Measure of Sampling Adequacy was >0.72 for every
variable. Factor 1 shows high loading for the results of Alertness,
Digit Span, and Verbal and Nonverbal Subtests of the
Performance Evaluation System, while factor 2 shows high
loading for the results of Recurring Figures, Recurring Words, and
Divided and Selective Attentiveness. Since in both of these factors 4
loadings (of 4 variables) were >0.6, this indicates a good
interpretation of the factor structures regardless of the sample size
(generalizing interpretation of factor structure is
allowed28 ). Therefore, a summarization to (1) NPSsum1
(summarizing results of Alertness, Digit Span, and Verbal and Nonverbal
Subtests of Performance Evaluation System) and (2) NPSsum2
(summarizing results of Recurring Figures, Recurring Words, and Divided
and Selective Attentiveness) was allowed regardless of sample
size.28
Summarized Neuropsychological Test Results (NPSsum1 and NPSsum2) in
Neuropsychologically Abnormal and Normal Patients
For NPSsum1, neuropsychological abnormal patients (n=22)
showed highly significantly reduced T values compared with the 35
neuropsychologically normal patients (33.02±3.22 versus 64.29±6.35;
P<0.00001). For NPSsum2, neuropsychologically abnormal
patients also showed highly significantly reduced T values compared
with neuropsychologically normal patients (31.40±6.11 versus
65.84±8.67; P<0.00001). A T value of 31 to 33 corresponds
to a percentile rank of 4, which shows the severe neuropsychological
impairment of the neuropsychologically abnormal group compared with the
normal group with T values in the range of 64 to 66, which correspond
to a percentile rank of 92 to 95.
Comparisons of Regional/Global CBF and MRGlu Values, Summarized
Test Results (NPSsum1 and NPSsum2), Degree of Atrophy, and
Microangiopathy Score
To ascertain whether deficits in NPSsum1 or NPSsum2 are linked to
global and regional CBF/MRGlu, these parameters were
compared between the groups with normal and poor neuropsychological
test performance. There were significantly lower global
CBF/MRGlu and rCBF/rMRGlu values in every ROI in NPSsum1 impaired
patients compared with NPSsum1 normal patients or age-matched controls,
while NPSsum1 normal patients and controls showed no significant
differences (Table 4). ANOVAs performed for NPSsum1 (homogeneity of
variances was given) with the 3 factors neuropsychological test result,
degree of atrophy, and microangiopathy score showed these global and
regional CBF/MRGlu reductions to be an effect of neuropsychological
grouping (except for rMRGlu in the basal ganglia and thalamus) and the
degree of atrophy in every ROI, but not of the microangiopathy score in
any ROI (Table 5).
|
|
There also were significantly lower global CBF/MRGlu and rCBF/rMRGlu
values in every ROI in NPSsum2 impaired patients compared with NPSsum2
normal patients or age-matched controls, while NPSsum2 normal patients
and controls showed no significant differences (Table 6). Likewise,
ANOVAs performed for NPSsum2 (homogeneity of variances was given) with
the 3 factors neuropsychological test result, degree of atrophy, and
microangiopathy score showed these global and regional CBF/MRGlu
reductions to be an effect of neuropsychological grouping (except for
rMRGlu in the basal ganglia and thalamus) and the degree of atrophy in
every ROI, but not of the microangiopathy score in any ROI (Table 7).
|
|
These results consistently show the same results as those
obtained by comparing every single rCBF/rMRGlu ROI value with
each of the 9 neuropsychological tests since there were highly
significantly lower rCBF/rMRGlu values (P<0.0005) in the
neuropsychologically abnormal group for all 9 single neuropsychological
tests in most of the ROIs even after correction of the error for
multiple testing. Similarly, multiple ANOVAs again showed these
reductions to be an effect of the neuropsychological grouping and/or
the degree of atrophy, while none of the ANOVAs could show a
significant influence of the degree of CMA (ie, the microangiopathy
score) on decreased rCBF/rMRGlu values in neuropsychologically abnormal
patients in any brain region (ROI).
Spearman correlation coefficients between the 6 variables atrophy,
microangiopathy, global perfusion (global CBF), global MRGlu, and the 2
summarized neuropsychological test results (NPSsum1 and NPSsum2) showed
strong and highly significant positive correlations between global CBF
and NPSsum1/NPSsum2 (r=0.61/r=0.57,
P<0.0005), global MRGlu and NPSsum1/NPSsum2
(r=0.60/r=0.58, P<0.0005), and strong
and highly significant negative correlations of atrophy with global
CBF/MRGlu (r=-0.61/r=-0.70;
P<0.0005) and with NPSsum1/NPSsum2
(r=-0.64/r=-0.69; P<0.0005).
