You are on page 1of 7

Topical Review

Section Editor: Marc Fisher, MD

Progression of Leukoaraiosis and Cognition


Reinhold Schmidt, MD; Katja Petrovic, PhD; Stefan Ropele, PhD; Christian Enzinger, MD;
Franz Fazekas, MD

Background and Purpose—Leukoaraiosis is used interchangeably with the term white matter lesions on MRI and seen to
some degree in more than half of the routine scans in older persons. Clinicians often struggle to explain the implications
of these findings to their patients. Recent data on the progression rate of ischemic white matter damage and its cognitive
consequences may help in patient counseling and have implications on treatment trials in vascular cognitive impairment.
Summary of Review—Leukoaraiosis progresses over time. Its extent at baseline is an important predictor for the
subsequent rate of lesion progression. Subjects with punctate abnormalities on MRI have a low tendency for
progression, individuals with early confluent and confluent changes tend to progress rapidly. Differences in
measurement methods and cohort composition make it difficult to compare progression rates reported by different
studies. Nevertheless, in community-dwelling cohorts, white matter lesions volume increased by as much as one quarter
per year in subjects with confluent abnormalities at baseline. Progression of leukoaraiosis relates to cognitive decline,
but this association is complex and modulated by other morphological factors like brain atrophy.
Conclusions—Evidence for rapid progression of widespread leukoaraiosis and the associated cognitive decline in domains
particularly affected by cerebral small vessel disease has set the stage for exploratory clinical trials in vascular cognitive
impairment using white matter lesions progression as a surrogate marker. (Stroke. 2007;38:2619-2625.)
Key Words: cognition 䡲 leukoaraiosis 䡲 MRI
Downloaded from http://ahajournals.org by on March 8, 2019

O ne of the challenges when evaluating patients with


cognitive impairment is the interpretation of leukoara-
iosis shown by neuroimaging. Stenosis or occlusion of small
with small vessel disease which completely eludes detection
by MRI are microinfarcts in the gray matter.4
The role of white matter damage as a contributor to
cerebral vessels with sudden or chronic ischemia leading to vascular dementia is still controversial. This is also evident
incomplete white matter infarction is considered to play a from the diverging conclusions of the 2 most recent sets of
central role in the pathogenesis of leukoaraiosis. Disturbed diagnostic criteria on vascular dementia.5,6 Both the Califor-
autoregulation in diseased small vessels and subsequent nia5 and the National Institute of Neurological Disorders and
fluctuations in blood flow in response to changes of systemic Stroke (NINDS)-Association Internationale pour la Recher-
blood pressure have also been implicated with complemen- che et l’Enseignement en Neurosciences (AIREN) criteria6
tary factors such as damage to the blood-brain barrier and consider leukoaraiosis. However, in the former,5 only re-
chronic leakage of fluid to the white matter.1 So far, it is search status is attributed to the impact of leukoaraiosis,
undetermined how well small vessel pathology maps onto whereas the latter6 include it among imaging findings diag-
radiological markers such as white matter hyperintensities nostic for vascular dementia if more than a quarter of the total
seen in standard MR sequences. Yet, conventional MRI is white matter is affected.
unlikely to identify the full spectrum of small vessel disease– Given the high prevalence of leukoaraiosis in the general
related cerebral damage. Diffusion tensor and magnetization elderly population,7 it is particularly difficult to ascribe a
transfer–imaging studies indicate that small vessel disease lowest threshold at which these lesions contribute to cogni-
also causes loss of white matter tract integrity in the so-called tive impairment.8 Early cross-sectional studies reported con-
normal-appearing white matter, a finding that has been flicting results probably attributable to low statistical power
related to cognitive decline in elderly individuals.2,3 Another and the frequent use of insensitive cognitive measures.9
pathological correlate of cognitive impairment in subjects Meanwhile, large population-based studies clearly support an

Received March 23, 2007; final revision received May 22, 2007; accepted June 7, 2007.
From the Department of Neurology (R.S., K.P., S.R., C.E., F.F.) and, the Division of Neuroradiology (R.S., C.E.), Department of Radiology, Medical
University of Graz, Austria.
Correspondence to Reinhold Schmidt, MD, Professor of Neurology, Department of Neurology, Medical University of Graz, Auenbruggerplatz 22,
8036- Graz, Austria. E-mail reinhold.schmidt@meduni-graz.at
© 2007 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.107.489112

