REVIEWS White matter hyperintensities, cognitive impairment and dementia: an update Niels D. Prins and Philip Scheltens Abstract | White matter hyperintensities (WMHs) in the brain are the consequence of cerebral small vessel disease, and can easily be detected on MRI. Over the past three decades, research has shown that the presence and extent of white matter hyperintense signals on MRI are important for clinical outcome, in terms of cognitive and functional impairment. Large, longitudinal population-based and hospital-based studies have confirmed a dose-dependent relationship between WMHs and clinical outcome, and have demonstrated a causal link between large confluent WMHs and dementia and disability. Adequate differential diagnostic assessment and management is of the utmost importance in any patient, but most notably those with incipient cognitive impairment. Novel imaging techniques such as diffusion tensor imaging might reveal subtle damage before it is visible on standard MRI. Even in Alzheimer disease, which is thought to be primarily caused by amyloid, vascular pathology, such as small vessel disease, may be of greater importance than amyloid itself in terms of influencing the disease course, especially in older individuals. Modification of risk factors for small vessel disease could be an important therapeutic goal, although evidence for effective interventions is still lacking. Here, we provide a timely Review on WMHs, including their relationship with cognitive decline and dementia. Prins, N. D. & Scheltens, P. Nat. Rev. Neurol. advance online publication 17 February 2015; doi:10.1038/nrneurol.2015.10

Introduction

Alzheimer Center, Department of Neurology, VU University Medical Center, Neuroscience Campus Amsterdam, De Boelelaan 1117, 1081 HV Amsterdam, Netherlands (N.D.P., P.S.). Correspondence to: P.S. [email protected]

White matter hyperintensities (WMHs) are frequently seen on MRI scans of the brain in older people.1,2 The underlying pathology of these lesions mostly reflects demyelination and axonal loss as a consequence of chronic ischaemia caused by cerebral small vessel disease (microangiopathy). The prevalence and severity of WMHs increase with age, and in association with arterial hypertension.3,4 Clear evidence exists that WMHs lead to cognitive decline and have a role in the aetiology of dementia.5–8 WMHs, together with lacunar infarcts and cerebral microbleeds, are considered to be the primary pathology in subcortical ischaemic vascular dementia (VaD).9–11 Furthermore, vascular factors and cerebrovascular pathologies such as WMHs are increasingly recognized to be involved in the aetiology of Alzheimer disease (AD).12–15 In individual patients with cognitive impairment, it is often challenging to determine the extent to which WMHs contribute to the clinical picture. The total volume and location of WMHs are important determinants of the clinical relevance of these lesions,16–18 and visual and automated rating scales have been developed to assess the severity and progression of WMHs.19–23 Furthermore, evidence is accumulating that novel imaging techniques, such as diffusion tensor imaging (DTI), can detect subtle impairments in white matter tract integrity before they can be seen on conventional Competing interests The authors declare no competing interests.

MRI, suggesting that WMHs represent only the extreme end of a continuous spectrum of white matter injury.24,25 MRI also enables the measurement of progression of WMHs over time.26–28 As regards the underlying pathology, there is little doubt that the majority of these lesions are ischaemic in origin.29 However, a number of conditions can be considered in the differential diagnosis of WMHs on MRI,30 especially in younger middle-aged individuals. Results from prospective studies have provided strong evidence that WMHs cause cognitive decline, in particular of information processing speed,7,31–33 and might lead to dementia directly or indirectly through several ­different mechanisms.5,8 In dementia research, the focus is largely on the amyloid hypothesis, and the influence of cerebral small vessel disease, including WMHs, is sometimes neglected. However, in order to delay the onset and progression of dementia, vascular factors such as WMHs may be more important than the presence of amyloid, and should, therefore, be acknowledged and addressed. In view of the ongoing debate about the roles of amyloid versus vascular changes, especially in elderly individuals, a timely Review on WMHs, including their relationship with ­cognitive decline and dementia, is warranted.34,35

Imaging of WMHs

In 1987, Hachinski, Potter and Merskey introduced the term ‘leukoaraiosis’ to designate bilateral and symmetrical areas in white matter of the periventricular region and centrum semiovale that appeared hypodense on

NATURE REVIEWS | NEUROLOGY

ADVANCE ONLINE PUBLICATION  |  1 © 2015 Macmillan Publishers Limited. All rights reserved

REVIEWS Key points ■■ White matter hyperintensities (WMHs) are commonly seen on brain MRI in older people, and result from chronic ischaemia associated with cerebral small vessel disease ■■ The histopathology of WMHs is heterogeneous, with tissue damage ranging from slight disentanglement of the matrix to varying degrees of myelin and axonal loss ■■ This heterogeneity might partly explain the weak clinicoradiological associations found in patients with WMHs ■■ WMHs cause cognitive decline—in particular of information processing speed—and may lead to executive dysfunction and, ultimately, dementia ■■ Although progression of WMHs has been associated with dementia and dependency, little evidence is available that reduction of WMH progression can prevent functional decline

Box 1 | Visual rating of WMHs according to Fazekas19 Fazekas 0 No WMHs Fazekas 1 Focal or punctate lesions: ■■ Single lesions ≤9 mm ■■ Grouped lesions 20 mm in any diameter ■■ No more than connecting bridges between individual lesions Fazekas 3 Confluent lesion: ■■ Single lesions or confluent areas of hyperintensity ≥20 mm in any diameter Abbreviation: WMHs, white matter hyperintensities.

CT.36 This neutral term was deemed suitable to define and describe white matter damage, and would also function as a trigger for further research aimed at identifying its clinical and imaging correlates. The term leukoarai­ osis is still in use to designate CT‑visible white matter lesions that are caused by cerebral small vessel disease, as opposed to demyelinating, infectious, toxic or metabolic processes. When these white matter changes are seen on MRI, they are usually referred to as WMHs, or white matter lesions. WMHs are typically defined as hyperintense on proton-­density, fluid-attenuated inversion recovery and T2-weighted images, without prominent hypointensity on T1-weighted scans. WMHs can be focal or multi­ focal and, as they become more extensive, they become confluent and may involve large areas of the white matter. In addition to the brain regions already mentioned, these lesions can be located in the white matter of the basal ganglia and thalamus.37 In their mild form, WMHs often appear as small ‘caps’ on the frontal and/or occipital horns and as thin ‘rims’ along the walls of the lateral ventricles on transverse sections (periventricular lesions, or PVLs), or as punctuate foci in the subcortical white matter, also referred to as ‘deep white matter’ (deep WMHs). As the severity of the lesions increases, the periventricular WMHs may extend into the s­ ubcortical white matter, where they can become confluent.