Microangiopathy did not correlate significantly with any of the other
variables (all r<0.09; P>0.5).
In the factorization of these 6 variables, Scree test
preceding-factor analysis revealed 2 factors with eigen values
1 (Table 8). The Kaiser-Meyer-Olkin Measure was 0.85, and the Measure
of Sampling Adequacy was always >0.75. Factor 1 combines global
perfusion and metabolism and shows strong positive
correlations with the 2 summarized neuropsychological test results, as
well as strong negative correlations with the degree of atrophy. Factor
2 is defined by the degree of microangiopathy and shows no strong
correlations with any of the other variables.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This study focused on a type of vascular brain disease caused by CMA without concomitant macroangiopathic vascular lesions. Occasionally, cerebral embolism has been discussed as a possible cause of LI, but this is a very rare event, if possible at all.10 Torvik29 found that showers of microemboli can cause lesions of lacune size in subcortical and cortical watershed areas and considered this the probable explanation for watershed infarctions in cases of carotid thrombosis. However, carotid thrombosis (macroangiopathy as well as cardiac embolism) was an exclusion criterion in our study, making it unlikely that Torvik's hypothesis applies to our patients. The age distribution, risk factor profile, and neurological symptomatology of our cohort are well comparable to cohorts with small-vessel disease from other studies in this field.30 Furthermore, the cohort exhibited no case with multi-infarction dementia caused by multiple macroangiopathy-induced brain lesions or brain embolism.
On initial MRI, LI and disseminated or even confluent DWML gave a
consistent morphological correlation with our patient group. As
judged by a council of 4 neuroradiological experts, LI were defined as
sharply bounded, round, or oval foci located at the basal ganglia,
capsula interna, lower corona radiata, and brain stem, 
10 mm in
diameter, and appearing hyperintense on T2- and hypointense on
T1-weighted MRI. The interobserver agreement was high. Regarding the
reliability of quantifying the morphological severity of CMA (LI,
leukoaraiosis), Schneider et al31 had demonstrated a fair
to good interrater agreement and suggested a council of 3 evaluators.
In particular, we focused our attention on thalamic LI, which could
possibly underlie the clinical syndrome of thalamic dementia. In
particular, bilateral polar lesions of the thalamus are seen as the
cause of the syndrome of thalamic dementia.32 All of our
patients were additionally screened for such classic syndromes at
admission, and none of them exhibited any of these. Perhaps no classic
dementia syndromes of the thalamus were observed because we limited the
maximum LI diameter to 10 mm; the narrower the definition of LI,
the more etiologically homogeneous the patient group will
be. One could argue, therefore, that the conclusions put forth here
cannot without exception be extrapolated to all types of LI. However,
Román33 showed that dementia was not related to the
number or location of the LI and emphasized the importance of DWML as
the cause of cortical disconnection leading to vascular dementia.
In this study the degree of brain atrophy was classified by a council of 4 investigators with neuroradiological experience on CT and T1-weighted MRI images. This semiquantitative assessment, also used by Leys et al2 in the evaluation of cerebral atrophy in patients with CMA, proved feasible and reliable and comparable in clinical value to the computerized volumetric method.2 34 35 36 Indeed, the strong correlation between the atrophy assessment and the reduced functional parameters (rCBF/rMRGlu, neuropsychological performance) clearly showed the usefulness of our approach for clinical practice, as we recently also showed for other groups of mentally ill patients.16
There is no agreement in the literature regarding the clinical relevance of DWML. DWML are even present in a number of elderly persons without any neurological signs.37 Fein et al38 examined patients with extensive DWML on MRI and found no major cognitive or focal neurological deficits. On the other hand, van Swieten et al39 observed that hypertensive patients with confluent DWML on T2-weighted MRI showed a more pronounced cognitive impairment than hypertensive patients with only "patchy or punctate" hyperintensities or than normointense patients. The true cause of DWML is still unproved. It is questionable whether these lesions reflect vascular dementia at all.40 41 42 Criteria for the diagnosis of this disorder have recently been proposed43 but have not yet gained general acceptance.
Several authors reported a decrease of rCBF or rMRGlu in the cortex as
well as in the white matter in microangiopathic
patients.5 6 7 8 9 Furthermore, DeCarli et al44
showed that in healthy subjects (aged 19 to 91 years) DWML were
correlated with reduced rMRGlu. In contrast, on the basis of data
obtained with retinal video fluorescence angiography, Schneider
et al11 questioned a causal relationship between white
matter hypodensity on CT (leukoaraiosis) and microangiopathic changes.