2619
2620 Stroke September 2007

association between leukoaraiosis and cognitive impair- very time-consuming and labor-intensive. Current semiauto-
ment.10 –12 The multinational Leukoaraiosis and Disability mated methods are mostly based on tissue segmentation and
Study (LADIS) also showed a strong association between still require considerable operator input. Despite training and
white matter lesions (WML) and cognitive dysfunction.13 guidelines for analysis, some differences between operators
Cross-sectional data cannot prove causality and thus can- are unavoidable. Therefore, most studies have used a side-
not determine whether the relationship between leukoaraiosis by-side comparison of lesion-load assessment. This may
and cognitive dysfunction stems from the ischemic white cause some overestimation of progression rate but is an
matter damage per se or from coexistent and secondary efficient way of reducing noise and artifactual influences on
phenomena also occurring with aging. This prompted a shift lesion change. Conceptually, techniques of coregistration and
from cross-sectional to longitudinal studies to establish a image subtraction should also enhance the sensitivity and
temporal relationship between change in lesion extent and reliability of detecting lesional changes, but such advantages
change in clinical presentation. are frequently offset by a parallel reduction in lesion contrast
Recent reports consistently showed that leukoaraiosis pro- from image reformatting. Moreover, subtraction techniques
gression over time is substantial, particularly in patients with rely on identical imaging protocols, a prerequisite that is
specific phenotypes.14 These findings led the European Task difficult to fulfill in long-term multicenter studies.
Force on Age-Related White Matter Changes to propose Neuropsychological test batteries assessing cognitive ef-
WML progression as a potential surrogate marker in thera- fects of leukoaraiosis progression need to be tailored toward
peutic trials of cerebral small vessel disease.15 Nonetheless, the pattern of dysfunction specific for vascular cognitive
longitudinal studies on white matter damage and its clinical impairment and distinct from that commonly associated with
consequences still pose many methodological challenges, eg, Alzheimer disease.9 A metaanalysis quantified the cognitive
how to best determine lesion progression and what other sequelae of leukoaraiosis in nondemented subjects and found
morphological variables to consider. No widely accepted processing speed, immediate and delayed memory, executive
cognitive tests for the assessment of neuropsychological functions, and indices of global cognitive functioning to be
consequences of leukoaraiosis exist. Another problem are especially affected.24 It was only recently that the NINDS-
high attrition rates in long-term studies on elderly individuals, Canadian Stroke Network recommended an optimized test
especially in protocols with repeat MRIs. protocol covering 4 distinctive cognitive domains including
executive control and activation state, language, visuospatial
Methods to Assess Leukoaraiosis Progression abilities, and memory.25
and Related Cognitive Change
Downloaded from http://ahajournals.org by on March 8, 2019

Semiquantitative visual assessment of leukoaraiosis has proven Magnitude of Progression of Leukoaraiosis


efficacious and several rating scales with different grades of Meanwhile, there is convincing evidence for leukoaraiosis
refinement have been developed and validated to determine progression from numerous studies using either visual
leukoaraiosis severity.16 These scales demonstrated good rating or semiautomatic measurement of lesion volume in
interrater agreement and correlated sufficiently with the population-based, community-dwelling and hospital-based
volume of white matter damage derived from fully quantita- studies.19,21,26 –34Even CT scanning can demonstrate newly
tive measurements.17 Unfortunately, these scales fail to pick developing leukoaraiosis as shown in a selected cohort from
up progression.17 We adapted the Fazekas scale18 and cate- the North American Symptomatic Carotid Endarterectomy
gorized leukoaraiosis progression as minor with 1 to 4 new Trial. From 596 patients without leukoaraiosis at entry, 18%
punctate lesions and as marked with a difference of ⬎4 developed restricted abnormalities and 3% widespread
lesions or a transition to early confluent or confluent white changes during a mean follow-up of 6.1 years.35 Similarly, all
matter abnormalities.19 However, by focusing on new lesions MRI studies so far also yielded lesion progression over time
and transition to higher grades alone, we probably underes- (Table 1). This table shows great variation in frequencies of
timated the rate of progression because WML progression progression and volume increases. A direct comparison
very commonly means enlargement of already existing between studies is difficult because of differences in study
changes rather than newly occurring abnormalities. groups and analysis techniques. Importantly, all investiga-
The Rotterdam Progression Scale considered both new tions agree that the extent of lesions at baseline is a major
focal lesion occurrence and lesion enlargement, and found predictor of subsequent progression.26,27,32,33 Age and con-
good agreement with a volumetric method after the scale had ventional risk factors for stroke appear to have relatively little
been adapted.20 The Cardiovascular Health study investiga- influence, at least over the observational periods reported so
tors calculated a change score by subtracting grades of far.14 In the Austrian Stroke Prevention Study,32 there was
abnormality on their 10-point scoring system at follow-up almost no increase in lesion volume in subjects with punctate
and observed modest intrareader reliability for change of 1 or abnormalities, whereas participants with a baseline finding of
more grades but only fair interrater agreement.21 Another early confluent or confluent changes showed a remarkably
group compared visual rating of lesion progression based on rapid increase in lesion volume. After 3 and 6 years of
the Scheltens scale22 with volumetric measurements and follow-up, subjects with confluent white matter abnormalities
concluded that such a quantitative assessment of leukoaraio- at baseline experienced a 5 cm3 and 13 cm3 increase in
sis offered a more reliable, sensitive, and objective instrument lesional volume, respectively. The Figure shows examples of
than visual rating scales in longitudinal studies.23 Most differential lesion progression depending on baseline lesion
researchers consider volumetric assessments superior but grade.
Schmidt et al Progression of Leukoaraiosis and Cognition 2621

