Introduction
‘Cerebral small vessel disease’ is an umbrella term that is used to describe a host of features present on brain imaging and a number of pathological processes occurring within the brain. Small vessel disease (SVD) has important clinical sequelae and may present as lacunar stroke, progressive cognitive decline, gait disorder, or frank dementia.Reference Wardlaw1 In life SVD is commonly diagnosed on the basis of brain imaging findings. Discrete lacunar infarcts and confluent white matter lesions (WML) are the most frequently described imaging surrogates of SVD.Reference Wardlaw1 With the advent of widespread magnetic resonance imaging (MRI) scanning, the presence of microhaemorrhages is also recognized to be a feature of SVD. To date this has been less well studied and the associations with clinical features are less clear.
The reported prevalence of these findings varies across the literature, depending on the population studied and the methods used. Several cross-sectional studies have demonstrated associations between imaging-defined SVD and specific clinical and cognitive characteristics such as mood changes, impairment in activities of daily living, gait difficulties, falls, global cognitive decline and deficits in executive functions.Reference Pohjasvaara, Mantyla, Ylikoski, Kaste and Erkinjuntti2–Reference Jokinen, Kalska, Mäntylä, Pohjasvaara, Ylikoski, Hietanen, Salonen, Kaste and Erkinjuntti4 More recently, a longitudinal study has demonstrated that progression of WML is associated with declining cognition and an increased risk of dementia.Reference Jokinen, Kalska, Ylikoski, Madureira, Verdelho, van der Flier, Scheltens, Barkhof, Visser, Fazekas, Schmidt, O'Brien, Waldemar, Wallin, Chabriat, Pantoni, Inzitari and Erkinjuntti5
Lacunar infarcts, long held to be the most ‘benign’ of stroke syndromes due to lower immediate mortality and functional disability compared with large artery stroke, may also be associated with poorer long-term outcomes, both in terms of functional ability and cognition, than previously understood.Reference Jokinen, Gouw, Madureira, Ylikoski, van Straaten, van der Flier, Barkhof, Scheltens, Fazekas, Schmidt, Verdelho, Ferro, Pantoni, Inzitari and Erkinjuntti6, Reference Melkas, Putaala, Oksala, Pohjasvaara, Oksala, Kaste, Karhunen and Erkinjuntti7 The prevalence of SVD changes on imaging increases with age;Reference van Dijk, Prins, Vrooman, Hofman, Koudstaal and Breteler8, Reference Vermeer, Longstreth and Koudstaal9 thus advancing age is a major risk factor for the development of SVD and a major cause of morbidity in older adults.
A number of distinct pathological processes have been identified within the brains of individuals with SVD, the most common of which is arteriosclerosis, or age and vascular risk factor-related SVD.Reference Pantoni10 This is characterized by loss of smooth muscle in the tunica media, with deposits of fibro-hyaline material within the vessel wall leading to luminal narrowing and/or occlusion. There remains, however, a lack of animal models of SVD and so the precise mechanisms by which these changes within the vasculature cause parenchymal damage remain unclear. Other forms of cerebral microangiopathy are also recognized, with cerebral amyloid angiopathy (CAA) and cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) being amongst the most frequently described.
This review will discuss the pathology and clinical presentations of SVD, the neuroimaging findings associated with SVD, and current evidence with regard to treatment and prevention. A literature search was performed using Medline, Embase and CINAHL databases between 1996 and October 2011. The search terms ‘elderly’ and ‘older adult’ were combined with the terms ‘small vessel cerebrovascular disease’, ‘white matter hyperintensities’, ‘neuroimaging’ and ‘cognitive impairment’. Further articles were obtained by searching reference lists. Only English language papers were included.