In the present study, done on a comparatively large number of 57
patients, no significant reduction in rCBF/rMRGlu could be shown to be
related to the degree of LI and DWML (Table 1).
Therefore, our hypothesis 1 could not be confirmed. This does not agree
with the studies cited above.5 6 7 8 9 44 45 The latter,
however, failed to consider brain atrophy in their assessment of
rCBF/rMRGlu, even though brain atrophy exists in most patients with
leukoaraiosis.2 46 Since at this time no generally
accepted score for quantifying the severity of DWML on MRI exists, we
developed a special microangiopathy score for our study, as mentioned
earlier. It could be argued that the failure to find a correlation
between DWML and reduced rCBF/rMRGlu may be due to the possibility of
our microangiopathy score not being sufficiently sensitive to
demonstrate such a relationship, rather than reflecting the fact that
no relationship exists. In our study, however, even patients with
severe confluent DWML and multiple (>10) LI without brain atrophy did
not show rCBF/rMRGlu reductions in any of the ROIs compared with
healthy age-matched controls (Table 1), which argues
strongly against this objection. The present study clearly shows
brain atrophy to be accompanied by a pronounced reduction in rCBF and
rMRGlu affecting both the gray and the white matter (Table 2).
Regarding vascular dementia, Erkinjuntti and Hachinski1
proposed a relationship between microangiopathic brain lesions and
cognitive deficits. There are conflicting claims concerning the
correlation between neuropsychological findings and changes in
functional/morphological imaging, with some authors reporting a
correlation between PET/MRI and neuropsychological impairment and
others not.38 39 45 Furthermore, there is not yet one
generally accepted test or battery for evaluating a person's
intellectual abilities, making comparison between studies difficult.
Often-used simple bedside tests such as the Mini-Mental State
Examination are also objects of controversy. According to
Poeck,47 this procedure does not meet the standards of
modern psychometric examinations. In this study, all patients underwent
a series of well-established and validated neuropsychological
tests18 19 20 21 22 23 to assess their cognitive and mnemonic
functions, as well as their attentiveness. In every test,
neuropsychologically abnormal patients showed highly significantly
reduced rCBF/rMRGlu in most brain regions. The microangiopathy score
had no significant effect in any of the cases. The same applies to the
summarized functional (global CBF/MRGlu) and neuropsychological
(NPSsum1/NPSsum2) results (Tables 3 to 7). The first part
of our hypothesis 2 (correlation of neuropsychological deficits with
severe DWML) could therefore not be confirmed.
Some authors suggested that the location of periventricular
white matter lesions could influence neuropsychological
status.48 Therefore, we did not perform only a global
analysis. In fact, we first analyzed the
periventricular white matter at each location
independently. Since there were no significant differences for
rCBF/rMRGlu between the various locations in the white matter, we could
summarize them as one periventricular white matter region
to increase the quantified volume of interest. Furthermore, the
distribution of white matter lesions did not differ significantly
between patients with and without neuropsychological deficits. In our
study (as evident from our classification into 4 degrees, Table 1), the size of hyperintensities varied considerably,
and therefore we defined white matter ROIs anatomically on MRI
regardless of the size a hyperintense area showed in any patient.
However, even the 18 patients with severe, confluent (degree 4, Table 1) hyperintensities—whose
periventricular white matter included ROIs that consisted
almost entirely of hyperintense rather than normointense white
matter—did not show reduced rCBF/rMRGlu values, compared with either
normointense white matter in the same patient or with age-matched
healthy controls free of hyperintensities. Therefore, we concluded that
smaller hyperintensities within a ROI of normointense white matter
would likewise show no reductions that could have been missed by
evaluating the whole ROI instead of isolating the hyperintensity.
In contrast to our findings, a study on patients with DWML on MRI
showed some associations with impaired cognitive
function,49 but the partial correlation coefficients
reported had not been adjusted for brain atrophy. According to our
results, brain atrophy is the main morphological determinant of
functional deficits. Furthermore, as determined by the ANOVAs,
neuropsychological deficits correlate with functional deficits
(rCBF/rMRGlu) even in the absence of brain atrophy. This confirms the
second part of our hypothesis 2 (correlation of neuropsychological
deficits with decreased rCBF/rMRGlu). The use of various statistical
methods, such as single comparisons followed by multiple ANOVAs,
univariate Spearman correlations, and
multivariate factor analyses (Tables 1 to 8),
consistently failed to show any significant correlations of
the morphological degree of CMA with either brain
perfusion/metabolism, neuropsychological impairment, or
atrophy. Again, the brain atrophy present in most patients with
DWML2 46 seems to be the decisive morphological correlate.
