Clinical Relevance of Cerebral Small Vessel Diseases.

Abstract

HomeStrokeVol. 51, No. 1Clinical Relevance of Cerebral Small Vessel Diseases Free AccessResearch ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticlePDF/EPUBClinical Relevance of Cerebral Small Vessel Diseases Marco Pasi, MD and Charlotte Cordonnier, MD, PhD Marco PasiMarco Pasi From the Department of Neurology, Univervité de Lille, Inserm U1171, Degenerative and Vascular Cognitive Disorders, CHU Lille, France. Search for more papers by this author and Charlotte CordonnierCharlotte Cordonnier Correspondence to Charlotte Cordonnier, MD, PhD, Department of Neurology, Univervité de Lille, Inserm U1171, Degenerative and Vascular Cognitive Disorders, CHU Lille, Lille, France. Email E-mail Address: [email protected] From the Department of Neurology, Univervité de Lille, Inserm U1171, Degenerative and Vascular Cognitive Disorders, CHU Lille, France. Search for more papers by this author Originally published22 Nov 2019https://doi.org/10.1161/STROKEAHA.119.024148Stroke. 2020;51:47–53is related toNew Treatment Approaches to Modify the Course of Cerebral Small Vessel DiseasesAdvanced Neuroimaging to Unravel Mechanisms of Cerebral Small Vessel DiseasesMultiple Faces of Cerebral Small Vessel Diseasesis related toGenetics of Cerebral Small Vessel DiseaseCerebral Autosomal Dominant Arteriopathy With Subcortical Infarcts and LeukoencephalopathyOther version(s) of this articleYou are viewing the most recent version of this article. Previous versions: November 22, 2019: Ahead of Print See related articles, p 9, p 12, p 21, p 29, p 38Until recent years, cerebral small vessel disease (SVD) has drawn little attention mostly because small cerebral arteries were difficult to observe in imaging and in vivo studies. Furthermore, although some fields of vascular neurology have been revolutionized with reperfusion therapies, stroke specialists have little to offer for diseases that affect small cerebral vessels. Nevertheless, SVDs play a crucial role in a large variety of conditions such as aging, stroke, cognitive impairment, and other age-related disabilities (motor and gait impairment, mood disorders, and urinary symptoms). The term SVD is used with various meanings according to the context (ie, clinical, research, neuropathology, and neuroimaging aspects). From a neuropathological perspective, SVD encompasses a group of pathologies that affect the perforating arteries, arterioles, capillaries that are located in the brain parenchyma or in the leptomeningeal vessels.1–3 Small vessel pathological processes that involve small veins and venules have not only been less frequently reported (such as venous collagenosis) but also less frequently be the object of pathology studies that mostly focus on arterial disease.1,3 Although recent investigational work has suggested that it may be possible to visualize pathologies in these small arteries with 7 Tesla magnetic resonance imaging (MRI), standard neuroimaging used in clinical practice does not yet allow the direct visualization of these vessels.4 Thus, the term SVD is also used to describe a range of neuroimaging and pathological features of parenchymal damage that can either be associated to clinical symptoms or clinically silent.5Two main forms of sporadic SVD have been described: arteriolosclerosis and cerebral amyloid angiopathy (CAA). The former describes a group of small vessel pathological processes. Arteriolosclerosis has a predilection for the deep lenticulostriate arteries that are vulnerable to poorly controlled and long-standing hypertension or diabetes mellitus.6 The second most common form of sporadic SVD is CAA, a pathological process characterized by the progressive accumulation of ß-amyloid protein in the wall of small cortical and leptomeningeal arterioles and arteries.7Cerebral SVDs can manifest with a widespread and heterogeneous range of clinical manifestations that often lead to functional disability in the late stages of the disease. Diseases that affect small arteries and arterioles may present abruptly and dramatically like in intracerebral hemorrhage (ICH) and lacunar stroke. Differently, an insidious clinical course may be associated with progressive cognitive decline, development of mood disorders, and motor problems related to small vessel pathologies. Finally, imaging markers of cerebral SVD, not associated with a clear-cut symptomatology, are among the most common incidental