TABLE 1. Leukoaraiosis Progression in MRI Studies in the Elderly


Follow-Up Progression
Study Sample Subject No. in Years Measurement Frequency Volume Increase
29
Wahlund Normals 13 5 Scheltens Scale 92% 4.6 cm3
30
Veldink Normals & cognitively 14 2 Scheltens Scale 57% 䡠䡠䡠
impaired
Schmidt19 Community-dwelling 273 3 Modified Fazekas 17.9%
Scale
Whitman31 Normals 70 4 Grid method 䡠䡠䡠 1.1 cm3
32
Schmidt Community-dwelling 296 6 Semiautomated 17.2% 1.4 cm3
volume measurement
Taylor26 Community-dwelling 117 2 Semiautomated 䡠䡠䡠 1.4 cm3
volume measurement
Van Dijk33 Population-based 636 3.3 Modified Rotterdam 27% periventricular, 䡠䡠䡠
Scale 32% subcortical
Longstreth21 Population-based 1919 5 Cardiovascular Health 28% 䡠䡠䡠
Study Scoring
Ten Dam34 Trial cohort 535 2.8 Automated volume 䡠䡠䡠 1.0 cm3
measurement
Garde28 Octogenarians 26 3.8 Semiautomated 䡠䡠䡠 2.4 cm3
volume measurement
Sachdev27 Community-dwelling 51 3 Automated volume 䡠䡠䡠 6.5 cm3
measurement

Almost two thirds with early confluent and all subjects and confluent white matter abnormalities are ischemic, pro-
with confluent lesions demonstrated progression beyond gressive, and thus malignant. Progression of white matter
measurement error over 6 years. This was seen in none of the abnormalities was also seen in subjects with dementia of
subjects with a normal baseline MRI and only 14% of those various etiologies, but these small studies did not show an
Downloaded from http://ahajournals.org by on March 8, 2019

with punctate foci. These data together with findings from effect of the dementing illness per se. Progression rates were
pathological correlations36,37 indicate that punctate WMLs are again influenced primarily by severity of lesions at baseline
probably of mixed origin and benign, whereas early confluent rather than by the type of dementia or cognitive decline.38 A
recent investigation showed that leukoaraiosis progression
occurs throughout the brain but with some regional varia-
tion.27 Frontal lesions progressed the most, occipital the least.
Lesions in the deep white matter progressed more than those
in the periventricular white matter, a similar finding was
reported by the PROspective Study of Pravastatin in the
Elderly at Risk (PROSPER) study, but only in women.39

Cognitive Consequences of Progression


of Leukoaraiosis
Reports on the cognitive consequences of lesion progression
have been published only recently (Table 2).21,28,40,41 They
consistently demonstrated an association between progres-
sion of leukoaraiosis and cognitive decline. In a Danish
cohort of people born in 1914 an increase in WML volume
correlated with a declining verbal IQ. The Cardiovascular
Health Study21 reported WML progression by at least 1 grade
in 28% of study participants. They experienced greater
decline on the modified Mini-Mental State Examination
Baseline (upper row) and 6-year follow-up (lower row) in 3 study and the Digit-Symbol Substitution Test than those without
participants of the Austrian Stroke Prevention Study. Subject A
progression. Similar data were seen in the MRI-substudy
had punctate lesions at baseline and no change after 6 years.
Subjects B and C show early confluent WMLs and confluent PROSPER41 in which periventricular rather than deep
abnormalities at baseline, respectively. Typically, these subjects white matter lesion progression affected cognitive function-
experienced substantial leukoaraiosis progression after 6 years. ing. The Austrian Stroke Prevention Study40 on 329 elderly
As shown in B and C, the predominant pattern of progression is
lesion growth at the sites of preexisting lesions rather than community-dwelling volunteers demonstrated a complex in-
occurrence of new abnormalities. teraction of progression of leukoaraiosis, brain atrophy and
2622 Stroke September 2007

TABLE 2. Decline of Cognitive Abilities in Longitudinal MRI Studies in the Elderly


Subject Follow-Up
Study Sample No. in Years Cognitive Tests Cognition-Associated MRI-Results Cognitive Change
21
Longstreth Population- 1919 5 Modified MMSE, DSST Worsening white matter grade Greater decline
based
Garde28 Octogenerians 26 3.8 Wechsler Adult Intelligence a) Baseline WML volume, b) WML a) Performance
Scale volume increase decline, b) Verbal
Decline
Van den Heuvel41 Trial cohort 554 3 Picture Word Learning, Symbol a) Higher periventricular WML Reduced mental
Digit Modalities, Stroop volume at baseline, b) Larger processing speed
progression of periventricular WMLs
Schmidt40 Community- 243 6 Learning and Memory, Increasing WML volume, loss of Decline on memory,
dwelling Wisconsin Card Sorting, Trail brain volume conceptual reasoning
Making B, Digit Span, Complex and visuo-practical
Reaction Time, Purdue Pegboard skills
MMSE indicates Mini-Mental State Examination; DSST, Digit Symbol Substitution Test.