At present, no gold standard has been established for the assessment of WMHs, and various methods are available to quantify the presence and severity of these lesions on CT or MRI, ranging from visual rating to semi-­ automated and fully automated methods. Visual rating scales offer the advantage of being quite fast and reliable when employed by an experienced rater. Furthermore, they do not require sophisticated and expensive postprocessing facilities. On the other hand, semi-automated and automated methods have the advantage of providing exact WMH volumes, which are desirable when one is looking for subtle associations. Widely used visual rating scales include those introduced by Fazekas and Scheltens. The Fazekas scale (Box 1) rates WMHs in the periventricular and subcortical region combined, on a 0–3-point scale.38 The Scheltens scale (Figure 1) rates WMHs separately in the periventricular region on a 0–6-point scale, and in the subcortical region on a 0–24-point scale, on the basis of the size and number of the lesions. This scale also includes ratings for basal ganglia and infratentorial areas.20 Figure 2 shows examples of brain scans graded according to the Fazekas scale. In addition to these scales, Wahlund and co-workers introduced a scale applicable to both CT and MRI, which is easy to use and compare between modalities.37 The Fazekas and Scheltens scales were designed for cross-sectional rating of WMHs. These scales show good intraobserver and interobserver agreement for measuring WMHs at a single time point. Furthermore, the results obtained on these scales correlate closely with volumetric assessments.39 However, given that these visual rating scales are time-consuming to apply, have a ceiling effect and use rather broad categories for severity, they are not suitable for measuring WMH progression.40 Although semiquantitative visual scales have been developed for measuring WMH change, the automated WMH detection methods that are now available allow the most precise and rapid quantification of WMH lesion load and WMH progression through use of image subtraction.41–43 WMHs may represent only the extreme end of a continuous spectrum of white matter injury, creating a need for imaging approaches that can detect more-subtle changes. DTI is a technique that enables white matter tract integrity to be assessed. DTI measures of white matter integrity have been linked to episodic memory function.25 Injury to white matter tracts that connect frontal and temporal cortex, and frontal cortex and striatum, might to some extent explain the differences in memory performance between older individuals. Impaired white matter integrity, as measured by DTI, has been independently, and in a stepwise manner, associated with an increased risk of future WMH development.44 Studies in patients with mild cognitive impairment (MCI) showed that DTI measures of white matter tract integrity accurately predicted cognitive performance, and cerebral small vessel disease was associated with microstructural damage in fibre tracts subserving the default mode network.24,45 These data emphasize that DTI might provide complementary information for understanding the time course of ­ageing-associated white matter degeneration.

2  |  ADVANCE ONLINE PUBLICATION

www.nature.com/nrneurol © 2015 Macmillan Publishers Limited. All rights reserved

REVIEWS Periventricular WMHs (total score 0–6) Frontal caps (0–2) Occipital caps (0–2) Bands (0–2)

Deep WMHs (total score 0–24) Frontal (0–6) Parietal (0–6) Occipital (0–6) Temporal (0–6)

Basal ganglia WMHs (total score 0–30) Caudate nucleus (0–6) Putamen (0–6) Globus pallidus (0–6) Thalamus (0–6) Internal/external capsule (0–6)

Infratentorial WMHs (total score 0–24) Cerebellum (0–6) Mesencephalon (0–6) Pons (0–6) Medulla (0–6)

Score per WMH location (total score 0–78) 0, None 1, 5 5, >10 mm; n ≥1 6, Confluent

Score per WMH type (total score 0–6) 0, None 1, Smooth halo (>1–5 mm) 2, Large confluent lesions

Total score 0–84 20 Figure 1 | Visual rating of WMHs according to Scheltens.Nature Abbreviation: Reviews |WMH, Neurology white matter hyperintensity.

T2* gradient-recalled echo and susceptibility-weighted MRI sequences visualize another type of cerebral small vessel disease: cerebral microbleeds. These small, punctuate hypointense lesions correspond to areas of haemo­ siderin breakdown products from prior microscopic haemorrhages.46 Microbleeds are associated with WMHs and lacunar infarcts on MRI, as well as increased cerebral amyloid‑β (Aβ) on PET, and in this sense they form the link between arteriolosclerosis and cerebral amyloid ­angiopathy (CAA).47–50

Epidemiology of WMHs Incidence and prevalence The reported prevalence and incidence of WMHs varies between studies as a result of differences in imaging techniques and rating methods used, and in the populations that were studied.2,51 Some studies only included healthy individuals, whereas others took random samples from the general population, or from dementia or stroke cohorts. In the general population, the prevalence of any WMHs ranges from 39% to 96%. The Helsinki Ageing Study and Austrian Stroke Prevention Study reported relatively low prevalence rates for any WMHs of 39% and 45%, respectively, which can be explained by the low-field-strength MRI that was used in the former, and by the fact that people with neurological or psychiatric disease or abnormal neurological examination were excluded from the latter.52,53 In the population-based Cardiovascular Health Study and Rotterdam Scan Study, both of which only included participants above the age of 60 years, the prevalence of any WMHs was high: 96% and 95%, respectively.1,54 In a case–control study by Barber et al.,55 the prevalence of any periventricular WMHs was 100% in participants with VaD, AD or dementia with Lewy bodies (DLB), compared with 92% in controls. Subcortical WMHs and hyperintensities in the basal ganglia were present in 96% of patients with VaD, 89% of patients with

AD and 85% of patients with DLB, compared with 73% of controls. A Korean study in patients with stroke found an associ­ ation between frequency of severe WMHs and stroke subtype.56 A group of patients with large artery disease had a higher prevalence of WMHs than did the other groups in the study (55.4% in the large artery disease group, compared with 30.3% of patients with lacunar stroke and 14.3% of patients with cardioembolic stroke).

Progression Data on progression of WMHs have been provided by both population-based and hospital-based cohort studies. In the Rotterdam Scan Study, a semiquanti­tative scale was used to assess changes in periventricular and subcortical WMH severity.40 Of the 668 participants, who had a mean age of 71 years and underwent MRI and neuropsychological testing at baseline and at 3 years’ follow-up, 39% showed visible progression of WMHs (32% in the subcortical white matter, and 27% in the periventricular white matter regions).28 Predictors of progression were baseline severity of leukoaraiosis (as expected), age, blood pressure, current smoking, and the presence of lacunar infarcts. Atrial fibrillation, carotid atherosclerosis and homocysteine were not related to WMH progression. Progression of WMHs in the peri­ ventricular regions was associated with parallel declines in information processing speed and general cognition, as well as decline on the Mini-Mental State Examination (MMSE), whereas progression of subcortical WMHs was not associated with cognitive decline. The LADIS (Leukoaraiosis And DISability) study is a European multicentre collaboration that was started in 2001 with the aim of assessing the independent role of WMHs in predicting disability in people aged 65–84 years.57 Individuals who showed no impairment or were impaired on only one item of the Instrumental Activities of Daily Living (IADL) scale, presenting with different grades of WMH severity, were enrolled in a hospital-based setting, and were monitored for up to 3 years at 11 European centres. WMHs were categorized as mild, moderate or severe on the Fazekas scale.19 Stratified enrolment according to WMH severity allowed relative oversampling of individuals with mild and severe WMHs. WMH progression was measured with a semiquantitative visual scale very similar to the one used in the Rotterdam Scan Study.40 WMH progression was seen in 290 (73.6%) of the 394 study participants, and was again associated with baseline WMH severity: 57% of participants with punctate WMHs, 89% with early confluent WMHs, and 84% with confluent WMHs at ­baseline showed WMH progression after 3 years.27 In the Austrian Stroke Prevention Study, a semiautomated volumetric method was used to measure WMH progression in 329 older people from the general population, who underwent repeated MRI scanning and cognitive testing at baseline and at 3‑year and 6‑year follow-up.58 After 6 years, the median increase in white matter lesion load was 0.2 cm3 (interquartile range 0.0–0.8 cm3), with a maximum of 31.4 cm3.