findings on MRI.8 Population-based studies suggest that they are associated with subtle neurological symptoms, and their presence confer an increased risk of stroke and cognitive impairment.9,10The neuropathological substrates of SVDs lead to both ischemic and hemorrhagic consequences. Hence, stroke specialists may face challenging clinical situations in balancing hemorrhagic and ischemic risk, especially when antithrombotic medication may be needed.In this special focus on clinical consequences of SVDs, we aim to provide a comprehensive overview of the main clinical manifestations of SVD. We will also explore the clinical impact of SVDs not associated with a clear-cut symptomatology on the development of disabling conditions such as stroke and dementia.Acute Ischemic Expression of SVD: Lacunar StrokesCerebral SVD results in approximately one-fourth of all acute ischemic strokes, mainly in the form of lacunar infarction. According to the TOAST (Trial of ORG 10172 in Acute Stroke Treatment) criteria, lacunar strokes are defined as small subcortical brain infarcts visible on MRI (normal computed tomography scan if evaluated in the acute phase) <1.5 cm in axial diameter and associated with one of the traditional clinical lacunar syndromes.11 The STRIVE (Standards for Reporting Vascular Changes on Neuroimaging) proposed the term of recent small subcortical infarcts to define the neuroimaging evidence of recent infarction with an axial diameter of 2 cm in diffusion-weighted imaging MRI sequences in the territory of one perforating arteriole (deep cerebral white matter, basal ganglia, thalamus, or pons), with imaging features or clinical symptoms consistent with a lesion occurring in the previous few weeks.5 Although lacunar strokes may occasionally result from mechanisms of brain ischemia, such as cardiac embolism or carotid artery stenosis, most result from intrinsic diseases of the small deep perforating arteries.12,13 Hypertension and diabetes mellitus are said to be strongly associated with lacunar stroke.14,15 However, in studies using risk factor free ischemic stroke subtype definitions, there was only a marginal excess of hypertension with lacunar versus nonlacunar infarction and no difference for diabetes mellitus.15 From the literature, there is no clear specific association between other modifiable risk factors such as smoking and excessive alcohol consumption when compared to nonlacunar ischemic stroke.15Few studies have evaluated functional outcomes in lacunar stroke patients: the proportion of dependent patients is around 42% at 3 years.16 Stroke recurrence is reported to be similar than other stroke subtypes with a risk of recurrence after 1 year of about 4.6% per year.16Lacunar strokes are considered as a common cause of cognitive impairment: ≈30% of patients with lacunar stroke will be cognitively impaired in the 4 years following the index event.17 This proportion is similar to nonlacunar stroke (23%). Cognitive impairment following lacunar strokes seems more related with an underlying progressive process (namely ongoing cerebral SVD) rather than from the acute lesions itself. Cognitive impairment in lacunar stroke patients is generally considered to be domain specific mostly involving executive functions, attention, and psychomotor speed. However, results of a systematic review suggest that impaired cognition in this stroke subtype appears less selective than previously thought, involving all major cognitive domains.18The SPS3 trial (Secondary Prevention of Small Subcortical Strokes) is the largest study to date that enrolled MRI confirmed lacunar stroke.19 The SPS3 trial tested 2 randomized interventions: clopidogrel and aspirin versus aspirin alone and 2 target levels of systolic blood pressure.19 Dual antiplatelet therapy increased the incidence of major hemorrhage and mortality rate, and thus, the trial was stopped early. Furthermore, the risk of recurrent stroke was not significantly reduced by dual antiplatelet therapy. The SPS3 was also designed to test whether a systolic blood pressure target of <130 mm Hg compared with 130–149 mm Hg would be associated with a reduction of all strokes (ischemic and hemorrhagic). Despite a 19% nonsignificant reduction in all strokes, the authors reported a significant reduction in ICH.20 These findings suggest that a strict blood pressure control is likely to be beneficial in patients with recent lacunar stroke.Currently, antiplatelet monotherapy