Corpus Callosum Atrophy in Patients With Leukoaraiosis May Indicate Global Cognitive Impairment
Background and Purpose—The extent of white matter high-intensity lesions (WMLs) on T2-weighted MR images may be an indicator of cognitive impairment, especially impairment related to frontal lobe dysfunction. However, it is unclear whether the extent of WMLs is an independent predictor of cognitive impairment. In patients with extensive WMLs, atrophy of the corpus callosum may be an important predictor of global cognitive impairment. The purpose of this study was to investigate the relation of the extent of WMLs and callosal size with cognitive functions in a patient population with a wide range of extent of WMLs.
Methods—We studied 62 patients, aged 49 to 86 years, who underwent MRI because of neurological symptoms and were diagnosed as having lacunar stroke or no specific neurological disease: 28 with lacunar infarcts and 34 without. Multivariate analysis was used to test the independent predictive value of patient age, sex, educational level, other medical illness, lacunar infarct, corpus callosum area, and extent of WMLs with respect to scores of Mini-Mental State Examination or verbal fluency task.
Results—Only callosal size and age were significant independent predictors of the scores of the Mini-Mental State Examination, while only the extent of WMLs was an independent predictor of the score of the verbal fluency task.
Conclusions—Callosal atrophy may be an important predictor of global cognitive impairment in patients with WMLs, whereas the extent of WMLs per se may be related to impairment of frontal lobe function independent of callosal atrophy.
White matter high-intensity lesions (WMLs) on T2-weighted MR images are frequently detected in elderly people.1 2 In subjects without a history of stroke, the extent of WMLs may be related to the degree of cognitive impairment.2 Among patients with dementia of presumed vascular origin, the extent of WMLs may also be correlated with the severity of dementia.1 However, it remains controversial whether the extent of WMLs per se is an independent predictor of cognitive dysfunction and, if so, what type of cognitive function is influenced by the extent of WMLs.
White matter damage may play a role in the pathogenesis of dementia. However, the relationship between the extent of WMLs and global cognitive function is not simple. Not all subjects with extensive WMLs have dementia,3 indicating the contribution of additional factors that determine global cognitive function in subjects with extensive WMLs. If the extent of the WMLs on T2-weighted MRI is of the same degree, the difference of pathologies in the WMLs may determine the severity of white matter damage.4 When located in the deep and subcortical white matter, WMLs may mostly reflect ischemic damage and may correspond to not only focal rarefaction of myelin but also loss of fibers and sometimes lacunar infarctions, according to the histopathologic characteristics.5 The severity of axonal disruption in the WMLs may be important as a determinant for the degree of cognitive impairment. The extent of WMLs, however, cannot differentiate white matter damage with axonal loss from that without axonal loss. A previous study showed that in patients with lacunar infarcts and extensive WMLs of similar degree, the severity of global cognitive impairment was varied but was correlated with the severity of atrophy of the corpus callosum.6 The corpus callosum is composed of interhemispheric fibers traversing the subcortical white matter. In patients with WMLs, callosal atrophy may result from axonal disruption due to white matter damage. Thus, the WMLs with callosal atrophy may indicate a more severe form of white matter damage with axonal loss, whereas the WMLs without callosal atrophy may correspond to pathologies without axonal loss, including demyelination. Callosal atrophy may parallel the total loss of fibers in the white matter because the severity of ischemic damage of a nerve fiber in the white matter may not be affected by the direction of the fiber.7 Thus, if global cognitive impairment is related to the severe damage of the white matter in subjects with WMLs, callosal atrophy may be an important indicator of global cognitive decline.6 8 Until now, few studies have investigated the relation between callosal size per se and cognitive function in patients with WMLs.
Although the extent of WMLs may not be an independent predictor of global cognitive dysfunction, it may be correlated with specific cognitive deficits, particularly those related to impairment of frontal lobe functions. The severity of axonal disruption in the WMLs may determine the type of cognitive deficit caused by the WMLs. Demyelination may reduce the speed of neuronal connectivity, while axonal disruption would cause loss of the connectivity. In multiple sclerosis, which mainly impairs the white matter by demyelination, WMLs are associated with cognitive slowing.9 Thus, mild WMLs without axonal damage may influence specific cognitive functions, such as those involving the speed of mental processing and attentional abilities (subcortical–frontal lobe functions), without affecting performance on neuropsychological tests that focus mainly on cortical functions such as language and visuospatial abilities.10 11 Then, the extent of WMLs may be correlated with the severity of frontal lobe dysfunctions independent of the degree of callosal atrophy. However, few studies have investigated the relation between the extent of WMLs and frontal lobe functions after controlling for callosal size. In addition, it is unclear whether WMLs in the frontal region, with or without anterior callosal atrophy, can by themselves account for the frontal lobe dysfunction.