NATURE REVIEWS | NEUROLOGY

ADVANCE ONLINE PUBLICATION  |  3 © 2015 Macmillan Publishers Limited. All rights reserved

REVIEWS Fazekas 1

Fazekas 2

Fazekas 3

Nature Reviews Figure 2 | Axial fluid-attenuated inversion recovery images illustrating the| Neurology Fazekas scores. The scoring system is outlined in Box 1.

Pathophysiology of WMHs

The differential diagnosis of WMHs is extensive, but clinical and pathological data suggest that ischaemia is the cause of these lesions in the majority of older people.29 The most consistently reported histological substrates of WMHs are diffuse myelin rarefaction with sparing of the subcortical U fibres, astrogliosis, spongiosis, axonal loss, and widened perivascular spaces.29,59 However, the histopathology of WMHs is heterogeneous, as shown by correlative MRI and postmortem studies that reported tissue damage ranging from slight disentanglement of the matrix to varying degrees of myelin and axonal loss (Figures 3 and 4).60 This heterogeneity might partly explain the clinicoradiological paradox— that is, the weak correlation between MRI abnormalities and clinical symptomatology—with regard to WMHs. According to population-based studies, the main risk factors for the development of white matter lesions are greater age, arterial hypertension and cardio­vascular disease. 26,61–64 Although diabetes is a risk factor for athero­sclerosis and incident brain infarcts, most studies found no association or only a weak association between diabetes and WMHs. 65–68 The combination of art­ eriolo­sclerosis and hyaline wall thickening of the long penetrating arterioles on the one hand and impaired autoregulation on the other could result in hypoxia and ischaemia of the white matter. Impaired blood–brain

barrier permeability as a result of ischaemia might lead to leakage of macromolecules, followed by astrocyte activation. 4 Alternative mechanisms have also been proposed. Aβ deposition in the medial and adventitial layers of meningocortical arteries and arterioles may lead to obstruction of the vessel luminae and to the loss of vascular smooth muscle cells, resulting in impaired cerebral autoregulation.69,70 Another factor that has been implicated in the formation of deep WMHs is colla­ genous thickening and occlusion of deep periventricular-­ draining veins.71 In degenerative dementia, WMHs have been attributed to Wallerian degeneration.72 Given that WMHs and other features of cerebral small vessel disease are often comorbid with AD pathology, a recurring question is whether small vessel disease and AD pathology interact. Data from the Neuropathology Group of the Medical Research Council Cognitive Function and Ageing Study (MRC CFAS) showed no association between vascular risk factors and an increased burden of AD pathology in the brains of older people.73 The same study group found that the apolipoprotein E (APOE) genotype, a well-known genetic risk factor for AD, was associated with the key features of AD pathology but not with cerebrovascular diseases other than CAA.74 This finding is consistent with evidence that WMHs correlate with amyloid burden in APOE ε4 noncarriers but not in APOE ε4 carriers, suggesting that in the latter individuals the effects of APOE ε4 override the effects of WMHs.75 WMHs on MRI and amyloid load on PET were also found to independently relate to the integrity of fibre tracts, as measured with DTI. 76 This observation is in line with imaging data suggesting that WMHs and amyloid burden each correlate with specific neuro­ psychiatric symptoms. 77 Data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) showed that both WMHs and reduced cerebrospinal fluid levels of Aβ1–42, a biomarker for AD pathology, were independently associated with brain atrophy rates in controls.78 However, the association between WMHs and overall brain atrophy rate was not present in patients with MCI or AD. Reduced levels of Aβ1–42 and elevated temporal WMH volumes were associated with smaller entorhinal cortex and hippocampus volumes in patients with MCI. These data support the view that WMHs are associated with increased brain atrophy in the context of AD pathology in pre-dementia stages.79 Several monogenic conditions involve the cerebral small vessels and predispose individuals to diffuse WMHs, stroke and VaD. These hereditary conditions are rare, and are characterized by onset of clinical manifestations at around 50 years of age. White matter lesions are a prominent feature in CADASIL (cerebral auto­somal dominant arteriopathy with subcortical infarcts and leukoencephalopathy).80,81 This condition is caused by mutations in the NOTCH3 gene, which encodes a receptor that is expressed in vascular smooth muscle cells. Similar WMHs are found in the recessively inherited CARASIL, which is caused by mutations in the serine protease HTRA1 gene that result in reduced function

4  |  ADVANCE ONLINE PUBLICATION

www.nature.com/nrneurol © 2015 Macmillan Publishers Limited. All rights reserved

REVIEWS a

c

b

cerebral haemorrhage with amyloidosis (HCHWA) is a rare form of CAA that can present with WMHs.88 Cerebral WMHs are also seen in multiple sclerosis,89 leukodystrophies,90 mitochondrial encephalopathies91 and certain infections, and after cerebral anoxia from cardiac arrest or carbon monoxide poisoning.30 In the elderly population, however, most of these conditions are rare.

d

WMHs and the development of dementia Reviews | Neurology Figure 3 | Direct matching of WMHs on postmortem MRI Nature scans and histochemically stained sections. Images show a coronal frontal brain slice at the level of the basal ganglia (right hemisphere) from a patient with Alzheimer disease. a | Formalin-fixed brain section with the hippocampal area removed. b | T2-weighted postmortem MRI scan with WMH (red circle). c,d | Histochemically stained hemispheric sections. Bodian silver stain (part c) reveals axonal loss, and Luxol fast blue–cresyl violet staining (part d) reveals demyelination. Abbreviation: WMH, white matter hyperintensity.

a

c

b

d

e

Figure 4 | Direct matching of WMHs on postmortem MRI scans histochemically Natureand Reviews | Neurology stained sections. Images show a coronal prefrontal brain slice (right hemisphere) of a 98-year-old individual without cognitive impairment. a | T2-weighted and b | 3D fluid-attenuated inversion recovery postmortem MRI scans showing periventricular WMHs (red circle). c–e | Hemispheric sections histochemically stained with haematoxylin and eosin (part c), Bodian silver (part d) and Luxol fast blue–cresyl violet (part e). Bodian silver reveals axonal loss, and Luxol fast blue–cresyl violet reveals demyelination. Abbreviation: WMHs, white matter hyperintensities.

of the HTRA1 protein as a repressor of transforming growth factor β.81,82 Evidence from family and twin studies suggest that genetic factors contribute to WMH pathogenesis. However, genetic association studies, conducted using candidate genes, have so far failed to demonstrate a convincing association between WMHs and genetic variants.83 A meta-analysis of genome-wide association studies for WMH burden in 9,361 stroke-free individuals of European descent from seven community-based cohorts identified susceptibility genotypes for WMHs at the chromosome 17q25 locus.84 The association between these genotypes and WMH volume was confirmed in patients with ischaemic stroke, and in individuals of ­non-European origin.85,86 A number of conditions can be considered in the differential diagnosis of WMHs on MRI.30 Besides ischaemic WMHs, several other white matter changes have a hypoxic–ischaemic origin. CAA results from deposition of Aβ in the media and adventitia of small arteries and capillaries of the leptomeninges and cerebral cortex, and is an important cause of lobar intracerebral haemorrhage and cognitive impairment in older people. CAA is associated with measures of small vessel disease on MRI, including cerebral microbleeds and WMHs.87 Hereditary