Therefore, Leys et al2 questioned whether
neuropsychological impairment in such patients might be due to atrophy
rather than due to DWML.
Until now, little effort had been made to establish a connection between brain atrophy and vascular dementia. Using regression analysis, Kobari et al6 identified atrophy and age as the essential factors causing leukoaraiosis. Unfortunately, the rCBF decrease detected in the same study was not adjusted for atrophy. Drayer50 described ventricular dilation and a broadening of the sulci in patients with Binswanger's disease as indicators of brain atrophy. As a potential correlate, electron microscopy has shown a clear reduction of nerve fiber density in the frontal white matter, as well as a reduced oligodendrocyte and astrocyte density in the deep white matter in patients with vascular dementia.51
We found rCBF/rMRGlu reductions in both the cortex and the white matter correlating with neuropsychological impairment. Indeed, not one of our cases showed reductions only in the cortex or in the white matter. The question of whether disconnection or direct damage to the cortex (eg, incomplete infarction), independent of subcortical disease, is responsible for dementia is not easy to answer. The fact that LI/DWML do not correlate with cognitive impairment does not support the disconnection theory often cited as an explanation for reduced parameters in the cortex, ie, the cerebral cortex disconnected from its DWML-affected projecting subcortical structures.33 A follow-up examination of our patients after 1 year revealed no progression of brain atrophy or of any other cortical damage. Thus, the cortical damage theory also seems unlikely in this context.
Most of our patients showed hypometabolism/hypoperfusion both cortically as well as subcortically, without either a frontal or a parietotemporal predominance. One patient did exhibit frontally predominant reductions, with a slighter but nevertheless significant reduction of the other values. Likewise, 6 patients showed parietotemporally predominant reductions, again with a slighter but nevertheless significant reduction of the other values. In these latter 6 cases, it is not possible to definitely rule out an additional diagnosis of Alzheimer's disease. This does not, however, change the main findings of our study.
The present results allow no conclusion as to whether brain atrophy (which correlates with a substantial loss of performance on neuropsychological tests in patients with CMA) is an epiphenomenon or whether it reflects a disease (microangiopathy)-specific feature. This is an important question that warrants further research.
Conclusions
In our study we showed that neuropsychological deficits are not
correlated with LI and DWML. Fein et al38 showed that
patients with extensive DWML on MRI need not necessarily exhibit
cognitive, behavioral, or focal neurological deficits. On the basis of
data obtained with retinal video fluorescence angiography,
Schneider et al11 questioned a causal relationship between
DWML and microangiopathic changes. Neuropsychological deficits are
correlated, however, with reduced functional PET/SPECT
parameters (rCBF/rMRGlu) and, if present, with brain
atrophy. The conclusion, therefore, must be that LI/DWML are merely
epiphenomena that may morphologically characterize patients with CMA,
but that the coincidence of neuropsychological deficits and LI/DWML in
itself does not allow a diagnosis of vascular dementia. This need not
invalidate the concept of vascular dementia since the functional
parameters do correlate with cognitive impairment. Of
course, there is a vascular component, but it cannot be detected
morphologically using the extent of LI and that of DWML as
criteria.
|
Acknowledgments |
|---|
This study was supported by a grant from the Deutsche Forschungsgemeinschaft (DFG RI 415/4-1). E-B. Ringelstein, O. Sabri, R. Schneider, D. Hellwig, M. Schreckenberger, H-J. Kaiser, and U. Buell were responsible for conception and design of the study; O. Sabri, D. Hellwig, M. Schreckenberger, M. Mull, and H-J. Kaiser for data evaluation; and E-B. Ringelstein, R. Schneider, and U. Buell for critical revision of the article. Thanks are also due to A. Rodón for translation and editing and to C. Doherty, S. Dettki, and S. Luerken for patient management. Furthermore, we would also like to thank B. Fimm for assistance with the analysis of the neuropsychological data and H.J. Kunert and K. Willmes for assistance with the statistical analysis.
Received August 31, 1998; revision received December 22, 1998; accepted December 22, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
1. Erkinjuntti T, Hachinski V. Rethinking vascular dementia. Cerebrovasc Dis. 1993;3:3–23.
2. Leys D, Pruvo JP, Scheltens PH, Rondepierre P, Godefroy O, Leclerc X, de Reuck J. Leuko-araiosis: relationship with the types of focal lesions occurring in acute cerebrovascular disorders. Cerebrovasc Dis. 1992;2:169–176.