In this study we analyzed the relation among regional WMLs, regional size of the corpus callosum, and cognitive functions in a patient population with variable degrees of the severity of WMLs by using multivariate analysis. We adopted 2 cognitive measures: the Mini-Mental State Examination (MMSE)12 and the verbal fluency (VF) task. The MMSE is a nontimed task that is accepted as a measure of global cognitive function, while the VF task is a timed task that is relatively sensitive to frontal lobe dysfunction. The purpose of this study was to determine whether callosal atrophy is specifically correlated with impairment of MMSE (global cognitive function) and whether the extent of WMLs is specifically correlated with impairment of the VF tasks (frontal lobe function).
Subjects and Methods
We studied 62 consecutive right-handed outpatients, aged 49 to 86 (mean±SD, 68±8) years. All subjects were prospectively selected from patients who underwent MRI to diagnose or rule out cerebrovascular disease because of neurological symptoms. Inclusion criteria were as follows: (1) patients with a history of lacunar stroke, a clinical presentation consistent with one of the classic lacunar syndromes described by Fisher,12A and MRI evidence of lacunar infarcts that appeared responsible for symptoms; or (2) patients who underwent MRI because of headache or dizziness and showed normal neurological findings and no specific neurological diseases other than tension-type headache, irrespective of MRI evidence of lacunar infarcts or any degree of WMLs. Exclusion criteria were as follows: (3) cortical infarct on MRI; (4) strategically located lacunar infarcts causing dementia: infarcts in the genu of the internal capsule, thalamus, and caudate nucleus; (5) significant stenosis of the cervical or intracranial arteries on MR angiography; and (6) complications of other neurological or psychiatric disorders, including alcohol abuse and depression.
Twenty-eight patients had lacunar infarcts on MRI. A lacunar infarct was identified as an increased signal intensity on both proton-weighted and T2-weighted images with decreased signal intensity on T1-weighted images. No patients showed lesions with a diameter of >1.5 cm. The group included 26 patients with a symptomatic lacunar stroke and 2 with asymptomatic lacunar infarcts in the putamen. Nineteen men and 9 women, aged 52 to 81 (mean±SD, 69±6) years, were included. Six patients had a history of hypertension, and 3 others had diabetes mellitus. None had ischemic heart disease. The 34 patients without lacunar infarcts included 13 men and 21 women, aged 49 to 86 (mean±SD, 68±9) years. Five of them had hypertension, 4 others had diabetes mellitus, and 2 others had ischemic heart disease. No patient satisfied the criteria for dementia according to the Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition.13 All had normal laboratory test results for syphilis serology, vitamin B12 levels, and thyroid hormone levels. In all patients with a symptomatic lacunar stroke, the MRI and neuropsychological evaluations were performed at least 2 months after the ischemic event.
Cognitive function was screened with the MMSE12 and the VF task. Verbal fluency was assessed as phonological word fluency, in which a subject was asked to produce within 1 minute as many words as possible that begin with a specified letter (the Japanese letter “ka”). MMSE is a global cognitive test developed for use in dementia screening. This test may reflect mainly memory and posterior cortical function and may be less sensitive to frontal-executive cognitive deficits, while the VF task is a timed task that can monitor frontal lobe function.
Magnetic Resonance Imaging
MRI was performed with a Signa unit (General Electric) operating at a field strength of 1.5 T. T1-weighted sagittal and axial images were obtained with the use of a spin-echo pulse sequence (repetition time, 400 ms; echo time, 15 ms). Axial proton-weighted and T2-weighted images were also obtained with spin-echo pulse sequences. Repetition time for both images was 3000 ms. Echo time for proton-weighted images was 34 ms and for T2-weighted images was 100 ms. The slice thickness was 3 mm for sagittal images and 5 mm for axial images. Sections were contiguous in the sagittal plane and had an intersection gap of 1.8 mm in the axial plane.