Equivocal results from cross-sectional studies have contributed to a long debate on whether WMHs on MRI are truly related to cognitive decline and dementia.92–94 Results from prospective studies have provided clear evidence that small vessel disease, including WMHs, causes cognitive decline, and that WMHs are particularly related to reductions in information processing speed and executive dysfunction, consistent with prevailing subcortical damage.7,31–33,95 In the prospective Cardiovascular Health Study, the combination of WMHs and infarcts carried the worst prognosis with regard to long-term survival, self-rated health, and limitation in activities of daily living (ADLs).96 In the Rotterdam Scan Study, progression of WMHs in the periventricular regions was paralleled by declines in information processing speed and general cognition, as well as decline on the MMSE.28 The LADIS study reported 2.4-year followup data for its observational cohort of 633 older people, and found a transition rate of 29.5% to death or disability in ADLs for those with severe WMH (Fazekas 3) at entry, compared with 10% in those with only mild changes.27,97 In this study, the progression of WMHs rather than the baseline level correlated best with cognitive decline. Both baseline WMH severity and WMH progression have also been linked to dementia risk.5,6,98 In the Rotterdam Scan Study, WMHs and brain infarcts at baseline were strongly associated with an increased risk of subsequent development of dementia, including AD.8,98 This temporality provides strong evidence for a causal relationship between WMHs and cognitive decline and dementia. Several mechanisms have been proposed for the association between WMHs and cognitive decline and dementia. Small vessel disease can interrupt prefrontal subcortical loops directly, or might indirectly lead to cognitive decline through incident stroke. 8 Damage to neurotransmitter systems, in particular the cholinergic system, could also account for the cognitive sequelae of WMHs.99 Vascular risk factors, such as hypertension, diabetes and hyperhomocysteinaemia, may be related to cognitive impairment through other mechanisms.100–102 The picture is further complicated by the presence of comorbid brain pathologies, of which AD encephalo­ pathy is the most prevalent. Indeed, it has been suggested that WMHs can lead to dementia through interaction with AD pathology.35,103,104 Animal models provide an additional means of studying the mechanisms of cognitive decline due to small vessel disease, including WMHs. WMHs and cognitive deficits have been reported in various mouse models

NATURE REVIEWS | NEUROLOGY

ADVANCE ONLINE PUBLICATION  |  5 © 2015 Macmillan Publishers Limited. All rights reserved

REVIEWS of cardiovascular disease. Surgical animal models, such as the bilateral common carotid artery stenosis hypo­ perfusion mouse model, the transverse aortic constriction model, and myocardial infarction models, display the key features of vascular cognitive impairment.105,106 Observations from epidemiological and pathological studies challenge the conventional distinction between AD and VaD, as defined by diagnostic criteria. In a community-based neuropathological study, AD‑type and vascular pathology were the main pathological corre­lates of dementia in elderly people, although most patients had mixed disease.35 No clear thresholds of these features that predicted dementia status were apparent. Imaging studies have provided evidence for combined involvement of AD and vascular pathology in cognitive decline, and have found an additive effect of WMHs and hippocampal atrophy on MRI, as well as an associ­ ation between periventricular WMHs and hippocampal atrophy.107,108 These findings suggest that dementia in older people is often a heterogeneous condition in which different pathologies contribute to the clinical dementia syndrome. Most older people have some WMHs, and it is challenging to determine the degree to which WMHs account for the observed cognitive deficits and/or dementia. This issue is not easy to resolve, and in clinical practice, expert judgement is frequently used to determine the causal relationship between the presence of WMHs and the clinical picture. Some investigators have suggested that at least 25% of the total white matter should be affected to support a diagnosis of VaD, and one study in 51 healthy individuals aged 19–91 years noted that a WMH volume of greater than 0.5% of intracranial volume was associated with increased ventricular volume, reduced brain volume, and reduced cognitive scores.109–111 The nature and severity of cognitive impairment associated with WMHs will also depend on the volume and location of lesions, and probably on other factors such as brain reserve.112,113 As a general rule, one can state that in order to determine that WMHs are the cause of the cognitive impairment, the lesions should be extensive and confluent, and a temporal relationship to the cognitive impairment, demonstrated by repeated MRI, would also be highly supportive.

WMHs as a target for therapy

Treatment approaches for WMHs can be categorized as primary and secondary prevention, and symptomatic and disease-modifying treatment of VaD. The aim of primary prevention with regard to WMHs is to promote good health and remove relevant risk factors before the cognitive symptoms occur.114 Secondary prevention strat­egies are initiated after cognitive or mood symptoms have become manifest, with the goal of preventing the ­progression towards dementia. Targets for primary and secondary prevention of dementia associated with WMHs include the known vascular risk factors hypertension, obesity, diabetes and hypercholesterolaemia. In particular, high diastolic blood pressure and hypertension have been shown to be risk

factors for both the presence of WMHs and progression of WMHs over time.63,64 In one study, WMH progression was significantly greater in people with uncontrolled untreated hypertension than in those with uncontrolled treated hypertension, indicating that blood pressure control is likely to be effective in the prevention of WMH occurrence and worsening.26 Several clinical trials of anti­hyper­tensive drugs have found reductions in cardiovascular morbidity, but only the Syst-Eur trial additionally reported a reduced incidence of dementia in the treatment group.115 In another study involving participants with long-standing type 2 diabetes and at high risk of cardiovascular events, intensive blood pressure control and fibrate therapy in the presence of controlled LDL cholesterol levels did not produce a measurable effect on cognitive decline at 40 months of follow-up.116 Cholesterol is important in dementia, not only because of its relationship with cardiovascular disease, but also owing to its role in amyloid metabolism. In the PROSPER study, the effects of pravastatin 40 mg daily on the progression of ischaemic brain lesions were evaluated though use of repeated brain MRI.117 After a mean treatment period of 33 months, no difference in lesion progression was observed between the pravastatintreated and placebo groups. To date, no trials related to weight reduction or anti-obesity drugs and dementia have been conducted. In the EVA trial, a randomized, controlled clinical trial with 123 participants, intensive vascular care was compared with standard care in patients with AD who exhibited concomitant cerebrovascular lesions on MRI.118,119 Individuals receiving vascular care showed less WMH progression, but the two groups did not differ with regard to number of new lacunes and change in global cortical atrophy or medial temporal lobe atrophy. 118 In addition, no differences in clinical function were observed between the intervention and control groups.119 The PreDIVA study, a large randomized trial with 6‑year follow-up in 3,700 older people, will study whether nurse-led vascular care in the primary care setting can decrease the incidence of dementia.120 The effects of physical exercise on WMH progression will be studied in the Australian Imaging Biomarkers and Lifestyle Flagship Study of Aging (AIBL) Active trial.121 In this single-blind randomized controlled trial, 156 parti­cipants with subjective memory complaints or MCI, aged 60 years and older, will undergo 24 months of moderate, home-based physical exercise and a behavioural intervention package, and change in WMHs on MRI will be one of the outcome measures. Symptomatic treatment with the cholinesterase inhibitors donepezil and rivastigmine has been studied in randomized clinical trials in patients with VaD. A combined analysis of two large trials in patients with VaD showed that patients who were treated with donepezil performed better on measures of cognition, global function and IADL function than did those treated with placebo.122 On the basis of these findings, and the fact that the drug was well tolerated, donepezil could be considered as an option in the symptomatic treatment of patients with