Pathology
Arteriosclerosis
The most prevalent pathological process identified in SVD is arteriosclerosis, or age and vascular risk factor-related SVD.Reference Pantoni10 This is characterized by fibro-hyaline deposition within the walls of the end arteries and arterioles that supply the deep matter of the brain. This process results in thickening of the vessel wall and subsequent narrowing of the vessel lumen. Conventional teaching has been that this then leads to chronic hypoperfusion of the brain parenchyma, with ischaemic demyelination and low-grade infarction occurring as a result.Reference Jellinger11 More recently, however, some researchers have proposed alternative hypotheses as to the pathogenesis of cerebral SVD, such as endothelial dysfunction, inflammation and breakdown of the blood–brain barrier.Reference Wardlaw1, Reference Pantoni10 As yet, the exact relationship between the microvascular abnormalities seen in SVD and the associated parenchymal damage remain unclear.
It is increasingly recognized that arteriosclerosis is likely to be a systemic disease of microvasculature throughout the body, rather than being confined solely to the cerebral vasculature.Reference Thompson and Hakim12 In particular, a number of studies have demonstrated associations between retinal microvascular disease,Reference Wong, Klein, Sharrett, Nieto, Boland, Couper, Mosley, Klein, Hubbard and Szklo13, Reference Wong, Klein, Sharrett, Couper, Klein, Liao, Hubbard and Mosley14 renal impairmentReference Khatri, Wright, Nickolas, Yoshita, Paik, Kranwinkel, Sacco and DeCarli15, Reference Wada, Nagasawa, Iseki, Takahashi, Sato, Arawaka, Kawanami, Kurita, Daimon and Kato16 and cerebral SVD. Cerebral SVD also shares a risk factor profile with large vessel vascular disease, with the strongest association being with systolic hypertension.
Cerebral amyloid angiopathy
Cerebral amyloid angiopathy (CAA) is characterized by the progressive accumulation of congophilic amyloid beta protein within the walls of small to medium-sized arterioles and arteries. Amyloid deposition and the development of neurofibrillary plaques and tangles are the pathological hallmark of Alzheimer's disease (AD), but amyloid deposition is also recognized to occur in the brains of the general elderly population and its prevalence increases with age.Reference Pantoni10 Clinically CAA may be associated with large lobar haemorrhagesReference Smith and Eichler17 but, in imaging studies, it has also been noted to co-exist with features of SVD, such as WML, microhaemorrhages and infarction.Reference Pantoni10 Thus although these two forms of microangiopathy may have differing pathological characteristics, the prevalence of both seems to increase with advancing age, and both may co-exist in the same patients.
Interaction between SVD and AD pathology
In the past, dementia due to AD and vascular dementia were held to be clinically distinct entities. Since the late 1990s, however, a radical revision of this concept has occurred and it is now recognized that AD and ischaemic cerebrovascular disease share a number of risk factors and clinical features and that the two processes may co-exist and interact in a synergistic fashion in patients with dementia.
The first study to illustrate a clinical interaction between Alzheimer pathology and cerebrovascular disease was the Nun Study, in which serial cognitive assessments were undertaken of 678 nuns in later life in the mid-west USA and the results related to brain pathology at post-mortem.Reference Snowdon, Greiner, Mortimer, Riley, Greiner and Markesbery18 Deep lacunar infarcts at post-mortem were associated with an odds ratio for dementia of 20.7. However, vascular disease on its own was only weakly associated with poor cognitive function and dementia, requiring co-existent AD pathology for major cognitive impairment to be evident. When lacunar infarction was present, fewer neuropathological lesions of AD were associated with dementia than in those without infarcts. However, nuns are an atypical group, likely to have a lifestyle associated with lower levels of vascular disease than the general population. The true burden of small vessel cerebrovascular disease is probably better shown in the MRC Cognitive Function Ageing Study (CFAS), in which an epidemiological cohort of older subjects was followed to post-mortem.19 There was a large overlap in AD pathology in those who died with dementia compared with non-demented subjects. Small vessel cerebrovascular disease was common in both groups, but with an excess in those with dementia. It was concluded that the combination of SVD and AD was particularly important in the expression of dementia in later life. These post-mortem studies provide convincing evidence for the common interaction of AD pathology with SVD in determining whether a patient develops dementia in later life.
Neuroimaging
In neuroimaging, the term SVD is often (misleadingly) used to refer only to the ischaemic parenchymal consequences of the underlying pathology. As we have discussed, however, the consequences of SVD are heterogeneous, encompassing both ischaemic (lacunar infarcts, WML) and haemorrhagic (microhaemorrhages) lesions. The main imaging techniques used in routine clinical practice to identify cerebral SVD are computed tomography (CT) and MRI.