cognitive functioning. Increasing volume of leukoaraiosis Current Studies Probably Underestimate the
correlated significantly with loss of brain volume and perfor- Progression of Leukoaraiosis and Related
mance decline in tests of memory, conceptualization, and Cognitive Dysfunction
visuopractical skills. However, the associations between Substantial participant attrition was seen in all longitudinal
changes in WML load and cognitive functioning were no studies assessing leukoaraiosis progression and requesting
longer significant when the change in brain volume was study participants to undergo repeat MRIs and complex
added to the models suggesting that cognitive decline related investigational protocols. In the Rotterdam Scan Study,33
directly to loss of brain substance with progression of lesion 38% of participants who underwent brain MRI at baseline did
burden. One explanation for the mediating effect of brain not undergo the 5-year follow-up scan. In the Austrian Stroke
atrophy in the relation between WML progression and cog- Prevention Study, 43% and 52% of participants were lost to
nitive decline is concomitant Alzheimer disease.42 However, the 3-year and 6-year follow-up scanning, respectively.40 A
Downloaded from http://ahajournals.org by on March 8, 2019

in the Austrian study, the subgroup of participants with no very similar drop-out rate was seen in the Cardiovascular
brain atrophy but extensive white matter changes at baseline Health Study21 in which 42% of subjects did not undergo a
experienced a substantially greater loss of brain volume second MRI after 5 years. All 3 studies reported that patients
during subsequent years than their counterparts without white who took part in the follow-up examinations were younger,
matter foci. This supports the view that concomitant Alzhei- had less vascular risk factors and higher educational levels
mer pathology is not the sole explanation for increasing loss than the total study population. Therefore, successive evalu-
of brain parenchyma in patients with progressing leukoaraio- ations reflect the course of brain changes and the cognitive
sis. Interestingly, in the context of subcortical ischemic performance of a progressively more healthy and elite subset
disease, dementia with substantial cortical and hippocampal of the original sample. It thus appears justified to assume a
atrophy has been shown to occur in the absence of any greater rate of leukoaraiosis progression and its clinical
Alzheimer pathology.43 There are several mechanisms by consequences in those study participants who were lost to
which subcortical ischemic disease could lead to loss of brain follow-up.
volume and subsequent cognitive deficits. Among them are Also, some underestimation regarding evolving neuropsy-
chronic functional alterations related to ischemic white matter chological dysfunction may come from practice effects.
damage with subsequent structural changes of the cortex. Repeated confrontation with psychological tests is likely to
Alternatively, progressing leukoaraiosis might be paralleled improve performance and subsequently lead to underestima-
by increasing cortical microangiopathy with microscopic tion of the rate of cognitive deterioration. An analysis of
infarctions and loss of neurons invisible on conventional active community residents showed that improvement attrib-
MRI. A similar scenario has been suggested in multiple utable to retesting can be as large as the average decline
sclerosis where cognitive deterioration also relates more to between ages 49 and 70, or the average decline between ages
70 and 80.46 Such effects are of particular importance in
the magnitude of brain parenchymal loss than to the extent of
studies on elderly populations without dementia from which
WML burden.44 Gray matter pathology beyond the resolution
most current data on the natural course of leukoaraiosis were
of standard MRI sequences is considered to play an important
derived. A time interval of 6 years is needed between test
role.45 More data are needed on the mechanisms involved in
exposures to eliminate practice effects.47 In the available
the relationship between evolving leukoaraiosis and cognitive
studies, the intervals between testing were shorter, in part
deterioration. Nevertheless, based on current evidence, there
because such long periods between follow-ups pose a serious
is no doubt that, on a group level, leukoaraiosis progression is
risk of even higher drop-out rates. It is common practice to
paralleled by cognitive decline.
use parallel test forms at follow-up in order to reduce practice
Schmidt et al Progression of Leukoaraiosis and Cognition 2623