White Matter High-Intensity Lesions
To grade WMLs, axial T2-weighted images were visually evaluated, separately for each hemisphere. The presence, location, and degree of WMLs were assessed according to a recently described scale.14 In brief, the scale required a separate evaluation of the anterior and posterior regions of each hemisphere on 3 MRI slices: 1 through the choroid plexus of the posterior horns, 1 through the cella media, and 1 through the centrum semiovale. The first 2 slices were used to evaluate the anterior region: the region around the anterior horn of the lateral ventricles; and the last 2 slices were used to evaluate the posterior region: the white matter around the posterior part of the body of the lateral ventricles and the posterior part of the centrum semiovale.
WMLs were identified as increased signal-intensity areas on T2-weighted images. Most WMLs were visualized as high-intensity areas on proton-weighted images and as normointense areas or poorly delineated hypointense areas on T1-weighted images. Focal lesions were judged present when the diameter was ≥2 mm. The severity of WMLs on T2-weighted images, including areas of infarct on T1-weighted images, was graded on a 2-point scale in each of the 2 regions on the 3 slices according to their number and configuration. A score of 0 denoted no lesion or only a single lesion. A score of 1 denoted multiple focal lesions; a score of 2 denoted multiple confluent lesions scattered throughout the white matter. WML scores in the anterior or posterior region were calculated by summing the scores in each region in both hemispheres: maximum, 8, and minimum, 0.
The grading of WMLs was performed by 1 investigator who was blinded to the clinical status of the patients. Before this study, the observer reliability of the grading of the severity of WMLs with scores of 0 to 16 was evaluated on the basis of blind rereview of 50 MRI scans in patients with strokes (100 hemispheres). There were high intraobserver and interobserver reliabilities for the grading of white matter lesions. The intraclass correlation coefficient for intraobserver reliability was 0.98 and for interobserver reliability was 0.94.
The extent of atrophy of the corpus callosum on midsagittal images was measured by using a computer-assisted image analyzer (FDM98-1; Photron) and a personal computer (PC-9801; Nihon Electric Co), as previously described.15 16 In brief, each MRI scan (1 scan for each subject) was recorded with a video camera and digitized with the use of the image analyzer utilizing a 256×256 data matrix and a 64-step gray scale. For the tracing technique, an outline of each structure was done manually with a mouse cursor. The number of pixels that had signal intensities that corresponded to the predetermined level that was set for the region of interest was then counted. The level was automatically determined as the range from the maximum pixel value to the mean value of the maximum and minimum values, where the maximum value was the maximum pixel value of the corpus callosum and the minimum value was that of the background (ie, the cerebrospinal fluid space). We divided each corpus callosum into 2 parts by drawing 2 lines at the anterior border of the rostrum and the caudal end of the splenium at right angles to a tangent that connects the most inferior points of the splenium and the rostrum. The area between these 2 lines was then divided into halves, with 1 additional perpendicular line used to produce a total of 2 regions, ie, the anterior and posterior portions. There is some evidence that the anterior and posterior halves of the corpus callosum connect with the respective cortical region, including the frontal and parieto-temporo-occipital cortex.17 The regional areas of the corpus callosum were measured. To control for variation in the skull size, we measured the areas of the midline internal skull surface by manually tracing the line through the inner table, foramen magnum, clivus, sellar diaphragm, and jugum sphenoidale, and we then calculated the regional callosal area/skull area ratio (a percentage).18
All measurements were performed by 1 investigator who was unaware of the clinical status of the patients. Before this study, the observer reliability of our procedure was evaluated in 20 subjects with or without neurological diseases. There was high intraobserver reliability (r=0.98, P<0.001).
We compared the clinical background between any 2 groups by using Student’s t test or the χ2 test, as appropriate. Repeated-measures ANOVA was performed for the partial area measurements of the corpus callosum to assess whether the relationship between 2 groups differed according to the region. Then we compared the callosal areas of the 2 groups by Student’s t test to detect regional differences. Statistical significance was indicated by P<0.025 (0.05/2) with the use of Bonferroni’s correction for multiple comparisons. We compared the regional WML scores of the 2 groups using the Mann-Whitney U test when the Kruskal-Wallis 1-way ANOVA revealed main group effects. Statistical significance in the Mann-Whitney U test was indicated by P<0.025 with the use of Bonferroni’s correction for multiple comparisons.