CT scanning
The typical CT features of SVD are patchy hypodensities present in the periventricular and subcortical areas (leukoariosis). Leukoariosis on CT is felt to be an infrequent finding in a healthy elderly population (compared with WML on MRI, which is a common finding in epidemiological studies of elderly populations; presumably this reflects the increased sensitivity of MRI for detecting these lesions), but is present in a majority of patients who are felt to have a clinically significant vascular component to their dementia.Reference Garcia and Brown20 Leukoariosis is also associated with an increased prevalence of neurological impairments, including gait disorder, lower limb weakness, upgoing plantars and primitive reflexes.Reference Steingart, Hachinski, Lau, Fox, Diaz, Cape, Lee, Inzitari and Merskey21 Post-mortem studies have confirmed that CT findings of SVD correlate reasonably well with pathological findings suggestive of SVD at autopsy,Reference Rossi, Joachim, Geroldi, Combrinck, Esiri, Smith and Frisoni22 and demyelination, reactive gliosis and arteriosclerosis have all been noted to occur in proportion to the degree of leukoariosis seen on CT. Lacunar infarcts may also be seen on CT, but are more frequently appreciated on MRI scanning.
MRI
On MRI scanning, SVD may be identified in a number of ways. WML are bilateral, symmetrical and often confluent lesions affecting the periventricular and subcortical white matter, which appear hyperintense on T2-weighted and FLAIR sequences. Lacunar infarcts may be identified in the acute phase as hyperintense areas on DWI sequences, or as hypointense areas on T1-weighted sequences.Reference Gouw, Seewann, van der Flier, Barkhof, Rozemuller, Scheltens and Geurts23 Lacunar infarcts are most typically found in the basal ganglia, thalamus, pons and internal capsule. Microhaemorrhages have more recently been identified as a signifier of SVD, and can be appreciated on gradient echo sequences.
Early MRI studies noted white matter abnormalities in up to a third of elderly patients,Reference Hunt, Orrison, Yeo, Haaland, Rhyne, Garry and Rosenberg24 and in 2005 the Cardiovascular Health Study identified white matter abnormalities in over 80% of an epidemiological cohort of elderly subjects.Reference Longstreth, Arnold, Beauchamp, Manolio, Lefkowitz, Jungreis, Hirsch, O'Leary and Furberg25 While in the past the clinical significance of white matter changes on MRI was debated, with some suggesting that it was a variant of normal ageing, there is now an extensive body of work which confirms that MRI-defined white matter abnormalities are associated with cognitive decline and dementia, functional disability, gait disorders and mood disorders.Reference Pohjasvaara, Mantyla, Ylikoski, Kaste and Erkinjuntti2–Reference Jokinen, Kalska, Ylikoski, Madureira, Verdelho, van der Flier, Scheltens, Barkhof, Visser, Fazekas, Schmidt, O'Brien, Waldemar, Wallin, Chabriat, Pantoni, Inzitari and Erkinjuntti5 An exhaustive review of the clinical correlations of white matter changes on MRI are beyond the scope of this review; interested readers may wish to refer to the detailed review of this subject recently published by Patel and Markus.Reference Patel and Markus26
While the associations between white matter changes and poorer long-term functional and cognitive outcomes are now well recognized, the longer term effects of lacunar infarction have been less well studied. In the past, lacunar infarction has been considered a relatively benign stroke subtype, largely due to the lower immediate mortality and disability associated with lacunar stroke in comparison with large vessel stroke. A recent ultra-long follow-up trial, however, has suggested that stroke attributable to SVD (as defined by the presence of a clinical lacunar syndrome, imaging findings and the presence of risk factors) may be associated with a poorer long-term prognosis than other stroke subtypes.Reference Melkas, Putaala, Oksala, Pohjasvaara, Oksala, Kaste, Karhunen and Erkinjuntti7 The SVD-labelled cohort of this trial, however, was small in numbers, so results should be interpreted with caution. ‘Silent’ lacunar infarction (i.e. the presence of imaging changes in keeping with infarction in the absence of a history of a clinical stroke event) may be identified in up to 20% of the healthy elderly population and in up to 50% of selected series.Reference Vermeer, Longstreth and Koudstaal9 The presence of lacunar infarcts on MRI is associated with an increased risk of future stroke, cognitive decline and dementia,Reference Vermeer, Longstreth and Koudstaal9 though it is not yet known if treating individuals with evidence of silent infarction on MRI reduces the risk of subsequent events.