3. Filley ChM, Franklin GM, Heaton RK, Rosenberg NL. White matter dementia: clinical disorders and implications. Neuropsychiatry Neuropsychol Behav Neurol. 1989;239–254.
4. Fisher CM. The arterial lesions underlying lacunes. Acta Neuropathol (Berl). 1969;12:1–15.
5. Reiche W, Weiller C, Weigmann R, Kaiser HJ, Büll U, Schneider R, Ringelstein EB. Vergleich von MRT- und SPECT-Befunden bei Patienten mit zerebraler Mikroangiopathie. Nuklearmedizin. 1991;30:161–169.[Medline] [Order article via Infotrieve]
6.
Kobari M, Meyer JS, Ichio M. Leuko-araiosis, cerebral
atrophy and cerebral perfusion in normal aging. Arch Neurol. 1990;47:161–165.
7.
Yao H, Sadoshima S, Kuwabara Y, Ichiya Y, Fujishima M.
Cerebral blood flow and oxygen consumption in patients with vascular
dementia of the Binswanger type. Stroke. 1990;21:1694–1699.
8.
Kawamura J, Meyer JS, Terayama Y, Weathers S.
Leukoaraiosis correlates with cerebral hypoperfusion in vascular
dementia. Stroke. 1991;22:609–614.
9.
Mielke R, Herholz K, Grond M, Kessler J, Heiss W-D.
Severity of vascular dementia is related to volume of
metabolic impaired tissue. Arch Neurol. 1992;49:909–913.
10. Ringelstein EB, Koschorke S, Holling A, Thron A, Lambertz H, Minale C. CT patterns of proven embolic brain infarctions. Ann Neurol. 1989;26:759–765.[Medline] [Order article via Infotrieve]
11.
Schneider R, Rademacher M, Wolf S. Lacunar infarcts and
white matter attenuation: ophthalmological and microcirculatory aspects
of the pathophysiology. Stroke. 1993;24:1874–1879.
12. Ringelstein EB, Weiller C, Weckesser M, Weckesser S. Cerebral vasomotor reactivity is significantly reduced in low-flow as compared to thromboembolic infarctions: the key role of the circle of Willis. J Neurol Sci. 1994;121:103–109.[Medline] [Order article via Infotrieve]
13. Kaiser HJ, Sabri O, Wagenknecht G, Lege B, Hellwig D, Buell U. A method of correlating and merging cerebral morphology and function by a special head holder. Nuklearmedizin. 1994;33:123–126.[Medline] [Order article via Infotrieve]
14. Phelps ME, Huang SC, Hoffmann EJ, Selin C, Sokoloff L, Kuhl DE. Tomographic measurement of local cerebral glucose metabolic rate in humans with (F-18) 2-fluoro-2-deoxyglucose: validation of a method. Ann Neurol. 1979;6:371–388.[Medline] [Order article via Infotrieve]
15. Reivich M, Alavi A, Wolf AP, Fowler J, Russell, Arnett C, MacGregor RR, Shiue CY, Atkins H, Anand A, Dann R, Greenberg JH. Glucose metabolic rate kinetic model parameter determination in man: the lumped constant and rate constants for (18-F) fluorodeoxyglucose and (11-C) deoxyglucose. J Cereb Blood Flow Metab. 1985;5:179–192.[Medline] [Order article via Infotrieve]
16. Sabri O, Erkwoh R, Schreckenberger M, Owega A, Sass H, Buell U. Correlation of positive symptoms to hyperperfusion or hypoperfusion of cerebral cortex in never-treated schizophrenics. Lancet. 1997;349:1735–1739.[Medline] [Order article via Infotrieve]
17. Talairach J, Szlika G. Atlas d'anatomie stéréotaxique du télencéphale. Paris, France: Masson; 1967.
18. Sturm W, Willmes K, Horn W. Leistungsprüfsystem für 50-90jährige (LPS 50+). Göttingen, Germany: Hogrefe; 1993.
19. Sturm W, Willmes K. Verbaler Lerntest. Mödling, Germany: Schuhfried; 1994.
20. Sturm W, Willmes K. Nonverbaler Lerntest. Mödling, Germany: Schuhfried; 1994.
21. Wechsler D. Wechsler Memory Scale—Revised. New York, NY: Psychological Corporation; 1987.
22. Zimmermann P, Fimm B. Testbatterie zur Aufmerksamkeitsprüfung (TAP). Freiburg, Germany: PSYTEST; 1993.
23. von Cramon DY, Mai N, Ziegler W. Neuropsychologische Diagnostik. 1st ed. Weinheim, Germany: VCH; 1993.