Multiple linear regression analysis was used to test the independent predictive value of WMLs and callosal atrophy with respect to cognitive performance. We applied this analysis to the scores of the MMSE or VF task as the dependent variable and patient age, sex, educational level, presence of other medical illness (hypertension, diabetes mellitus, and ischemic heart disease), presence of lacunar infarct, regional size of the corpus callosum, and regional WML scores as the independent variables. Statistical significance was indicated by P<0.05.
Table 1⇓ shows the clinical characteristics of patients with and without lacunar infarct. The distribution of demographic factors, except for sex, was not significantly different between the 2 groups, but a significant decrease in the scores of the MMSE and VF tasks and a significant increase in the WML score were found in patients with lacunar infarct. The callosal areas in patients with lacunar infarct had a tendency to be smaller than those in patients without, but this did not reach statistical significance.
In univariate analysis for all patients, the MMSE scores were significantly correlated with the anterior-half callosal/skull area ratio, while the scores of the VF task were significantly correlated with both anterior and posterior WML scores (Table 2⇓). No significant relation was found between the total WML score and the total callosal area/skull area ratio (r=−0.12, P=0.35). Neither age nor educational level was significantly correlated with the scores of the MMSE or VF task. The scores of neither the MMSE nor the VF task were significantly different between patients of the 2 sexes or between patients with and without hypertension, diabetes mellitus, or ischemic heart disease.
When patient age, sex, level of education, presence of lacunar infarct, presence of other medical illness (hypertension, diabetes mellitus, and ischemic heart disease), anterior and posterior WML scores, and the anterior- and posterior-half callosal area/skull area ratios were entered into a stepwise multiple linear regression analysis, it produced a model including the anterior-half callosal area/skull area ratio and the patient age with a correlation coefficient of 0.44 for the MMSE score: MMSE score=1.354(anterior-half callosal area/skull area ratio)− 0.079(age)+29.4 (P<0.002) (Table 3⇓, model 1). In this model, the anterior-half callosal area/skull area ratio accounted for 11.8% of the variance of the MMSE score, whereas age accounted for 7.5% of the variance. The other variables did not significantly contribute to the magnitude of the correlation. After we controlled for the effects of age, sex, level of education, presence of other medical illness, presence of lacunar infarct, anterior and posterior WML scores, and the posterior-half callosal area/skull area ratio, the anterior-half callosal area/skull area ratio was also a significant independent predictor of the MMSE score (Table 3⇓, model 2).
When patient age, sex, level of education, presence of lacunar infarct, presence of other medical illness, anterior and posterior WML scores, and anterior and posterior-half callosal area/skull area ratios were entered into a stepwise multiple linear regression analysis, it produced a model including the anterior WML score with a correlation coefficient of 0.354 for the score of the VF task: VF task score=−0.523(anterior WML score)+8.385 (P<0.005). In this model, the anterior WML score accounted for 12.5% of the variance of the score of the VF task. The other variables did not significantly contribute to the magnitude of the correlation. After we controlled for the effects of age, sex, level of education, presence of other medical illness, presence of lacunar infarct, and anterior and posterior callosal area/skull area ratios, the anterior WML score was a significant independent predictor of the score of the VF task (Table 4⇓). However, the effect of the anterior WML score on the score of the VF task was not independent of the effect of the posterior WML score because a strong correlation was found between the anterior and posterior WML scores (r=0.87, P<0.001).
This study showed that when patients with a variety of WMLs and no cortical lesions were assessed with the MMSE and VF tasks, callosal atrophy was specifically correlated with poor performance in the MMSE, whereas the extent of WMLs was specifically correlated with poor performance in the VF task. Multiple regression analysis revealed that among patient age, sex, educational level, presence of other medical illness, presence of lacunar infarct, midsagittal corpus callosum area, and WML score, only anterior-half callosal area/skull area ratio and age were significant independent predictors of the MMSE score. The effect of the anterior callosal area was independent from that of the posterior callosal area. The WML score was not a significant predictor of the MMSE score. On the other hand, the WML score, especially in the anterior region, was an independent predictor of the score of the VF task. The effect of the WML score in the anterior region was not independent of that in the posterior region, because a strong correlation was found between the WML score in the anterior region and that in the posterior region. The callosal size was not a significant predictor of the score of the VF task.