Microhaemorrhages appear on MRI as small (5–10 mm), well demarcated, round lesions on gradient echo sequences. They are commonly seen in a number of cerebrovascular diseases, including ischaemic stroke, haemorrhagic stroke, AD and vascular dementia.Reference Werring, Gregoire and Cipolotti27 It has been hypothesized (but not proven) that the distribution of microhaemorrhages may relate to the underlying vessel pathology, with ‘deep’ lesions being associated with arteriosclerosis and more superficial lesions being associated with CAA or AD pathology. Unlike the better-understood WML, the associations between microhaemorrhages and cognitive impairment/dementia are not currently well understood. A recent publication, however, has suggested that cerebral microhaemorrhages on MRI may be associated with poorer outcomes in elderly patients.Reference Altmann-Schneider, Trompet, de Craen, van Es, Jukema, Stott, Sattar, Westendorp, van Buchem and van der Grond28 This substudy of the PROSPER trial found that elderly individuals with >1 microhaemorrhage on MRI had a 6-fold increase in risk of stroke-related death, and a 2-fold increase in risk of cardiovascular death. These findings suggest that cerebral microhaemorrhages are likely to be of clinical significance, but further research in this area is needed to fully elucidate the clinical correlations of these lesions.
Novel MRI techniques may, in the near future, shed further light on the issue of white matter changes in SVD. Diffusion tensor imaging is a new technique that promises to provide more information regarding the microstructure of white matter within the brain and functional MRI may provide information regarding connectivity between brain regions. The ongoing RUN-DMC studyReference van Norden, de Laat, Gons, van Uden, van Dijk, van Oudheusden, Esselink, Bloem, van Engelen, Zwarts, Tendolkar, Olde-Rikkert, van der Vlugt, Zwiers, Norris and de Leeuw29 will use these novel techniques in a prospective cohort of healthy elderly subjects to assess the effects of SVD on cognition and function.
Clinical symptoms and signs
Cognitive impairment
Research investigating the relationship between SVD and cognitive impairment/dementia has been complicated by a lack of consensus regarding definitions and diagnostic criteria for both the imaging surrogates of SVD and the resulting cognitive sequelae. The definition of what constitutes ‘cognitive impairment’ may vary considerably, as do definitions of dementia; the proportion of a population labelled as having dementia can vary up to tenfold depending on the criteria used.Reference Erkinjuntti, Ostbye, Steenhuis and Hachinski30 In addition, there are no widely applied standards as to what defines SVD in imaging studies; many of the studies use WML load as indicators of degree and progression of SVD, but methods used to quantify WML severity may be subjective (such as rating scales based on visual assessment) and inconsistent across studies.
The classical concept of vascular dementia is that it has an abrupt onset and that the patient shows stepwise deterioration in cognitive function. However, it is now appreciated that many patients with vascular dementia will present insidiously and decline progressively. The cognitive domains that appear to be most affected in vascular dementia are executive function and speed of information processing,Reference Price, Jefferson, Merino, Heilman and Libon31, Reference Reed, Mungas, Kramer, Ellis, Vinters, Zarow, Jagust and Chui32 while memory may be relatively well preserved, at least in the early stages. As such, these patients may have significant cognitive impairment, often resulting in significant limitations of function, but without meeting criteria for a diagnosis of dementia (as these criteria are often weighted towards significant disturbance of memory).
Attempts to determine the aetiology of dementia clinically are often of limited benefit, but applying detailed neuropsychological tests can identify patterns of cognitive dysfunction that may be used to attempt to differentiate vascular dementia from dementia due to other pathologies.