24. Becker M, Sturm W, Willmes K, Zimmermann P. Normierungsstudie zur Aufmerksamkeitstestbatterie (TAP) von Zimmermann und Fimm. Z Neuropsychol.. 1996;7:3–15.
25. Bortz J. Varianzanalytische Methoden. In: Bortz J, ed. Statistik für Sozialwissenschaftler. 4th ed. Berlin, Germany: Springer; 1993:223–407.
26. Flury B, Riedwyl H. Faktorenanalyse. In: Flury B, Riedwyl H, eds. Angewandte multivariate Statistik. Stuttgart, Germany: Fischer; 1983:135–149.
27. Kaiser HF. An index of factorial simplicity. Psychometrika. 1974;39:31–36.
28. Guadagnoli E, Velicer WF. Relation of sample size to the stability of component patterns. Psychol Bull. 1988;103:265–275.[Medline] [Order article via Infotrieve]
29.
Torvik A. The pathogenesis of watershed infarctions in
the brain. Stroke. 1984;15:221–223.
30.
Babikian V, Ropper AH. Binswanger's disease: a review.
Stroke. 1987;18:2–12.
31. Schneider R, Kluge R, Willmes K. Interrater agreement for CT scans of patients with lacunar infarcts and leuko-araiosis. Acta Neurol Scand. 1991;84:527–530.[Medline] [Order article via Infotrieve]
32.
Levasseur M, Baron JC, Sette G,
Legault-Démaré F, Pappata S, Mauguière F,
Benoît N, Tran Dinh S, Dégos JD, Laplane D, Mazoyer B.
Brain energy metabolism in bilateral paramedian thalamic
infarcts: a positron emission tomography study. Brain. 1992;115:795–807.
33.
Román GC. From UBOs to Binswanger's disease:
impact of magnetic resonance imaging on vascular dementia research.
Stroke. 1996;27:1269–1273.
34. Leys D, Pruvo JP, Petit H, Gaudet Y, Clarisse J. Maladie d'Alzheimer: analyse statistique des résultats du scanner X. Rev Neurol (Paris). 1989;145:134–139.[Medline] [Order article via Infotrieve]
35. Le May M, Stafford JL, Sándor T, Albert M, Haykal H, Zamani A. Statistical assessment of perceptual CT scan ratings in patients with Alzheimer dementia. J Comput Assist Tomogr. 1986;10:802–809.[Medline] [Order article via Infotrieve]
36. Laffey PA, Peyster RG, Nathan R, Haskin ME, McGinley JA. Computed tomography and aging: results in a normal elderly population. Neuroradiology. 1984;26:273–278.[Medline] [Order article via Infotrieve]
37.
Chimowitz MI, Awad IA, Furlan AJ.
Periventricular lesions on MRI: facts and theories.
Stroke. 1989;20:963–967.
38.
Fein G, van Dyke C, Davenport L, Turetsky B,
Brant-Zawadzki M, Zatz L, Dillon W, Valk P. Preservation of normal
cognitive functioning in elderly subjects with extensive white-matter
lesions of long duration. Arch Gen Psychiatry. 1990;47:220–223.
39. van Swieten JC, Geyskes GG, Derix MMA, Peeck BM, Ramos LM, van Latum JC, van Gijn J. Hypertension in the elderly is associated with white matter lesions and cognitive decline. Ann Neurol. 1991;30:825–830.[Medline] [Order article via Infotrieve]
40. Brust JCM. Vascular dementia reconsidered, part 2. Cerebrovasc Dis. 1993;18:2–12.
41.
Drachman DA. New criteria for the diagnosis of vascular
dementia: do we know enough yet? Neurology. 1993;43:243–245.
42. O'Brian MB. Vascular dementia reconsidered, part 1. Cerebrovasc Dis. 1993;3:24–25.
43.
Román GC, Tatemichi TK, Erkinjuntti T, Cummings
JL, Masdeu JC, García JH, Amaducci L, Orgogozo JM, Brun A,
Hofman A. Vascular dementia: diagnostic criteria for
research studies: report of the NINDS-AIREN International Workshop.
Neurology. 1993;43:250–260.
44.
DeCarli C, Murphy DGM, Tranh M, Grady CL, Haxby JV,
Gillette JA, Salerno JA, Gonzales-Aviles A, Horwitz B, Rapoport SI,
Schapiro MB. The effect of white matter hyperintensity volume on brain
structure, cognitive performance, and cerebral
metabolism of glucose in 51 healthy adults.
Neurology. 1995;45:2077–2084.