effects. These may, however, not only come from a general use leukoaraiosis progression as a surrogate marker in
familiarization with the examination procedures but also be a clinical trials been adopted in the PROSPER34 trial and the
consequence of decreased test anxiety. If this is the case, Perindopril Protection Against Recurrent Stroke Study
better performance is independent of cognitive test forms (PROGRESS).50 PROSPER did not show any effects of statin
used. It is noteworthy that the relevance of practice effects treatment on volume change of leukoaraiosis,34 but statin also
appears to be different for different cognitive domains. They had no effect on cerebral blood flow and brain volume.51 By
were described to be more pronounced for memory whereas contrast, the MRI substudy of PROGRESS50 revealed a lower
there is little, if any, effect on tests assessing speed.48 Practice incidence of WMLs under angiotensin-converting enzyme
effects may also affect randomized trials. It is important that inhibitor treatment with the treatment effect being most
study participants in the control and experimental arm expe- pronounced in the subgroup of subjects with advanced
rience identical practice levels. leukoaraiosis at study entry. The lack of an effect of statin use
on progression of morphological and functional variables
Conclusions seen in the PROSPER substudy needs to be considered with
Significant leukoaraiosis progression is seen in elderly indi- caution because patients were recruited irrespective of the
viduals who present with moderate to severe white matter presence of white matter lesions at baseline. Also, different
changes irrespective of clinical symptoms.19,21,26 –34 Leuko- from PROGRESS, stratification according to white matter
araiosis progression is accompanied by a modest decline in lesion severity was not done.34 Given the complexity of the
performance on neuropsychological tests. Future studies will associations between progression of white matter damage and
cognitive decline,40 measuring the increase of leukoaraiosis
have to better determine the magnitude of cognitive and
volume in isolation probably does not suffice to fully deter-
functional impairment associated with the full range of
mine treatment response in small vessel disease trials. It is
leukoaraiosis burden in elderly subjects. Currently, it is
equally important to assess other morphological expressions
unknown how often leukoariosis is associated with clinically
of the disease. Even though changes in brain volume cannot
relevant functional deficits. The LADIS project included
unequivocally be attributed to cerebral small vessel disease,
subjects based on their burden of leukoaraisosis irrespective
they deserve special attention as possible surrogate markers
of their clinical presentation and supports the clinical rele- in treatment trials on this disease as loss of brain parenchyma
vance of these brain abnormalities, because subjects with was found to be more closely related to cognitive decline than
severe white matter changes performed significantly worse the change in leukoaraiosis volume.40 Another focus needs to
on global cognitive measures such as the Mini-Mental State be put on white matter appearing normal on standard MRI
Examination and showed rapid functional decline as early as sequences. This will require longitudinal application of tech-
Downloaded from http://ahajournals.org by on March 8, 2019

after the first year of follow-up in comparison to those niques to quantify tissue integrity like mean diffusivity,
individuals with only mild leukoaraiosis at baseline.13,49 The fractional anisotropy or assessment of magnetization transfer
relationship between burden of leukoaraiosis and cognitive ratios.3,52 A recent longitudinal study in patients with CADA-
functioning is likely to be influenced by other morphological SIL established correlations between changes of diffusion
and clinical factors. Evolving brain atrophy mediated the tensor imaging histogram metrics and clinical measures over
association between leukoaraiosis progression and cognitive time.53
deterioration in community-dwelling older persons using Selection of appropriate patient groups will be an impor-
automated techniques to study the change of brain morphol- tant issue in trial planning. Focus on individuals with already
ogy over 6 years.40 The existing data cannot rule out that the existing substantial leukoaraiosis is mandatory because base-
progression of both leukoaraiosis and related cognitive dys- line lesion load is the most significant predictor of lesion
function is underestimated because of a selective drop-out of progression. Deterioration of neuropsychological perfor-
study participants. Other important but underappreciated mance has also been greater in patients with marked leuko-
factors are learning effects and cerebral adaptation to tissue araiosis at baseline than in those with only mild abnormali-
damage, an issue that has attracted little attention in studies ties.28 This is further supported by a recent observation of the
on the cognitive sequelae of small vessel disease so far. LADIS study.49 Individuals with severe white matter changes
Consequently, several issues need to be considered when were found to have a 2- to 3-fold increased risk of becoming
planning interventional therapeutic trials in small vessel more impaired as early as after the first year of follow-up in
disease. Clearly, leukoaraiosis progression represents one of comparison to those individuals with only mild leukoaraiosis
the most convincing and readily measurable markers of at baseline.
progressing small vessel disease.15 It has already been sug- Finally, interventional studies will have to concentrate not
gested by the European Task Force on Age-Related White only on cognitive functions but the entire spectrum of clinical
Matter Changes that trials on cerebral small vessel disease abnormalities potentially attributable to cerebral small vessel
with primary clinical end points may use lesion load progres- disease. This is the only way to attribute adequate importance
sion as a secondary outcome variable to evaluate therapeutic to an illness that is almost endemically encountered in our
effects earlier and with higher sensitivity than by clinical older populations and urgently awaits targeted therapeutic
observations alone.15 Based on data of the Austrian Stroke measures.
Prevention Study, the Task Force estimated that 195 subjects
with confluent lesions would be required per treatment arm to
Sources of Funding
The Austrian Stroke Prevention Study was supported by grants of the
demonstrate a 20% reduction in the rate of disease progres- Austrian Science Fund and the Jubiläumsfonds of the Austrian
sion over a 3-year period. Only recently has the proposal to National Bank.
2624 Stroke September 2007