Diagnostic criteria for vascular cognitive impairment
The most commonly used diagnostic criteria for vascular dementia in research studies are the DSM-IV criteria, the ICD-10 criteria and the NINDS-AIREN criteria.Reference Román, Tatemichi, Erkinjuntti, Cummings, Masdeu, Garcia, Amaducci, Orgogozo, Brun and Hofman33 These all vary in how dementia is defined, what types of cerebrovascular diseases are included, what other disorders are excluded and the neuroimaging correlates needed to make the diagnosis.
While making clear distinctions between vascular dementia and dementia attributable to other pathologies (notably AD) is of course essential in clinical trials, in clinical practice it is of more relevance to identify if there is likely to be a vascular component to a patient's cognitive impairment/dementia, rather than seeking to apply rigid diagnostic criteria. As such, the National Institute of Neurological Disorders and Stroke – Canadian Stroke Network Vascular Cognitive Impairment Harmonization Standards have suggested the use of the term ‘vascular cognitive impairment’ (VCI) to encompass ‘cognitive impairment that is caused by or associated with vascular factors’.Reference Hachinski, Iadecola, Petersen, Breteler, Nyenhuis, Black, Powers, DeCarli, Merino, Kalaria, Vinters, Holtzman, Rosenberg, Wallin, Dichgans, Marler and Leblanc34
Gait and mobility disorder
The classic gait disorder in SVD is short stepped and shuffling (marche a petits pas/gait apraxia), with the arms held abducted and the elbows flexed. The arms swing normally from the shoulders, the base tends to be normal width and there is no festination, propulsion, excessive hip or knee flexion, or tendency to sway unduly on stopping.Reference Ishii, Nishihara and Imamura35–Reference Nutt, Marsden and Thompson37 This condition may also be described as vascular or arteriosclerotic Parkinsonism.Reference Critchley36 It has been hypothesized that problems of initiation of gait may be associated with damage to the supplementary motor cortex and its connections, whereas equilibrium apraxia may be due to damage to the pre-motor cortex and connections.Reference Liston, Mickelborough, Bene and Tallis38 Transferring out of a chair or bed and descending stairs may be particularly difficult. The head, neck and trunk often move in one piece, with associated postural instability and a tendency to fall backwards. However, gait ataxia, with a tendency to fall backwards, may also be seen.Reference Bazner, Oster, Daffertshofer and Hennerici39 In a subject with cognitive impairment, associated features of early gait disturbance, unsteadiness and frequent unprovoked falls are highly suggestive of a significant vascular component.
Parkinsonian features, particularly bradykinesia, gait disorder and tremor, are found in association with evidence of cerebrovascular disease on CT, more so with lacunar infarcts than territorial strokes, and in association with WML on neuroimaging.Reference Van Zagten, Lodder and Kessels40
Muscle tone, power, reflexes and co-ordination
Hypertonicity is often present, usually of lead pipe character and sometimes asymmetrical. The lower limbs are more often affected than the arms, neck or trunk, and proximal muscles affected more than distal.
When weakness occurs, it tends to be in the same pattern as the hypertonicity described above. Difficulty in walking is often out of proportion to muscular weakness. In the late stages, when severely affected patients may become bedridden, there may be adductor spasm and a cerebral diplegia.Reference Ishii, Nishihara and Imamura35, Reference Critchley36 Limb reflexes may be brisk and/or asymmetrical. Crossed adductor responses may develop and plantars may be extensor. Some patients may develop primitive reflexes.Reference Ishii, Nishihara and Imamura35, Reference Critchley36
Many patients with SVD have co-ordination difficulties. Difficulty with fine movements may occur before reduction in muscular power is evident.Reference Villardita41 Rapid alternating movements tend to be slow and irregular.
Speech and swallowing
The voice is often quiet and many patients are dysarthric.Reference Ishii, Nishihara and Imamura35 Verbal fluency, category fluency and word list generation are reduced.Reference Lafosse, Reed, Mungas, Sterling and Jagust42, Reference Carew, Lamar, Cloud, Grossman and Libon43 Clinically significant dysphagia is uncommon, but many patients will report minor swallowing difficulties. Severe dysphagia may occur, and is almost always associated with dysarthria as a result of pseudobulbar palsy.