Our patients included patients with symptomatic lacunar infarct and stroke-free patients with headache or dizziness. No report has shown a difference between the nature of WMLs in patients with and without lacunar infarction. Because WMLs are closely associated with lacunar infarct as a subtype of stroke,19 lacunar infarcts and WMLs may be 2 types of expression of arteriolosclerosis.20 In this study patients with lacunar infarct tended to have high WML scores, small callosal size, and low scores on the MMSE and VF task compared with patients without lacunar infarct. Thus, in addition to the presence of lacunar infarct, both WMLs and callosal atrophy may be related to cognitive decline.
MRI measures of brain atrophy may be as important as or more important than WMLs for predicting cognitive deterioration. Thus, to assess whether the correlation of the extent of WMLs with cognitive functions is substantial, the relationship between the extent of WMLs and cognitive functions must be analyzed after controlling for measures of brain atrophy.1 This study showed that the extent of WMLs was a predictor for scores of the VF task, independent of callosal atrophy, but not for scores of the MMSE. A previous study including more extensive neuropsychological testing also showed that WMLs were specifically correlated with scores of tests involving subcortical–frontal lobe functions, including the VF task, independent of ventricular enlargement, while ventricular enlargement was correlated with scores of tests that focus mainly on cortical functions, including the MMSE.11 Thus, the extent of WMLs may be an independent predictor of specific cognitive deficits, particularly those related to impairment of frontal lobe functions, but not of global cognitive disturbance.21 However, a recent large population-based study showed that the presence of periventricular, not subcortical, WMLs was an independent predictor of the MMSE score after adjustment for the degree of cortical atrophy and ventricular enlargement.22 This result may agree with the association of callosal atrophy with the MMSE score in the present study because the commisural tract as well as the long association tract may be disrupted mainly by periventricular WMLs. Further investigations are required to determine the relationship among separate regions of WMLs, callosal atrophy, and global cognitive disturbance.
The degree of the severity of WMLs, as indicated by the extent of callosal atrophy, may be related to global cognitive impairment. Callosal atrophy has been shown in vascular dementia.6 8 23 24 A previous study in patients with lacunar infarcts and extensive WMLs of similar degrees suggested that callosal atrophy may be related to the process that determines the severity of global cognitive impairment.6 This notion was supported by the findings in patients with a wide range of degrees of extent of WMLs in this study: callosal atrophy was a predictor of global cognitive impairment independent of the extent of WMLs.
Predominant damage in the frontal white matter may cause characteristic impairment of frontal lobe functions in patients with WMLs. In this study anterior WML scores were related to the scores of the VF task, but the control for posterior WML scores resulted in this significant relation becoming nonsignificant. Thus, the attempt to attribute frontal lobe dysfunction to the WMLs in the anterior region was difficult, as shown in previous studies.10 25 In contrast, the anterior-half callosal/skull area ratio was related to the MMSE score, which was independent of the posterior-half callosal/skull area ratio. Therefore, global cognitive decline in the patients studied here may be attributable to the severe damage in the frontal white matter, which may cause disconnection between the frontal cortex and other regions. Pathological studies also suggested the importance of multiple small infarctions in the frontal white matter for the development of vascular dementia.26 27
This study has some limitations. We included stroke-free patients with headache or dizziness rather than healthy community volunteers as a control group. We did not think that this was a major problem since they were otherwise neurologically normal. Alcohol abuse, depression, duration of hypertension, and cortical brain atrophy can influence cognitive status. Inclusion of these factors as continuous variables might have affected the relation of WMLs or callosal atrophy to cognitive functions. Some factors other than white matter damage may contribute to callosal atrophy in subjects with WMLs. The size of the corpus callosum varies in the normal population. Decreased callosal size without WMLs may be caused by aging or other unknown factors. These factors may have led to the significant but weak correlation between callosal size and MMSE score in the patients studied here, in whom pathological decreases of callosal size and MMSE score were probably small. This interpretation is supported by the finding that callosal atrophy was strongly correlated with global cognitive impairment in a patient sample with extensive WMLs and a wide range of global cognitive function.6 The correlation between WML score and VF task score was also weak in this study, although a wide range of both of these scores was found in our patient sample. Thus, the effect of WMLs per se on specific cognitive functions may be significant but subtle.28 However, this study included only a limited neuropsychological evaluation, used a simple ordinal scale, and has the limitations of a cross-sectional study. Further longitudinal studies using more sophisticated neuropsychological measures of executive functions and quantitative measurement of WMLs volume are required to determine whether the effect of WMLs per se on specific cognitive functions is substantial and independent of callosal atrophy.