Disclosures 19. Schmidt R, Fazekas F, Kapeller P, Schmidt H, Hartung H. MRI white


matter hyperintensities: three-year follow-up of the Austrian Stroke Pre-
None.
vention study. Neurology. 1999;53:132–139.
20. Prins N, van Straaten E, van Dijk E, Simoni M, van Schijndel R,
Vrooman H, Koudstaal P, Scheltens P, Breteler M, Barkhof F. Measuring
References progression of cerebral white matter lesions on MRI: visual rating and
1. Pantoni L. Pathophysiology of age-related cerebral white matter changes. volumetrics. Neurology. 2004;62:1533–1539.
Cerebrovasc Dis. 2002;13:7–10. 21. Longstreth WJ, Arnold A, Beauchamp NJ, Manolio T, Lefkowitz D,
2. Shenkin S, Bastin M, Macgillivray T, Deary I, Starr J, Rivers C, Jungreis C, Hirsch C, O’Leary D, Furberg C. Incidence, manifestations,
Wardlaw J. Cognitive correlates of cerebral white matter lesions and and predictors of worsening white matter on serial cranial magnetic
water diffusion tensor parameters in community-dwelling older resonance imaging in the elderly: The Cardiovascular Health Study.
people. Cerebrovasc Dis. 2005;20:310 –318. Stroke. 2005;36:56 – 61.
3. Fazekas F, Ropele S, Enzinger C, Gorani F, Seewann A, Petrovic K, 22. Scheltens P, Barkhof F, Leys D, Pruvo J, Vermersch P, Steinling M, Valk
Schmidt R. MTI of white matter hyperintensities. Brain. 2005;128: J. A semiquantitative rating scale for the assessment of signal hyperin-
2926 –2932. tensities on magnetic resonance imaging. J Neurol Sci. 1993;114:7–12.
4. Kovari E, Gold G, Herrmann F, Canuto A, Hof P, Bouras C, Giannakopoulos 23. Van den Heuvel D, ten Dam V, de Craen A, Admiraal-Behloul F, van Es
P. Cortical microinfarcts and demyelination affect cognition in cases at high A, Palm W, Spilt A, Bollen E, Blauw G, Launer L, Westendorp R, van
risk for dementia. Neurology. 2007;68:927–931. Buchem M; PROSPER Study Group. Measuring longitudinal white
5. Chui H, Victoroff J, Margolin D, Jagust W, Shankle R, Katzman R. matter changes: comparison of a visual rating scale with a volumetric
Criteria for the diagnosis of ischemic vascular dementia proposed by the measurement. AJNR Am J Neuroradiol. 2006;27:875– 878.
state of California Alzheimer’s disease diagnostic and treatment centres. 24. Gunning-Dixon F, Raz N. The cognitive correlates of white matter abnor-
Neurology. 1992;42:473– 480. malities in normal aging: a quantitative review. Neuropsychology. 2000;
6. Roman G, Tatemichi T, Erkinjuntti T, Cummings J, Masdeu J, Garcia J, 14:224 –232.
Amaducci L, Orgogozo J, Brun A, Hofman A, Moody D, O’Brien M, 25. Hachinski V, Iadecola C, Petersen R, Breteler M, Nyenhuis D, Black S,
Yamaguchi T, Grafman J, Drayer B, Bennet D, Fisher M, Ogata Powers W, DeCarli C, Merino J, Kalaria R, Vinters H, Holtzman D,
J, Kokmen E, Bermejo F, Wolf P, Gorelick P, Bick K, Pajeau A, Bell M, Rosenberg G, Dichgans M, Marler J, Leblanc G. National Institute of
DeCarli C, Culebras A, Korczyn A, Bogousslavsky J, Hartman A, Neurological Disorders and Stroke-Canadian Stroke Network vascular
Scheinberg P. Vascular dementia: diagnostic criteria for research studies. cognitive impairment harmonization standards. Stroke. 2006;37:
Report of the NINDS-AIREN International workshop. Neurology. 1993; 2220 –2241.
43:250 –260. 26. Taylor W, MacFall J, Provenzale J, Payne M, McQuoid D, Steffens D,
7. Van Dijk E, Prins N, Vermeer S, Koudstaal P, Breteler M. Frequency of Krishnan K. Serial MR imaging of volumes of hyperintense white matter
white matter lesions and silent lacunar infarcts. J Neurol Transm Suppl. lesions in elderly patients: correlation with vascular risk factors. Am J
2002;25:25–39. Roentgenol. 2003;181:571–576.
8. Bowler J, Steenhuis R, Hachinski V. Conceptual background to vascular 27. Sachdev P, Wen W, Chen X, Brodaty H. Progression of white matter
cognitive impairment. Alzheimer Dis Assoc Disord. 1999;13(Suppl 3): hyperintensities in elderly individuals over 3 years. Neurology. 2007;68:
S30–S37. 214 –222.
9. Desmond D. Cognition and white matter lesions. Cerebrovasc Dis. 2002; 28. Garde E, Lykke Mortensen E, Rostrup E, Paulson O. Decline in intel-
Downloaded from http://ahajournals.org by on March 8, 2019

13(Suppl 2):53–57. ligence is associated with progression in white matter hyperintensity