Urinary symptoms
Urinary incontinence, urgency and frequency are common in patients with SVD.Reference Ishii, Nishihara and Imamura35 Detrusor instability is traditionally believed to be the main aetiology, although urodynamic findings are highly variable in incontinent patients with clinical stroke or dementia.Reference Gelber, Good, Laven and Velhurst44, Reference Resnick, Yalla and Laurino45 It is likely that a wide spectrum of bladder dysfunction is present in incontinent patients with SVD.
Mood disturbance
Approximately 20% of patients with vascular dementia develop emotional lability.Reference Ishii, Nishihara and Imamura35 When present, this is highly suggestive of underlying cerebrovascular disease. Diurnal restlessness is more common in clinically diagnosed dementia due to vascular disease than in AD.Reference Molska, Paljarvi, Rinne, Rinne and Salko46 Depression is common and remains under-recognized in the post-stroke population. WML on MRI have been increasingly associated with late-onset depression.Reference Hickie and Scott47 Apathy, with slowing of cognition and motor function, may be present in patients with SVDReference Babikian and Ropper48 and care must be taken not to misinterpret this, or indeed an underlying mood disorder, as lack of motivation.
Risk factors for SVD
Unsurprisingly, SVD shares a similar risk factor profile with large vessel vascular disease. Hypertension is a major risk factor for all forms of ischaemic cerebrovascular damageReference Skoog, Lernfelt, Landahl, Palmertz, Andreasson, Nilsson, Persson, Odén and Svanborg49 and the vast majority of patients with severe SVD have a history of systolic hypertension.Reference Babikian and Ropper48, Reference Tohgi, Chiba and Kiamura50 Elevated nocturnal blood pressure (BP) and increased BP variability are also associated with SVD.
Other risk factors include raised cholesterol, diabetes, metabolic syndrome and smoking.Reference Zimerbaum, Frishman and Aronson51–Reference Ott, Slooter, Hofman, van Harskamp, Witteman, Van Broeckhoven, van Duijn and Breteler53 Systemic inflammation may also be a risk factor.Reference Schmidt, Schmidt, Curb, Masaki, White and Launer54
Alzheimer pathology and SVD share much more than just the clinical features of progressive cognitive decline. The two pathologies share numerous risk factors, including genetics (apolioprotein E4 status), hypertension, raised cholesterol, diabetes, metabolic syndrome, obesity, smoking, poor socioeconomic status and low levels of physical activity amongst others. In view of the shared risk factor profile and the synergistic effects of AD pathology and vascular pathology on cognition, some commentators have gone so far as to suggest that AD may in fact be a primary vasculopathy, though this remains putative at present.
Therapeutic aspects
Studies focusing specifically on the treatment and prevention of SVD are lacking. There is nothing to suggest that small vessel ischaemic events do not benefit from evidence-based therapies in the acute setting, including aspirin, thrombolysis and stroke unit care. However, the evidence for treatment of vascular risk factors and the prevention of cognitive decline in the longer term is less clear.