6  |  ADVANCE ONLINE PUBLICATION

www.nature.com/nrneurol © 2015 Macmillan Publishers Limited. All rights reserved

REVIEWS VaD. The results from an open-label, 30-week extension study suggested that donepezil improved cognition for up to 54 weeks in patients with VaD.123 No unexpected adverse effects attributable to the drug occurred. Patients who started to use donepezil in this extension study performed less well on the primary outcome measure than did those initiating donepezil in the double-blind study. A 6‑month trial of the cholinesterase inhibitor and central nicotinic receptor modulator galantamine was conducted in patients diagnosed with probable VaD or with AD plus cerebrovascular disease.124 The drug demonstrated efficacy with regard to cognition, as measured with the ADAS-Cog, and global functioning, as ­measured with the CIBIC-plus. Memantine is an N‑methyl‑d-aspartate receptor antagonist. A multicentre, 28-week trial of this drug in 288 patients with VaD showed beneficial effects on cognition, as measured with the MMSE and ADAS-Cog, but not on global function measured with the CIBIC-plus. No adverse effects were observed.125

Conclusions

Despite the fact that little new knowledge has been gathered in recent years, the large observational, cross-­ sectional and longitudinal studies reviewed in this article have provided clinically meaningful insights into the role of white matter changes in cognition and dementia. Small punctate lesions (Fazekas 1) are highly prevalent, but have little clinical relevance. The larger lesions, especially when confluent, do interfere with normal cognitive function, and can lead to disability within 1 year of assessment.126 1.

2.

3.

4.

5.

6.

7.

8.

Longstreth, W. T. Jr et al. Clinical correlates of white matter findings on cranial magnetic resonance imaging of 3301 elderly people. The Cardiovascular Health Study. Stroke 27, 1274–1282 (1996). van Dijk, E. J., Prins, N. D., Vermeer, S. E., Koudstaal, P. J. & Breteler, M. M. Frequency of white matter lesions and silent lacunar infarcts. J Neural Transm. Suppl. 62, 25–39 (2002). de Leeuw, F. E., de Groot, J. C. & van Gijn, J. Cerebral white matter lesions in the elderly: vascular risk factors and cognitive consequences [Dutch]. Ned. Tijdschr. Geneeskd. 145, 2067–2071 (2001). Pantoni, L. & Garcia, J. H. Pathogenesis of leukoaraiosis: a review. Stroke 28, 652–659 (1997). Brickman, A. M. et al. Reconsidering harbingers of dementia: progression of parietal lobe white matter hyperintensities predicts Alzheimer’s disease incidence. Neurobiol. Aging 36, 27–32 (2015). Debette, S. & Markus, H. S. The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis. BMJ 341, c3666 (2010). Prins, N. D. et al. Cerebral small-vessel disease and decline in information processing speed, executive function and memory. Brain 128, 2034–2041 (2005). Vermeer, S. E. et al. Silent brain infarcts and the risk of dementia and cognitive decline. N. Engl. J. Med. 348, 1215–1222 (2003).

9.

10.

11. 12.

13.

14.

15.

16.

Much has been learned about WMH progression and its clinical meaning. However, whether reduction of progression is a valid therapeutic target and, if so, what kind of intervention to use and when to start, is still open to question, and should be the subject of future studies.127 With the heterogeneity of small vessel disease in mind, future research should also focus on tailor-made symptomatic treatment of patients with vascular cognitive impairment. Recognition of WMHs (Fazekas 2 and 3) in a patient presenting with cognitive decline should prompt a careful work-up and treatment of vascular risk factors. On the basis of the LADIS findings, the presence of moderate to severe WMHs also alerts the clinician to monitor the patient and manage functional decline, including gait and incontinence. This is not a trivial undertaking, and cognitive impairment should never be used as excuse not to treat a patient. Review criteria We identified articles by searching PubMed using the following terms: “white matter lesions”, “white matter hyperintensities”, “leukoaraiosis”, “cerebral small vessel disease”, “vascular dementia”, “vascular cognitive impairment”, “cognitive dysfuntion”, “Alzheimer’s disease” and “dementia”. Terms were used as MeSH terms, All Fields terms and Title terms. We limited the search to full-text manuscripts published in English. The reference lists of identified papers were searched for further relevant articles. Articles were also identified through searches of the authors’ own files. The final selection was based on quality and relevance, as judged by the authors.

Gorelick, P. B. et al. Vascular contributions to cognitive impairment and dementia: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 42, 2672–2713 (2011). Pantoni, L. Cerebral small vessel disease: from pathogenesis and clinical characteristics to therapeutic challenges. Lancet Neurol. 9, 689–701 (2010). Chui, H. C. Subcortical ischemic vascular dementia. Neurol. Clin. 25, 717–740 (2007). Lorius, N. et al. Vascular disease and risk factors are associated with cognitive decline in the Alzheimer disease spectrum. Alzheimer Dis. Assoc. Disord. http://dx.doi.org/10.1097/ WAD.0000000000000043. O’Brien, J. T. & Markus, H. S. Vascular risk factors and Alzheimer’s disease. BMC Med. 12, 218 (2014). Kalaria, R. N. & Ihara, M. Dementia: vascular and neurodegenerative pathways—will they meet? Nat. Rev. Neurol. 9, 487–488 (2013). Akinyemi, R. O., Mukaetova-Ladinska, E. B., Attems, J., Ihara, M. & Kalaria, R. N. Vascular risk factors and neurodegeneration in ageing related dementias: Alzheimer’s disease and vascular dementia. Curr. Alzheimer Res. 10, 642–653 (2013). Biesbroek, J. M. et al. Association between subcortical vascular lesion location and cognition: a voxel-based and tract-based lesionsymptom mapping study. The SMART-MR study. PLoS ONE 8, e60541 (2013).

NATURE REVIEWS | NEUROLOGY

17. Yoshita, M. et al. Extent and distribution of white matter hyperintensities in normal aging, MCI, and AD. Neurology 67, 2192–2198 (2006). 18. Swartz, R. H., Sahlas, D. J. & Black, S. E. Strategic involvement of cholinergic pathways and executive dysfunction: does location of white matter signal hyperintensities matter? J. Stroke Cerebrovasc. Dis. 12, 29–36 (2003). 19. Fazekas, F., Chawluk, J. B., Alavi, A., Hurtig, H. I. & Zimmerman, R. A. MR signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. AJR Am. J. Roentgenol. 149, 351–356 (1987). 20. Scheltens, P. et al. A semiquantative rating scale for the assessment of signal hyperintensities on magnetic resonance imaging. J. Neurol. Sci. 114, 7–12 (1993). 21. Gouw, A. A. et al. Simple versus complex assessment of white matter hyperintensities in relation to physical performance and cognition: the LADIS study. J. Neurol. 253, 1189–1196 (2006). 22. Maillard, P. et al. An automated procedure for the assessment of white matter hyperintensities by multispectral (T1, T2, PD) MRI and an evaluation of its between-centre reproducibility based on two large community databases. Neuroradiology 50, 31–42 (2008). 23. Maldjian, J. A. et al. Automated white matter total lesion volume segmentation in diabetes. AJNR Am. J. Neuroradiol. 34, 2265–2270 (2013). 24. Zhang, Y. et al. MRI markers for mild cognitive impairment: comparisons between white matter integrity and gray matter volume measurements. PLoS ONE 8, e66367 (2013).