10. Longstreth W, Manolio T, Arnold A, Burke G, Bryan N, Jungreis C, volume. J Neurol Neurosurg Psychiatry. 2005;76:1289 –1291.
Enright P, O’Leary D, Fried L. Clinical correlates of white matter 29. Wahlund L, Almkvist O, Basin H, Julin P. MRI in successful ageing, a
findings on cranial magnetic resonance imaging of 3301 elderly people. 5-year follow-up study from eighth to ninth decade of life. Magn Res
The Cardiovascular Health Study. Stroke. 1996;27:1274 –1282. Imag. 1996;14:601– 608.
11. De Groot J, de Leeuw F, Oudkerk M, van Gijn J, Hofman A, Jolles J, 30. Veldink J, Scheltens P, Jonker C, Launer L. Progression of cerebral white
Breteler M. Cerebral white matter lesions and cognitive function: The matter hyperintensities on MRI is related to diastolic blood pressure.
Rotterdam Scan Study. Ann Neurol. 2000;47:145–151. Neurology. 1998;51:319 –320.
12. Au R, Massaro J, Wolf P, Young M, Beiser A, Seshadri S, D’Agostino R, 31. Whitman G, Tang T, Lin M, Baloh R. A prospective study of cerebral
white matter abnormalities in older people with gait dysfunction.
DeCarli C. Association of white matter hyperintensity volume with
Neurology. 2001;57:990 –994.
decreased cognitive functioning: The Framingham Heart Study. Arch
32. Schmidt R, Enzinger C, Ropele S, Schmidt H, Fazekas F. Progression of
Neurol. 2006;63:246 –250.
cerebral white matter lesions: 6-year results of the Austrian Stroke Pre-
13. Van der Flier W, van Straaten E, Barkhof F, Verdelho A, Madureia S,
vention study. Lancet. 2003;361:2046 –2048.
Pantoni L, Inzitari D, Erkinjuntti T, Crisby M, Waldemar G, Schmidt R,
33. Van Dijk E, Prins N, Vermeer S, Vrooman H, Hofman A, Koudstaal P,
Fazekas F, Scheltens P; on behalf of the LADIS Study Group. Small
Breteler M. C-reactive protein and cerebral small vessel disease: The
vessel disease and general cognitive function in non-disabled elderly: The
Rotterdam Scan study. Circulation. 2005;112:900 –905.
LADIS study. Stroke. 2005;36:2116 –2120.
34. Ten Dam V, van den Heuvel D, van Buchem M, Westendorp R, Bollen
14. Enzinger C, Fazekas F, Ropele S, Schmidt R. Progression of cerebral E, Ford I, de Craen A, Blauw G; PROSPER Study Group. Effect of
white matter lesions: clinical and radiological considerations. J Neurol pravastatin on cerebral infarcts and white matter lesions. Neurology.
Sci. 2007;257:5–10. 2005;64:1807–1809.
15. Schmidt R, Scheltens P, Erkinjuntti T, Pantoni L, Markus H, Wallin A, 35. Streifler J, Eliasziw M, Benavente O, Alamowitch S, Fox A, Hachinski V,
Barkhof F, Fazekas F; for the European Task Force on Age-Related Barnett H; North American Symptomatic Carotid Endarterectomy Trial
White Matter Changes. White matter lesion progression: a surrogate Group. Development and progression of leukoaraiosis in patients with
endpoint for trials in cerebral small vessel disease. Neurology. 2004;63: brain ischemia and carotid artery disease. Stroke. 2003;34:1913–1916.
139 –144. 36. Fazekas F, Kleinert R, Offenbacher H, Payer F, Schmidt R, Kleinert G,
16. Pantoni L, Simoni M, Pracucci G, Schmidt R, Barkhof F, Inzitari D. Radner H, Lechner H. The morphologic correlate of incidental punctate
Visual rating scales for age-related white matter changes (leukoaraiosis): MR white matter hyperintensities. AJNR Am J Neuroradiol. 1991;12:
can the heterogeneity be reduced. Stroke. 2002;33:2827–2833. 915–921.
17. Kapeller P, Barber R, Vermeulen R, Ader H, Freidl W, Almkvist O, 37. Fazekas F, Kleinert R, Offenbacher H, Schmidt R, Kleinert G, Payer F,
Moretti M, del Ser T, Vaghfeldt P, Enzinger C, Barkhof F, Schletens P, Radner H, Lechner H. Pathologic correlates of incidental MRI white
Inzitari D, Erkinjuntti T, Schmidt R, Fazekas F. Visual rating of age matter signal hyperintensities. Neurology. 1993;43:1683–1689.
related white matter changes on MRI-scale comparison, interrater 38. Burton E, McKeith I, Burn D, Firbank M, O’Brien J. Progression of white
agreement and the correlation with quantitative measurement. Stroke. matter hyperintensities in Alzheimer disease, dementia with lewy bodies,
2003;34:441– 445. and Parkinson disease dementia: a comparison with normal aging. Am J
18. Fazekas F, Chawluk J, Alavi A, Hurtig H, Zimmerman R. MR signal Geriatr Psychiatry. 2006;14:842– 849.
abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. AJR 39. van den Heuvel D, Admiraal-Behloul F, ten Dam V, Olofsen H, Bollen E,
Am J Roentgenol. 1987;149:351–356. Murray H, Blauw G, Westendorp R, de Craen A, van Buchem M;
Schmidt et al Progression of Leukoaraiosis and Cognition 2625