is recommended to prevent recurrent strokes after lacunar strokes because dual antiplatelet therapy might increase major bleeding risk without providing additional stroke reduction benefits.21 Recent evidence, coming from trials that enrolled all types of ischemic stroke patients in mostly Asian populations, suggest that cilostazol may have lower risks of hemorrhage than aspirin while offering similar efficacies in stroke prevention.22,23 A recent trial (PICASSO: Prevention of Cardiovascular Events in Asian Patients With Ischemic Stroke at High Risk of Cerebral Hemorrhage) evaluated the efficacy and safety of cilostazol versus aspirin. In this study, cilostazol was noninferior to aspirin for the prevention of cardiovascular events but did not reduce the risk of hemorrhagic stroke.24 Recently, a randomized controlled trial showed that in patients with minor ischemic stroke or high-risk transient ischemic attack (patients with presumed atrial fibrillation were excluded), those who received a combination of clopidogrel and aspirin had a lower risk of major ischemic events but a higher risk of major hemorrhage (intracranial and noncerebral) at 90 days than those receiving aspirin alone.25Regarding reperfusion therapy, patients with lacunar strokes benefit from intravenous thrombolysis like the other stroke subtypes.26 The presence of SVD, frequent in patients with lacunar stroke on pretreatment neuroimaging, might confer an increased risk of both remote cerebral hemorrhage and hemorrhagic transformation of the infarcted area but does not represent an absolute contraindication to thrombolysis.26–28Acute Hemorrhagic Expression of SVD: ICHNontraumatic, spontaneous ICH is one of the most feared and devastating manifestations of cerebral SVD in terms of mortality and morbidity.29 It affects ≈2 million people worldwide each year.30 It is a life-threatening event as more than half of patients will die at 1 year and 1 in 5 survivors will be independent at 1 year being also at high risk of ICH recurrence, dementia and other vascular event.30,31Spontaneous ICH results from rupture and bleeding of small arteries and arterioles into the brain parenchyma. Arterial bleeding results from the 2 most common forms of sporadic aged-related cerebral SVD: arteriolosclerosis and CAA. The location of the hemorrhage may suggest the dominant underlying microangiopathy (Figures 1 and 2). Pathological studies have shown that ICH secondary to arteriolosclerosis are generally located in the basal ganglia, thalamus, deep white matter, and pons.6 In CAA, ICH are mostly located in lobar location.32,33 How much cerebellar ICH relates to arteriolosclerosis versus CAA is uncertain. As in supratentorial ICH, preliminary results suggest that cerebellar hemorrhages restricted to the deep regions of the cerebellum (deep white matter and nuclei) are related to arteriolosclerosis, whereas when limited to superficial cerebellar regions (cortex and surrounding white matter), they are associated with CAA.33aDownload figureDownload PowerPointFigure 1. Ischemic and hemorrhagic magnetic resonance imaging (MRI) features of deep perforating arteriopathy. A–D, MRI features of both ischemic and hemorrhagic markers related to arteriolosclerosis with (1) A (susceptibility-weighted imaging) showing a left (deep) thalamic hemorrhage and a deep cerebral microbleeds (inset); (2) B illustrating a left thalamic lacune of presumed vascular origin (inset) and a right deep hematoma (*); (3) C (T2-weighted sequence) showing severe basal ganglia enlarged perivascular spaces (inset); (4) D (fluid-attenuated inversion recovery sequence) depicting extensive white matter hyperintensities with a peri-basal ganglia pattern (inset).Download figureDownload PowerPointFigure 2. Hemorrhagic and nonhemorrhagic magnetic resonance imaging (MRI) features of cerebral amyloid angiopathy that can be used in a clinical setting. A–D, MRI features of cerebral amyloid angiopathy with (1) A (susceptibility-weighted imaging) showing a lobar intracerebral hemorrhage (*), cortical superficial siderosis (arrow), and lobar cerebral microbleeds (inset); (2) B depicting (fluid-attenuated inversion recovery sequence) periventricular white matter hyperintensities with a prevalent posterior distribution (arrow) and white matter hyperintensities multiple subcortical spots (inset); (3) C showing (T2-weighted sequence) multiple enlarged perivascular spaces (EPVS) in the centrum semiovale (inset); (4) D (fluid-attenuated inversion recovery sequence) showing a left superficial (lobar) lacune (inset) and a right old lobar hemorrhage (*).Identifying the nature of the underlying SVD is clinically important as CAA related ICH is associated with a higher risk of recurrences (7 to 10% per year).34 In the few studies which evaluated well-phenotyped patients with arteriolosclerosis-related ICH evaluated on MRI, the risk of recurrent ICH was around 2% per year.34Brain MRI will bring clues to identify the nature of the underlying vessel disease (Figures 1 and 2). In ICH related to arteriolosclerosis,35,36 cerebral microbleeds (CMBs) are mainly restricted to deep brain regions and usually are not associated with cortical superficial siderosis (cSS; Figure 1).37 Furthermore, the use of fluid-attenuated inversion recovery, T1 and T2-weighted MRI sequences in deep ICH allow the evaluation of the severity and location of nonhemorrhagic markers. These markers are the presence of severe (>20) enlarged perivascular spaces (EPVS) located in the basal ganglia, a peri-basal ganglia pattern of white matter hyperintensities (WMH) following the peripheral outline of the basal ganglia, and lacunes in the basal ganglia (Figure 1).35,36,38Similarly, blood sensitive MRI sequences have revolutionized clinical practice in CAA by their ability to show both acute and chronic bleeding (Figure 2) and by contributing to the in vivo diagnosis of CAA.32,39 According to the modified Boston criteria, if a patient over 55 years of age with a symptomatic lobar ICH demonstrates ≥1 additional strictly lobar CMBs or ICH or a region with cSS, without clinical or radiological suspicion of an alternative cause, the diagnosis of probable CAA is made. Nonhemorrhagic radiological markers will also contribute to the positive diagnosis (even if not included yet in the modified Boston criteria): multiple EPVS in centrum semiovale, >10 subcortical fluid-attenuated inversion recovery WMH spots, severe posterior WMH, and radiological lacunes (<15 mm) in a peripheral (lobar) distribution (Figure 2). Recently, the pathologically validated Edinburgh computed tomography criteria (including finger like shaped ICH, subarachnoid extension and APOE ε4 genotype) have been proposed to identify in vivo CAA at admission in patients with a lobar ICH.40 In daily clinical practice, the situation is rarely so clear: ICH associated with a mixed pattern share an annual risk of ICH recurrence around 5% per year, and this hemorrhagic pattern appears to be driven mostly by vascular risk factors similar to patients with deep perforating vasculopathy.34Until now, no imaging markers have been precisely recognized to identify those among deep ICH patients who are at higher risk of recurrent ICH. Among patients with CAA, the risk of recurrent ICH seems heterogeneous. MRI may help to further stratify recurrence risks. Presence and extent of cSS seems to be an important radiological marker of recurrent ICH in CAA.37 In a recent meta-analysis, during a mean follow-up of 2.5 years, 92 patients out of 443 CAA-ICH patients experienced recurrent ICH (pooled risk ratio, 6.9% per year [95% CI, 4.2%–9.7% per year]). In adjusted pooled analysis, presence of disseminated cortical siderosis increased the risk of lobar ICH by 4×.41 Even in a cohort of patients with probable CAA and without a history of previous ICH, the presence of cSS was also a significant marker of first ever symptomatic lobar ICH.42Although recurrent ICH has been the major fear of stroke specialists, the risk of not only intracranial but also extracranial ischemic events (such as myocardial infarction) may be at least as frequent in ICH patient.43,44 Indeed, cerebral ischemic and hemorrhagic diseases share several major risk factors such as arterial hypertension. An important issue for clinicians will, therefore, be whether antithrombotic drugs (antiplatelet and anticoagulant) are potentially beneficial or detrimental in secondary prevention in ICH patients. Limited observational evidence suggest that restarting anticoagulation in ICH patients with atrial fibrillation is not associated with an higher rate of ICH recurrence and may even be related with better vital and functional outcome, even in patients with lobar ICH related to CAA.45 However, these are nonrandomized, observational data. Several randomized controlled trials are ongoing and will evaluate the possible impact of underlying SVD on the net clinical