ADVANCE ONLINE PUBLICATION  |  7 © 2015 Macmillan Publishers Limited. All rights reserved

REVIEWS 25. Lockhart, S. N. et al. Episodic memory function is associated with multiple measures of white matter integrity in cognitive aging. Front. Hum. Neurosci. 6, 56 (2012). 26. Verhaaren, B. F. et al. High blood pressure and cerebral white matter lesion progression in the general population. Hypertension 61, 1354–1359 (2013). 27. Schmidt, R. et al. White matter lesion progression in LADIS: frequency, clinical effects, and sample size calculations. Stroke 43, 2643–2647 (2012). 28. van Dijk, E. J. et al. Progression of cerebral small vessel disease in relation to risk factors and cognitive consequences: Rotterdam Scan study. Stroke 39, 2712–2719 (2008). 29. Pantoni, L. & Simoni, M. Pathophysiology of cerebral small vessels in vascular cognitive impairment. Int. Psychogeriatr. 15 (Suppl. 1), 59–65 (2003). 30. Román, G. C., Erkinjuntti, T., Wallin, A., Pantoni, L. & Chui, H. C. Subcortical ischaemic vascular dementia. Lancet Neurol. 1, 426–436 (2002). 31. De Groot, J. C. et al. Periventricular cerebral white matter lesions predict rate of cognitive decline. Ann. Neurol. 52, 335–341 (2002). 32. Garde, E., Lykke Mortensen, E., Rostrup, E. & Paulson, O. B. Decline in intelligence is associated with progression in white matter hyperintensity volume. J. Neurol. Neurosurg. Psychiatry 76, 1289–1291 (2005). 33. Swan, G. E. et al. Biobehavioral characteristics of nondemented older adults with subclinical brain atrophy. Neurology 54, 2108–2114 (2000). 34. Snowdon, D. A. et al. Brain infarction and the clinical expression of Alzheimer disease. The Nun Study. JAMA 277, 813–817 (1997). 35. Neuropathology Group. Medical Research Council Cognitive Function and Aging Study. Pathological correlates of late-onset dementia in a multicentre, community-based population in England and Wales. Neuropathology Group of the Medical Research Council Cognitive Function and Ageing Study (MRC CFAS). Lancet 357, 169–175 (2001). 36. Hachinski, V. C., Potter, P. & Merskey, H. Leukoaraiosis. Arch. Neurol. 44, 21–23 (1987). 37. Wahlund, L. O. et al. A new rating scale for agerelated white matter changes applicable to MRI and CT. Stroke 32, 1318–1322 (2001). 38. Fazekas, G. et al. Brain MRI findings and cognitive impairment in patients undergoing chronic hemodialysis treatment. J. Neurol. Sci. 134, 83–38 (1995). 39. Kapeller, P. et al. Visual rating of age-related white matter changes on magnetic resonance imaging: scale comparison, interrater agreement, and correlations with quantitative measurements. Stroke 34, 441–445 (2003). 40. Prins, N. D. et al. Measuring progression of cerebral white matter lesions on MRI: visual rating and volumetrics. Neurology 62, 1533–1539 (2004). 41. Wu, M. et al. A fully automated method for quantifying and localizing white matter hyperintensities on MR images. Psychiatry Res. 148, 133–142 (2006). 42. Gouw, A. A. et al. Reliability and sensitivity of visual scales versus volumetry for evaluating white matter hyperintensity progression. Cerebrovasc. Dis. 25, 247–253 (2008). 43. Schwarz, C., Fletcher, E., DeCarli, C. & Carmichael, O. Fully-automated white matter hyperintensity detection with anatomical prior knowledge and without FLAIR. Inf. Process. Med. Imaging 21, 239–251 (2009).

44. Maillard, P. et al. FLAIR and diffusion MRI signals are independent predictors of white matter hyperintensities. AJNR Am. J. Neuroradiol. 34, 54–61 (2013). 45. Papma, J. M. et al. Cerebral small vessel disease affects white matter microstructure in mild cognitive impairment. Hum. Brain Mapp. 35, 2836–2851 (2014). 46. Yates, P. A. et al. Cerebral microbleeds: a review of clinical, genetic, and neuroimaging associations. Front. Neurol. 4, 205 (2014). 47. Lee, S. H. et al. Dynamic temporal change of cerebral microbleeds: long-term follow-up MRI study. PLoS ONE 6, e25930 (2011). 48. Yates, P. A. et al. Incidence of cerebral microbleeds in preclinical Alzheimer disease. Neurology 82, 1266–1273 (2014). 49. Smith, E. E., Schneider, J. A., Wardlaw, J. M. & Greenberg, S. M. Cerebral microinfarcts: the invisible lesions. Lancet Neurol. 11, 272–282 (2012). 50. Koennecke, H. C. Cerebral microbleeds on MRI: prevalence, associations, and potential clinical implications. Neurology 66, 165–171 (2006). 51. Simoni, M. et al. Age- and sex-specific rates of leukoaraiosis in TIA and stroke patients: population-based study. Neurology 79, 1215–1222 (2012). 52. Ylikoski, A. et al. White matter hyperintensities on MRI in the neurologically nondiseased elderly. Analysis of cohorts of consecutive subjects aged 55 to 85 years living at home. Stroke 26, 1171–1177 (1995). 53. Schmidt, R. et al. Magnetic resonance imaging of the brain in diabetes: the Cardiovascular Determinants of Dementia (CASCADE) Study. Diabetes 53, 687–692 (2004). 54. de Leeuw, F. E. et al. Prevalence of cerebral white matter lesions in elderly people: a population based magnetic resonance imaging study. The Rotterdam Scan Study. J. Neurol. Neurosurg. Psychiatry 70, 9–14 (2001). 55. Barber, R. et al. White matter lesions on magnetic resonance imaging in dementia with Lewy bodies, Alzheimer’s disease, vascular dementia, and normal aging. J. Neurol. Neurosurg. Psychiatry 67, 66–72 (1999). 56. Lee, S. J. et al. The leukoaraiosis is more prevalent in the large artery atherosclerosis stroke subtype among Korean patients with ischemic stroke. BMC Neurol. 8, 31 (2008). 57. Pantoni, L. et al. Impact of age-related cerebral white matter changes on the transition to disability—the LADIS study: rationale, design and methodology. Neuroepidemiology 24, 51–62 (2005). 58. Schmidt, R. et al. White matter lesion progression, brain atrophy, and cognitive decline: the Austrian stroke prevention study. Ann. Neurol. 58, 610–616 (2005). 59. Pantoni, L. & Garcia, J. H. The significance of cerebral white matter abnormalities 100 years after Binswanger’s report. A review. Stroke 26, 1293–1301 (1995). 60. Gouw, A. A. et al. Heterogeneity of small vessel disease: a systematic review of MRI and histopathology correlations. J. Neurol. Neurosurg. Psychiatry 82, 126–135 (2011). 61. de Leeuw, F. E. et al. Aortic atherosclerosis at middle age predicts cerebral white matter lesions in the elderly. Stroke 31, 425–429 (2000). 62. de Leeuw, F. E. et al. Hypertension and cerebral white matter lesions in a prospective cohort study. Brain 125, 765–772 (2002). 63. Goldstein, I. B., Bartzokis, G., Guthrie, D. & Shapiro, D. Ambulatory blood pressure and the brain: a 5‑year follow-up. Neurology 64, 1846–1852 (2005).