In conclusion, callosal atrophy is an important predictor of global cognitive impairment in patients with WMLs on T2-weighted MR images, whereas the extent of WMLs per se may be related to impairment of frontal lobe function independent of callosal atrophy.
Reprint requests to Dr Hiroshi Yamauchi, Research Institute, Shiga Medical Center, 5-4-30 Moriyama, Moriyama-city, Shiga 524-8524, Japan.
- Received February 14, 2000.
- Revision received April 3, 2000.
- Accepted April 3, 2000.
- Copyright © 2000 by American Heart Association
Pantoni L, Garcia JH. The significance of cerebral white matter abnormalities 100 years after Binswanger’s report: a review. Stroke. 1995;26:1293–1301.
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: the Cardiovascular Health Study. Stroke. 1996;27:1274–1282.
Fazekas F, Kleinert R, Offenbacher H, Schmidt R, Kleinert G, Payer F, Radner H, Lechner H. Pathologic correlates of incidental MRI white matter signal hyperintensities. Neurology. 1993;43:1683–1689.
Yamauchi H, Fukuyama H, Ogawa M, Ouchi Y, Kimura J. Callosal atrophy in patients with lacunar infarction and extensive leukoaraiosis: an indicator of cognitive impairment. Stroke. 1994;25:1788–1793.
Tanaka Y, Tanaka O, Mizuno Y, Yoshida M. A radiologic study of dynamic process in lacunar dementia. Stroke. 1989;20:1488–1493.
Breteler MM, van Amerongen NM, van Swieten JC, Claus JJ, Grobbee DE, van Gijn J, Hofman A, van Harskamp F. Cognitive correlates of ventricular enlargement and cerebral white matter lesions on magnetic resonance imaging: the Rotterdam Study. Stroke. 1994;25:1109–1115.
Fisher CM. Lacunar strokes and infarcts: a review. Neurology.. 1982;32:871–876.
American Psychiatric Association. Organic mental syndromes and disorders. In: American Psychiatric Association, ed. Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition. Washington, DC: American Psychiatric Association; 1987:97–163.
van Swieten JC, Hijdra A, Koudstaal PJ, van Gijn J. Grading white matter lesions on CT and MRI: a simple scale. J Neurol Neurosurg Psychiatry. 1990;53:1080–1083.
Innocenti GM. General organization of callosal connections in the cerebral cortex. In: Jones EG, Peters A, eds. Cerebral Cortex. 5th ed. New York, NY: Plenum Press; 1986:291–353.
Laissy JP, Patrux B, Duchateau C, Hannequin D, Hugonet P, Ait-Yahia H, Thiebot J. Midsagittal MR measurement of the corpus callosum in healthy subjects and diseased patients: a prospective survey. AJNR Am J Neuroradiol. 1993;14:145–154.
Hijdra A, Verbeeten BJ, Verhulst J. Relation of leukoaraiosis to lesion type in stroke patients. Stroke. 1990;21:890–894.
van Swieten JC, van den Hout JH, van Ketel BA, Hijdra A, Wokke JH, van Gijn J. Periventricular lesions in the white matter on magnetic resonance imaging in the elderly: a morphometric correlation with arteriolosclerosis and dilated perivascular spaces. Brain. 1991;114:761–774.
Kameyama M. Vascular lesions in the frontal association field and dementia. Clin Psychiatry (Tokyo). 1973;15:357–366.
Ishii N, Nishihara Y, Imamura T. Why do frontal lobe symptoms predominate in vascular dementia with lacunes? Neurology. 1986;36:340–345.
Schmidt R, Fazekas F, Offenbacher H, Dusek T, Zach E, Reinhart B, Grieshofer P, Freidl W, Eber B, Schumacher M, Koch M, Lechner H. Neuropsychologic correlates of MRI white matter hyperintensities: a study of 150 normal volunteers. Neurology. 1993;43:2490–2494.