Antihypertensives
As mentioned above, systolic hypertension is one of the major risk factors for cerebral SVD, but the evidence for treatment of hypertension as a measure to reduce cognitive decline or dementia is conflicting. There is some initial trial evidence suggesting that treatment of hypertension prevents WML progression and associated cognitive decline, particularly in those with previous cerebrovascular events.Reference Godin, Tzourio, Malliard, Mazoyer and Dofouil55 The PROGRESS trial reported that a combination of ACE inhibitor and thiazide diuretic was associated with a 12% reduction in the incidence of dementia and a 34% reduction in the incidence of recurrent stroke and cognitive decline.Reference Tzourio, Anderson, Chapman, Woodward, Neal, MacMahon and Chalmers56 The Syst-Eur trial reported a 50% decrease in the incidence of dementia from 7.7 to 3.8 cases per 1000 patient-years in elderly patients treated with the calcium channel blocker nitrendipine,Reference Forette, Seux, Staessen, Thijs, Babarskiene, Babeanu, Bossini, Fagard, Gil-Extremera, Laks, Kobalava, Sarti, Tuomilehto, Vanhanen, Webster, Yodfat and Birkenhäger57 and a further trial suggested that nimodipine was associated with less deterioration in mini mental state examination (MMSE) score and executive function.Reference Pantoni, del Ser, Soglian, Amigoni, Spadari, Binelli and Inzitari58
In 2007, however, a meta-analysis of 12,091 patients from three large-scale antihypertensive trials (including the Syst-Eur trial) concluded that there was no evidence that BP lowering with antihypertensives prevented cognitive decline or dementia in elderly hypertensive patients without prior cerebrovascular disease.Reference McGuinness, Todd, Passmore and Bullock59 Furthermore, a large population-based cohort study has investigated the relationship between blood pressure and cognitive function in individuals aged between 55 and 88 years of age.Reference Euser, van Bemmel, Schram, Gussekloo, Hofman, Westendorp and Breteler60 This study has suggested that, in patients aged 65–74 years, higher blood pressure is associated with poorer cognitive function at follow-up, but in older patients (i.e. age >75 years) this relationship is reversed, with the strongest association between higher blood pressure and better cognitive function demonstrated in the ‘oldest old’ (age >85 years). This suggests that, in elderly and very elderly patients, aggressive blood pressure lowering may not always be desirable, and may, in fact, be associated with the potential for harm.
In summary, treatment of hypertension remains important in preventing large vessel cerebrovascular disease, cardiovascular disease and renovascular disease. Further studies are needed to clarify whether antihypertensive strategies play a role in the management of SVD, or whether indeed age-specific guidelines for the management of hypertension are necessary.
Statins
The stroke prevention by aggressive reduction of cholesterol levels (SPARCL) study suggests that patients with SVD and increased low density lipoprotein benefit from statin therapy to reduce stroke risk.Reference Amarenco, Benavente, Goldstein, Callahan, Sillesen, Hennerici, Gilbert, Rudolph, Simunovic, Zivin and Welch61 However, the effects of statins on cognitive decline or progression of white matter lesions in older adults is less clear. Observational studies have demonstrated that statins may reduce the risk of developing dementia, but randomized controlled trials are lacking.
The PROSPER study showed that pravastatin had no effect on cognitive function, including speed of information processing or verbal memory, over a 3.2 year period.Reference Houx, Shepherd, Blauw, Murphy, Ford, Bollen, Buckley, Stott, Jukema, Hyland, Gaw, Norrie, Kamper, Perry, MacFarlane, Meinders, Sweeney, Packard, Twomey, Cobbe and Westendorp62 Recruited patients were generally cognitively intact and results cannot be extrapolated to treatment of mild cognitive impairment or mild dementia. We are also unable to predict longer term clinical effects from this study. The Cardiovascular Health Study observed participants over a longer period (7 years).Reference Bernick, Katz, Smith, Rapp, Bhadelia, Carlson and Kuller63 Patients treated with statins were observed to have less cognitive decline than untreated subjects. The significant cognitive benefit was associated with a reduction in progression of cerebral infarcts in the treatment group, but no statistically significant difference between groups in white matter change progression.
Anti-platelet drugs
Data can be derived from subgroup analysis of several trials that anti-platelets reduce recurrence rates of ischaemic events.Reference Bousser, Eschwege, Haguenau, Lefaucconnier, Thibult, Touboul and Touboul64, 65 However, data regarding the prevention of cognitive decline in patients with small vessel disease is limited. One small, randomized controlled trial of aspirin found benefits on cognition in subjects with cerebrovascular disease and dementia.Reference Meyer, Rogers, McClintic, Mortel and Lotfi66
More recently, the combination of aspirin and dipyridamole was compared with clopidogrel in the PRoFESS trial.Reference Diener, Sacco, Yusuf, Cotton, Ounpuu, Lawton, Palesch, Martin, Albers, Bath, Bornstein, Chan, Chen, Cunha, Dahlöf, De Keyser, Donnan, Estol, Gorelick, Gu, Hermansson, Hilbrich, Kaste, Lu, Machnig, Pais, Roberts, Skvortsova, Teal, Toni, VanderMaelen, Voigt, Weber and Yoon67 There was no significant difference in cognitive decline or recurrent stroke between the two anti-platelet regimens.