8  |  ADVANCE ONLINE PUBLICATION

64. Gottesman, R. F. et al. Blood pressure and whitematter disease progression in a biethnic cohort: Atherosclerosis Risk in Communities (ARIC) study. Stroke 41, 3–8 (2010). 65. Brundel, M., Kappelle, L. J. & Biessels, G. J. Brain imaging in type 2 diabetes. Eur. Neuropsychopharmacol. 24, 1967–1981 (2014). 66. Tiehuis, A. M. et al. Metabolic syndrome, prediabetes, and brain abnormalities on MRI in patients with manifest arterial disease: the SMART-MR study. Diabetes Care 37, 2515–2521 (2014). 67. Exalto, L. G., van der Flier, W. M., Scheltens, P., Vrenken, H. & Biessels, G. J. Dysglycemia, brain volume and vascular lesions on MRI in a memory clinic population. J. Diabetes Complications 28, 85–90 (2014). 68. Knopman, D. S. et al. Vascular risk factors and longitudinal changes on brain MRI: the ARIC study. Neurology 76, 1879–1885 (2011). 69. van Dijk, E. J. et al. Plasma amyloid β, apolipoprotein E, lacunar infarcts, and white matter lesions. Ann. Neurol. 55, 570–575 (2004). 70. Vinters, H. V., Secor, D. L., Pardridge, W. M. & Gray, F. Immunohistochemical study of cerebral amyloid angiopathy. III. Widespread Alzheimer A4 peptide in cerebral microvessel walls colocalizes with gamma trace in patients with leukoencephalopathy. Ann. Neurol. 28, 34–42 (1990). 71. Moody, D. M., Brown, W. R., Challa, V. R. & Anderson, R. L. Periventricular venous collagenosis: association with leukoaraiosis. Radiology 194, 469–476 (1995). 72. Englund, E. Neuropathology of white matter changes in Alzheimer’s disease and vascular dementia. Dement. Geriatr. Cogn. Disord. 9 (Suppl. 1), 6–12 (1998). 73. Richardson, K. et al. The neuropathology of vascular disease in the Medical Research Council Cognitive Function and Ageing Study (MRC CFAS). Curr. Alzheimer Res. 9, 687–696 (2012). 74. Nicoll, J. A. et al. Association between APOE genotype, neuropathology and dementia in the older population of England and Wales. Neuropathol. Appl. Neurobiol. 37, 285–294 (2011). 75. Noh, Y. et al. White matter hyperintensities are associated with amyloid burden in APOE4 noncarriers. J. Alzheimers Dis. 40, 877–886 (2014). 76. Chao, L. L. et al. Associations between white matter hyperintensities and β amyloid on integrity of projection, association, and limbic fiber tracts measured with diffusion tensor MRI. PLoS ONE 8, e65175 (2013). 77. Kim, H. J. et al. The effects of small vessel disease and amyloid burden on neuropsychiatric symptoms: a study among patients with subcortical vascular cognitive impairments. Neurobiol. Aging 34, 1913–1920 (2013). 78. Barnes, J. et al. Vascular and Alzheimer’s disease markers independently predict brain atrophy rate in Alzheimer’s Disease Neuroimaging Initiative controls. Neurobiol. Aging 34, 1996–2002 (2013). 79. Guzman, V. A. et al. White matter hyperintensities and amyloid are independently associated with entorhinal cortex volume among individuals with mild cognitive impairment. Alzheimers Dement. 9 (5 Suppl.), S124–S131 (2013). 80. Chabriat, H. et al. Patterns of MRI lesions in CADASIL. Neurology 51, 452–457 (1998). 81. Tikka, S. et al. CADASIL and CARASIL. Brain Pathol. 24, 525–544 (2014). 82. Bianchi, S. et al. Two novel HTRA1 mutations in a European CARASIL patient. Neurology 82, 898–900 (2014).

www.nature.com/nrneurol © 2015 Macmillan Publishers Limited. All rights reserved

REVIEWS 83. Bersano, A. et al. The genetics of small-vessel disease. Curr. Med. Chem. 19, 4124–4141 (2012). 84. Fornage, M. et al. Genome-wide association studies of cerebral white matter lesion burden: the CHARGE consortium. Ann. Neurol. 69, 928–939 (2011). 85. Adib-Samii, P. et al. 17q25 locus is associated with white matter hyperintensity volume in ischemic stroke, but not with lacunar stroke status. Stroke 44, 1609–1615 (2013). 86. Tabara, Y. et al. Association of Chr17q25 with cerebral white matter hyperintensities and cognitive impairment: the J‑SHIPP study. Eur. J. Neurol. 20, 860–862 (2013). 87. Viswanathan, A. & Greenberg, S. M. Cerebral amyloid angiopathy in the elderly. Ann. Neurol. 70, 871–880 (2011). 88. Vinters, H. V. et al. Secondary microvascular degeneration in amyloid angiopathy of patients with hereditary cerebral hemorrhage with amyloidosis, Dutch type (HCHWA‑D). Acta Neuropathol. 95, 235–244 (1998). 89. McDonald, W. I. et al. Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis. Ann. Neurol. 50, 121–127 (2001). 90. van der Knaap, M. S., Breiter, S. N., Naidu, S., Hart, A. A. & Valk, J. Defining and categorizing leukoencephalopathies of unknown origin: MR imaging approach. Radiology 213, 121–133 (1999). 91. Yoshida, T. et al. MELAS and reversible vasoconstriction of the major cerebral arteries. Intern. Med. 52, 1389–1392 (2013). 92. de Groot, J. C. et al. Cerebral white matter lesions and cognitive function: the Rotterdam Scan Study. Ann. Neurol. 47, 145–151 (2000). 93. Kramer, J. H., Reed, B. R., Mungas, D., Weiner, M. W. & Chui, H. C. Executive dysfunction in subcortical ischaemic vascular disease. J. Neurol. Neurosurg. Psychiatry 72, 217–220 (2002). 94. Mungas, D. et al. MRI predictors of cognition in subcortical ischemic vascular disease and Alzheimer’s disease. Neurology 57, 2229–2235 (2001). 95. Carmichael, O. et al. Longitudinal changes in white matter disease and cognition in the first year of the Alzheimer disease neuroimaging initiative. Arch. Neurol. 67, 1370–1378 (2010). 96. Longstreth, W. T., Diehr, P. H., Yee, L. M., Newman, A. B. & Beauchamp, N. J. Brain imaging findings in elderly adults and years of life, healthy life, and able life over the ensuing 16 years: the Cardiovascular Health Study. J. Am. Geriatr. Soc. 62, 1838–1843 (2014). 97. LADIS Study Group. 2001–2011: a decade of the LADIS (Leukoaraiosis And DISability) Study: what have we learned about white matter changes and small-vessel disease? Cerebrovasc. Dis. 32, 577–588 (2011). 98. Prins, N. D. et al. Cerebral white matter lesions and the risk of dementia. Arch. Neurol. 61, 1531–1534 (2004). 99. Román, G. C. & Kalaria, R. N. Vascular determinants of cholinergic deficits in Alzheimer