benefit of antithrombotic treatments.46Finally, recent data suggest that ICH patients are at high risk of developing dementia.47–49 In the PITCH cohort (Prognosis of Intracerebral Hemorrhage), out of 218 patients (median age, 67.5 years) without preexisting dementia alive at 6 months follow-up, 63 patients developed new-onset dementia leading to an incidence rate of 14.2% at 1 year after ICH, and incidence reached 28% at 4 years. Lobar location of the ICH, cSS, CMBs, cortical atrophy, and older age were independent predictors of new-onset dementia, suggesting that underlying CAA is an important contributor. These findings suggest that an ongoing underlying cognitive impairment process takes place rather than new-onset dementia being induced by the ICH itself.CAA can also cause nontraumatic acute convexity subarachnoid hemorrhage located in the cerebral supratentorial sulci. Typically, acute convexity subarachnoid hemorrhage manifests with transient focal neurological episodes due to the presence of blood in the subarachnoid spaces of ≥1 sulci in contact or contiguous with clinically eloquent cerebral areas.7 Most published cases describe transient focal neurological episodes as recurrent, stereotyped, spreading paresthesia, usually lasting several minutes.50,51 Because transient focal neurological episodes may precede symptomatic ICH in the context of CAA, their recognition may reduce the risk of brain bleed by avoiding antithrombotic use after a potential misdiagnosis as a transient ischemic attack.SVD Markers on MRI: Clinical ImplicationsPerforming MRI for various clinical indications is becoming increasingly frequent, and MRI markers of SVD (WMH, lacunes of presumed vascular origin, CMBs, and EPVS) are among the most commonly reported incidental findings.8WMH and Lacunes of Presumed Vascular OriginThe best-defined radiological manifestations of cerebral SVD are WMH and lacunes of presumed vascular origin.WMH prevalence increases exponentially with age reaching, at any degree of severity, ≈90% in individuals older than 80 years.52 WMH are likely to be more common in individuals with history of ischemic and hemorrhagic strokes, dementia, migraine, or late-life depression.53–56 The most consistently identified risk factors for WMH are advanced age, hypertension, and diabetes mellitus.Large population-based studies suggest that WMH presence, especially when moderate or severe, is not an innocuous bystander. In a recent meta-analysis that took into account >14 000 people from the general population and participants with high risk for vascular events, WMH burden was associated with more than a doubled risk of ischemic stroke and a 3-folds higher risk for ICH, compared with patients with no or mild WMH burden.10 Furthermore, WMH burden was associated with a significant risk of developing subsequent dementia (including Alzheimer type) in both general population and population at high risk of vascular events. This latter finding is in line with a large pathology study of >1100 community-dwelling elderly people which showed that arteriolosclerosis increases the odds of probable and possible Alzheimer disease dementia, independently of the effect of infarcts and Alzheimer disease pathology.56aLacunes of presumed vascular origin have mostly been studied in population-based studies as the most prevalent expression of radiological cerebral ischemic lesions (excluding WMH). They most likely correspond to infarcts on pathology and are usually not associated with a clear-cut symptomatology. All the data reported in this section refer to studies that have evaluated radiological presumably cerebrovascular ischemic lesions (using heterogeneous terminology as brain infarct, silent brain infarct, covert brain infarct) that in >90% of cases are located in subcortical regions. This radiological entity has been reported in ≈20 % of elderly individuals and up to 50 % in high-risk population.9,57 Radiological cerebral ischemic lesions are not related to acute stroke-like symptoms and overt disabilities, but they are associated with more subtle neurological deficits.57 Therefore, the term covert brain infarct is sometimes used.57 Radiological cerebral ischemic lesions on MRI are relevant markers of increased risk of future cerebral ischemic event, and the risk of future ICH seems even higher.10 Furthermore, the presence of radiological cerebral ischemic lesions more than doubles the risk of dementia, including Alzheimer