45. Meyer E, Delpla P, Pertides E, Ethier R, Bès A, Hakim AM. PET metabolic and neuropsychological correlates of periventricular lucencies. In: Hartmann A, Kuschinsky W, Hoyer S, eds. Cerebral Ischemia and Dementia. Berlin, Germany: Springer-Verlag; 1991:368–373.
46.
Mirsen TR, Lee DH, Wong CJ, Díaz JF, Fox AJ,
Hachinski VC, Merskey H. Clinical correlates of white-matter changes on
magnetic resonance imaging scans of the brain. Arch Neurol. 1991;48:1015–1021.
47. Poeck K. Der Beitrag der Neuropsychologie zur Diagnose und Erforschung der Demenzen. In: Häfner H, Hennerici M, eds. Psychische Krankheiten und Hirnfunktion im Alter. Stuttgart, Germany: Gustav Fischer; 1992:91–103.
48. De Reuck J, Decoo D, Strijckmans K, Lemahieu I. Does the severity of leukoaraiosis contribute to senile dementia? A comparative computerized and positron emission tomographic study. Eur Neurol. 1992;32:199–205.[Medline] [Order article via Infotrieve]
49.
Longstreth WT Jr, Manolio TA, Arnold A, Burke GL, Bryan
N, Jungreis CA, Enright PL, O'Leary D, Fried L. Clinical correlates of
white matter findings on cranial magnetic resonance imaging of 3301
elderly people. Stroke. 1996;27:1274–1282.
50.
Drayer BP. Imaging of the aging brain, part II:
pathological conditions. Radiology. 1988;166:797–806.
51. Yamanouchi H. Loss of white matter oligodendrocytes and astrocytes in progressive subcortical vascular encephalopathy of Binswanger type. Acta Neurol Scand. 1991;83:301–305.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  P. M. Meyer, K. Strecker, K. Kendziorra, G. Becker, S. Hesse, D. Woelpl, A. Hensel, M. Patt, D. Sorger, F. Wegner, et al. Reduced {alpha}4{beta}2*-Nicotinic Acetylcholine Receptor Binding and Its Relationship to Mild Cognitive and Depressive Symptoms in Parkinson Disease Arch Gen Psychiatry, August 1, 2009; 66(8): 866 - 877. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Nitkunan, T. R. Barrick, R. A. Charlton, C. A. Clark, and H. S. Markus Multimodal MRI in Cerebral Small Vessel Disease: Its Relationship With Cognition and Sensitivity to Change Over Time Stroke, July 1, 2008; 39(7): 1999 - 2005. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P M Walker, D Ben Salem, M Giroud, and F Brunotte Is NAA reduction in normal contralateral cerebral tissue in stroke patients dependent on underlying risk factors? J. Neurol. Neurosurg. Psychiatry, May 1, 2006; 77(5): 596 - 600. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Scheid, K. Walther, T. Guthke, C. Preul, and D. Y. von Cramon Cognitive sequelae of diffuse axonal injury. Arch Neurol, March 1, 2006; 63(3): 418 - 424. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. G. Leblanc, J. F. Meschia, D. T. Stuss, and V. Hachinski Genetics of Vascular Cognitive Impairment: The Opportunity and the Challenges Stroke, January 1, 2006; 37(1): 248 - 255. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. R. Reed, J. L. Eberling, D. Mungas, M. Weiner, J. H. Kramer, and W. J. Jagust Effects of White Matter Lesions and Lacunes on Cortical Function Arch Neurol, October 1, 2004; 61(10): 1545 - 1550. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. M. Wen, V. C.T. Mok, Y. H. Fan, W. W.M. Lam, W. K. Tang, A. Wong, R. X. Huang, and K. S. Wong Effect of White Matter Changes on Cognitive Impairment in Patients With Lacunar Infarcts Stroke, August 1, 2004; 35(8): 1826 - 1830. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Tullberg, E. Fletcher, C. DeCarli, D. Mungas, B. R. Reed, D. J. Harvey, M. W. Weiner, H. C. Chui, and W. J. Jagust White matter lesions impair frontal lobe function regardless of their location Neurology, July 27, 2004; 63(2): 246 - 253. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ihara, H. Tomimoto, K. Ishizu, T. Mukai, H. Yoshida, N. Sawamoto, M. Inoue, T. Doi, K. Hashikawa, J. Konishi, et al. Decrease in Cortical Benzodiazepine Receptors in Symptomatic Patients With Leukoaraiosis: A Positron Emission Tomography Study Stroke, April 1, 2004; 35(4): 942 - 947. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. O'Sullivan, S. Singhal, R. Charlton, and H. S. Markus Diffusion tensor imaging of thalamus correlates with cognition in CADASIL without dementia Neurology, March 9, 2004; 62(5): 702 - 707. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M O'Sullivan, R G Morris, B Huckstep, D K Jones, S C R Williams, and H S Markus Diffusion tensor MRI correlates with executive dysfunction in patients with ischaemic leukoaraiosis J. Neurol. Neurosurg. Psychiatry, March 1, 2004; 75(3): 441 - 447. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. H. Sweet, R. H. Paul, R. A. Cohen, D. Moser, B. R. Ott, N. Gordon, J. N. Browndyke, P. Shah, and K. D. Garrett Neuroimaging Correlates of Dementia Rating Scale Performance at Baseline and 12-Month Follow-up among Patients with Vascular Dementia J Geriatr Psychiatry Neurol, December 1, 2003; 16(4): 240 - 244. [Abstract] [PDF]
|
|
|
|
|
|
F. C. Bakker, C. J.M. Klijn, A. Jennekens-Schinkel, I. van der Tweel, J. van der Grond, A. C. van Huffelen, C. A.F. Tulleken, and L. J. Kappelle Cognitive Impairment Is Related to Cerebral Lactate in Patients With Carotid Artery Occlusion and Ipsilateral Transient Ischemic Attacks Stroke, June 1, 2003; 34(6): 1419 - 1424. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Sabri, A. Owega, M. Schreckenberger, L. Sturz, B. Fimm, P. Kunert, P. T. Meyer, D. Sander, and J. Klingelhofer A Truly Simultaneous Combination of Functional Transcranial Doppler Sonography and H215O PET Adds Fundamental New Information on Differences in Cognitive Activation Between Schizophrenics and Healthy Control Subjects J. Nucl. Med., May 1, 2003; 44(5): 671 - 681. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. O'Sullivan, P. E. Summers, D. K. Jones, J. M. Jarosz, S. C.R. Williams, and H. S. Markus Normal-appearing white matter in ischemic leukoaraiosis: A diffusion tensor MRI study Neurology, December 26, 2001; 57(12): 2307 - 2310. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Zuccala, G. Onder, C. Pedone, L. Carosella, M. Pahor, R. Bernabei, and A. Cocchi Hypotension and cognitive impairment: Selective association in patients with heart failure Neurology, December 11, 2001; 57(11): 1986 - 1992. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-Y. Hsu, A.-T. Du, N. Schuff, and M. W. Weiner Magnetic Resonance Imaging and Magnetic Resonance Spectroscopy in Dementias J Geriatr Psychiatry Neurol, September 1, 2001; 14(3): 145 - 166. [Abstract] [PDF]
|
|
|
|
|
|
R. H. Paul, R. A. Cohen, D. Moser, B. R. Ott, T. Zawacki, N. Gordon, S. Bell, and W. Stone Performance on the Mattis Dementia Rating Scale in Patients with Vascular Dementia: Relationships to Neuroimaging Findings J Geriatr Psychiatry Neurol, March 1, 2001; 14(1): 33 - 36. [Abstract] [PDF]
|
|
|
|
|
|
H. Chabriat, S. Pappata, L. Ostergaard, C. A. Clark, M. Pachot-Clouard, K. Vahedi, A. Jobert, D. Le Bihan, and M. G. Bousser Cerebral Hemodynamics in CADASIL Before and After Acetazolamide Challenge Assessed With MRI Bolus Tracking Stroke, August 1, 2000; 31(8): 1904 - 1912. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Terborg, F. Gora, C. Weiller, and J. Rother Reduced Vasomotor Reactivity in Cerebral Microangiopathy : A Study With Near-Infrared Spectroscopy and Transcranial Doppler Sonography Stroke, April 1, 2000; 31(4): 924 - 929. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. H. Taber, J. G. Zimmerman, H. Yonas, W. Hart, and R. A. Hurley Applications of Xenon CT in Clinical Practice: Detection of Hidden Lesions J Neuropsychiatry Clin Neurosci, November 1, 1999; 11(4): 423 - 425. [Full Text] [PDF]
|
|
|
|
|
|
I. Puls, K. Hauck, K. Demuth, A. Horowski, M. Schlie{beta}er, P. Dorfler, P. Scheel, K. V. Toyka, K. Reiners, M. Schoning, et al. Diagnostic Impact of Cerebral Transit Time in the Identification of Microangiopathy in Dementia : A Transcranial Ultrasound Study Stroke, November 1, 1999; 30(11): 2291 - 2295. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||