disease and vascular dementia. Neurobiol. Aging 27, 1769–1785 (2006). 100. Kimura, S. et al. Pathogenesis of vascular dementia in stroke-prone spontaneously hypertensive rats. Toxicology 153, 167–178 (2000). 101. Strachan, M. W., Reynolds, R. M., Marioni, R. E. & Price, J. F. Cognitive function, dementia and type 2 diabetes mellitus in the elderly. Nat. Rev. Endocrinol. 7, 108–114 (2011). 102. Sachdev, P. S. et al. Homocysteine as a risk factor for cognitive impairment in stroke patients. Dement. Geriatr. Cogn. Disord. 15, 155–162 (2003). 103. Breteler, M. M. Vascular risk factors for Alzheimer’s disease: an epidemiologic perspective. Neurobiol. Aging 21, 153–160 (2000). 104. Kalaria, R. N. Small vessel disease and Alzheimer’s dementia: pathological considerations. Cerebrovasc. Dis. 13 (Suppl. 2), 48–52 (2002). 105. Hainsworth, A. H., Brittain, J. F. & Khatun, H. Preclinical models of human cerebral small vessel disease: utility for clinical application. J. Neurol. Sci. 322, 237–240 (2012). 106. Bink, D. I., Ritz, K., Aronica, E., van der Weerd, L. & Daemen, M. J. Mouse models to study the effect of cardiovascular risk factors on brain structure and cognition. J. Cereb. Blood Flow Metab. 33, 1666–1684 (2013). 107. van der Flier, W. M. et al. Medial temporal lobe atrophy and white matter hyperintensities are associated with mild cognitive deficits in nondisabled elderly people: the LADIS study. J. Neurol. Neurosurg. Psychiatry 76, 1497–1500 (2005). 108. Jang, J. W. et al. Effect of white matter hyperintensity on medial temporal lobe atrophy in Alzheimer’s disease. Eur. Neurol. 69, 229–235 (2013). 109. Erkinjuntti, T. et al. Do white matter changes on MRI and CT differentiate vascular dementia from Alzheimer’s disease? J. Neurol. Neurosurg. Psychiatry 50, 37–42 (1987). 110. Price, C. C., Jefferson, A. L., Merino, J. G., Heilman, K. M. & Libon, D. J. Subcortical vascular dementia: integrating neuropsychological and neuroradiologic data. Neurology 65, 376–382 (2005). 111. DeCarli, C. et al. The effect of white matter hyperintensity volume on brain structure, cognitive performance, and cerebral metabolism of glucose in 51 healthy adults. Neurology 45, 2077–2084 (1995). 112. Brickman, A. M. et al. White matter hyperintensities and cognition: testing the reserve hypothesis. Neurobiol. Aging 32, 1588–1598 (2011). 113. Murray, A. D. et al. The balance between cognitive reserve and brain imaging biomarkers of cerebrovascular and Alzheimer’s diseases. Brain 134, 3687–3696 (2011). 114. Skoog, I., Kalaria, R. N. & Breteler, M. M. Vascular factors and Alzheimer disease. Alzheimer Dis. Assoc. Disord. 13 (Suppl. 3), S106–S114 (1999). 115. Forette, F. et al. Prevention of dementia in randomised double-blind placebo-controlled

NATURE REVIEWS | NEUROLOGY

Systolic Hypertension in Europe (Syst-Eur) trial. Lancet 352, 1347–1351 (1998). 116. Williamson, J. D. et al. Cognitive function and brain structure in persons with type 2 diabetes mellitus after intensive lowering of blood pressure and lipid levels: a randomized clinical trial. JAMA Intern. Med. 174, 324–333 (2014). 117. ten Dam, V. H. et al. Effect of pravastatin on cerebral infarcts and white matter lesions. Neurology 64, 1807–1809 (2005). 118. Richard, E., Gouw, A. A., Scheltens, P. & van Gool, W. A. Vascular care in patients with Alzheimer disease with cerebrovascular lesions slows progression of white matter lesions on MRI: the evaluation of vascular care in Alzheimer’s disease (EVA) study. Stroke 41, 554–556 (2010). 119. Richard, E., Kuiper, R., Dijkgraaf, M. G., Van Gool, W. A. & Evaluation of Vascular care in Alzheimer’s disease. Vascular care in patients with Alzheimer’s disease with cerebrovascular lesions—a randomized clinical trial. J. Am. Geriatr. Soc. 57, 797–805 (2009). 120. Richard, E. et al. Prevention of dementia by intensive vascular care (PreDIVA): a clusterrandomized trial in progress. Alzheimer Dis. Assoc. Disord. 23, 198–204 (2009). 121. Cyarto, E. V. et al. Protocol for a randomized controlled trial evaluating the effect of physical activity on delaying the progression of white matter changes on MRI in older adults with memory complaints and mild cognitive impairment: the AIBL Active trial. BMC Psychiatry 12, 167 (2012). 122. Román, G. C. et al. Donepezil in vascular dementia: combined analysis of two large-scale clinical trials. Dement. Geriatr. Cogn. Disord. 20, 338–344 (2005). 123. Wilkinson, D. et al. The long-term efficacy and tolerability of donepezil in patients with vascular dementia. Int. J. Geriatr. Psychiatry 25, 305–313 (2010). 124. Erkinjuntti, T. et al. Efficacy of galantamine in probable vascular dementia and Alzheimer’s disease combined with cerebrovascular disease: a randomised trial. Lancet 359, 1283–1290 (2002). 125. Orgogozo, J. M., Rigaud, A. S., Stöffler, A., Möbius, H. J. & Forette, F. Efficacy and safety of memantine in patients with mild to moderate vascular dementia: a randomized, placebocontrolled trial (MMM 300). Stroke 33, 1834–1839 (2002). 126. Inzitari, D. et al. Changes in white matter as determinant of global functional decline in older independent outpatients: three year follow-up of LADIS (leukoaraiosis and disability) study cohort. BMJ 339, b2477 (2009). 127. de Leeuw, F. E., Barkhof, F. & Scheltens, P. Progression of cerebral white matter lesions in Alzheimer’s disease: a new window for therapy? J. Neurol. Neurosurg. Psychiatry 76, 1286–1288 (2005). Author contributions Both authors researched data for the article, discussed the content, wrote the article, and reviewed and edited the manuscript before submission.

ADVANCE ONLINE PUBLICATION  |  9 © 2015 Macmillan Publishers Limited. All rights reserved