PROSPER Study Group. Different progression rates for deep white 47. Zelinski E, Burnight K. Sixteen-year longitudinal and time lag changes in
matter hyperintensities in elderly men and women. Neurology. 2004;63: memory and cognition in older adults. Psychol Aging. 1997;12:503–513.
1699 –1701. 48. Ferrer E, Salthouse T, Stewart W, Schwartz B. Modeling age and retest
40. Schmidt R, Ropele S, Enzinger C, Petrovic K, Smith S, Schmidt H, processes in longitudinal studies of cognitive abilities. Psychol Aging.
Matthews P, Fazekas F. White matter lesion progression, brain atrophy, 2004;19:243–259.
and cognitive decline: The Austrian Stroke Prevention study. Ann Neurol. 49. Inzitari D, Simoni M, Pracucci G, Poggesi A, Basile A, Chabriat H,
2005;58:610 – 616. Erkinjuntti T, Fazekas F, Ferro J, Hennerici M, Langhorne P, O’Brien J,
41. Van den Heuvel D, ten Dam V, de Craen A, Admiraal-Behloul F, Olofsen Barkhof F, Visser M, Wahlund L, Waldemar G, Wallin A, Pantoni L;
H, Bollen E, Jolles J, Murray H, Blauw G, Westendorp R, van Buchem LADIS Study Group. Risk of rapid global functional decline in elderly
M. Increase in periventricular white matter hyperintensities parallels patients with severe cerebral age-related white matter changes: The
decline in mental processing speed in a non-demented elderly population. LADIS study. Arch Intern Med. 2007;167:81– 88.
J Neurol Neurosurg Psychiatry. 2006;77:149 –153. 50. Dufouil C, Chalmers J, Coskun O, Besancon V, Bousser M, Guillon P,
42. Mungas D, Jagust W, Reed B, Kramer J, Weiner M, Schuff N, Norman MacMahon S, Mazoyer B, Neal B, Woodward M, Tzourio-Mazoyer N,
D, Mack W, Willis L, Chui H. MRI predictors of cognition in subcortical Tzourio C; PROGRESS MRI Substudy Investigators. Effects of blood
ischemic vascular disease and Alzheimer’s disease. Neurology. 2001;57: pressure lowering on cerebral white matter hyperintensities in patients
2229 –2235. with stroke: The PROGRESS (perindopril protection against recurrent
43. Fein G, Di Sclafani V, Tanabe J, Cardenas V, Weiner M, Jagust W, Reed stroke study) magnetic resonance imaging substudy. Circulation. 2005;
B, Norman D, Schuff N, Kusdra L, Greenfield T, Chui H. Hippocampal 112:1644 –1650.
and cortical atrophy predict dementia in subcortical ischemic vascular 51. Ten Dam VB, FM, de Craen A, van den Heuvel D, Bollen E, Murray H,
disease. Neurology. 2000;55:1626 –1635. van Buchem M, Westendorp R, Blauw G; PROSPER Study Group. Lack
44. Zivadinov R, Sepcic J, Nasuelli D, De Masi R, Bragadin L, Tommasi M, of effect of pravastatin on cerebral blood flow or parenchymal volume
Zambito-Marsala S, Moretti R, Bratina A, Ukmar M, Pozzi-Mucelli R, loss in elderly at risk for vascular disease. Stroke. 2005;36:1633–1636.
Grop A, Cazzato G, Zorzon M. A longitudinal study of brain atrophy and 52. Wang C, Stebbins G, Nyenhuis D, de Toledo-Morrell L, Freels S,
cognitive disturbances in the early phase of relapsing-remitting multiple Gencheva E, Pedelty L, Sripathirathan K, Moseley M, Turner D, Gabrieli
sclerosis. J Neurol Neurosurg Psychiatry. 2001;70:773–780. J, Gorelick P. Longitudinal changes in white matter following ischemic
45. Bakshi R, Ariyaratana S, Benedict R, Jacobs L. Fluid-attenuated stroke: a three-year follow-up study. Neurobiol Aging. 2006;27:
inversion recovery magnetic resonance imaging detects cortical and jux- 1827–1833.
tacortical multiple sclerosis lesions. Arch Neurol. 2001;58:742–748. 53. Holtmannspotter M, Peters N, Opherk C, Martin D, Herzog J, Bruckmann
46. Rabbit P, Diggle P, Smith D, Holland F, McInnes L. Identifying and H, Samann P, Gschwendtner A, Dichgans M. Diffusion magnetic res-
separating the effects of practice and of cognitive ageing during a large onance histograms as a surrogate marker and predictor of disease pro-
longitudinal study of elderly community residents. Neuropsychologia. gression in cadasil: a two-year follow-up study. Stroke. 2005;36:
2001;39:532–543. 2559 –2565.
Downloaded from http://ahajournals.org by on March 8, 2019

You might also like