The ongoing secondary prevention of small subcortical strokes (SPS3) trial has been designed to focus on the treatment and prevention of SVD, in the light of limited previous evidence. Two different anti-platelet regimes and the effect of intensive blood pressure control on stroke recurrence and cognitive decline will be studied and hopefully guide future management.Reference Bousser, Eschwege, Haguenau, Lefaucconnier, Thibult, Touboul and Touboul64
Cognitive enhancers
Recognition of the frequent co-existence of vascular and Alzheimer's pathology in older adults with cognitive impairment opens up additional treatment and prevention possibilities. UK guidelines suggest the use of acetylcholinesterase inhibitors for the symptomatic treatment of those with mild to moderate AD. Some indirect evidence about the cognitive effect of anti-dementia drugs on patients with SVD can be extrapolated from trials of patients with vascular dementia. Acetylcholinesterase inhibitors (donepezil and galantamine), when compared with placebo in three trials, produced a small treatment effect on cognition.Reference Black, Román, Geldmacher, Salloway, Hecker, Burns, Perdomo, Kumar and Pratt68–Reference Wilkinson, Doody and Helme70 Benefit in terms of global functioning or activities of daily living were modest. The NMDA antagonist memantine has also been shown to be of benefit in AD, but studies in patients with vascular dementia have failed to demonstrate significant clinical improvement.Reference Wilcock, Möbius and Stöffler71, Reference Orgogozo, Rigaud, Stöffler, Möbius and Forette72
SVD in specific treatment settings
Thrombolysis
Evidence of benefit from intravenous tissue plasminogen activator (tPA) has been demonstrated up to 4.5 hours after onset in acute ischaemic stroke.Reference Hacke, Kaste, Bluhmki, Brozman, Dávalos, Guidetti, Larrue, Lees, Medeghri, Machnig, Schneider, von Kummer, Wahlgren and Toni73 The NINDS trial reported that all stroke subtypes, including small vessel infarcts, responded favourably to treatment.74 However, intracranial bleeding is the most concerning side-effect and efforts have been made to identify factors that could increase occurrence. Advancing age, hyperglycaemia and hypertension have been associated with increased risk of haemorrhagic transformation.Reference Palumbo, Boulanger, Hill, Inzitari and Buchan75 Additionally, studies have suggested that neuroimaging evidence of SVD may also indicate risk of haemorrhage. A multivariate analysis of the Canadian Alteplase for stroke effectiveness study (CASES) suggested that the presence of leukoariosis or multiple lacunes on neuroimaging was associated with an increased bleeding risk, independent of age or other factors.Reference Palumbo, Boulanger, Hill, Inzitari and Buchan75
Oral anticoagulants
As discussed earlier, one of the manifestations of SVD is cerebral haemorrhage. In the stroke prevention in reversible ischaemia trial (SPIRIT),Reference Gorter76 leukoariosis was, together with older age, the only independent predictor of major bleeding with anticoagulation commenced after cerebral ischaemia (odds ratio 2.7, 95% confidence interval 1.4–5.3). Therefore, care should be taken with the use of anticoagulation in patients with extensive small vessel ischaemic damage on neuroimaging.
Multidisciplinary team rehabilitation
There is little hard evidence to guide best rehabilitative practice for patients with SVD. The general principles of good care and comprehensive geriatric assessment should apply. Physiotherapy may aid gait/balance disorders and incontinence should be fully assessed and managed in an appropriate fashion. Speech and language therapy may play a role in patients with relevant deficits.
Conclusions
Cerebral SVD is a common pathology in older people and a major cause of morbidity and functional disability. Newer neuroimaging techniques continue to allow us to diagnose cerebral SVD with greater accuracy, but evidence-based strategies for preventing disease progression are lacking. In routine clinical practice, attention should be paid to controlling conventional cardiovascular risk factors and future trials should focus both on measures to prevent the development of SVD and strategies to halt its progression once diagnosed.
Conflicts of interest
None declared.