disease.9The nontrivial increased risk of both ischemic and hemorrhagic cerebral events in patients with WMH and lacunes raise the questions whether aspirin or antithrombotics medication may be considered as a valid therapeutic option in primary stroke prevention. In the absence of randomized controlled trial, guidelines on prevention of stroke in patients with silent cerebrovascular disease states that it is not clear whether WMH or silent brain infarcts is a sufficient reason for aspirin therapy in prevention of ischemic stroke.58Cerebral MicrobleedsCMBs are an hemorrhagic expression of chronic cerebral SVD that are present in 8% to 24% of the general population.59,60 Despite being considered as small areas of hemosiderin deposition from previous silent hemorrhages, CMBs are associated with an almost 2-fold increased risk of ischemic stroke.10 The risk is almost increased by 3× for future cerebral hemorragic events.10 The risk of either ischemic or hemorragic cerebral events is influenced by the anatomic distribution of CMBs. Indeed, deep CMBs are related with arteriolosclerosis, whereas CMBs restricted to superficial location (lobar) are related with CAA. Based on the Boston criteria, the majority of patients older than 55 years with multiple strictly lobar CMBs, even without ICH, and in the absence of an alternative diagnosis, may have CAA.39,61,62 In this population, the risk of future ICH may be around 5% per year.62 The risk of cognitive decline and dementia is also at stake and seems higher in patients with strictly lobar CMBs compared with those with deep CMBs.Because of their hemorrhagic substrate, several observational studies aimed at investigating whether CMBs predict symptomatic ICH in stroke patients treated with intravenous thrombolysis. In a recent individual patients data meta-analysis, CMBs were associated with a greater risk of symptomatic ICH and poor functional outcome in ischemic stroke patients undergoing thrombolytic therapy.28 This risk was meaningful in patients with >10 CMBs.63 However, this result is only based on nonrandomized observational data. Therefore, before intravenous thrombolysis when computed tomography scan is the first line imaging, MRI blood sensitive sequences are not mandatory to exclude the presence of CMBs. When MRI is performed, thrombolytic treatment should not be withheld in eligible patients solely because of CMBs.CMBs are found in ≈1 every 4 patients with ischemic stroke.64 In a large meta-analysis, both the presence of any CMBs and their increased numbers were significant predictors of ICH risk during follow-up.64 These results led to concerns on the safety of life-long anticoagulation use in patients with CMBs and nonvalvular atrial fibrillation.65 Unfortunately, no randomized controlled study to date have clearly determined the net clinical benefit of long term anticoagulation in patients with CMBs and a high risk of thromboembolic event. Therefore, routine MRI screening in patients before starting oral anticoagulation cannot be recommended.58Enlarged Perivascular SpacesIn recent years, EPVS have gained attention especially in the field of research as an imaging marker of cerebral SVD. EPVS (also known as Virchow-Robin spaces) are interstitial fluid-filled cavities surrounding the small penetrating vessels which function as the brain drainage system.66 Increasing evidence suggest that the topography of EPVS is characteristic of a specific underlying SVD type: (1) when located in the basal ganglia, EPVS are associated with arteriolosclerosis; (2) EPVS in the centrum semiovale are related to CAA.35 However, their clinical impact in the general population remains unclear. For example, in the Northern Manhattan Study, participants in the highest tertile of EPVS burden did not show a significantly higher risk of incident stroke.67 But, in a large population-based study (N=1778 participants), the presence of EPVS in the basal ganglia and white matter was associated with a significant increased risk of incident dementia.68Future Strategies to Prevent SVD Clinical ManifestationsAdvances in the understanding of the role of cardiovascular risk factors have shown a close association between their burden and dementia and cognitive impairment. Part of their detrimental effect on cognition might be explained by presence and progression of cerebral SVD. Recently, the American Heart Association provided an initial definition of optimal brain health and guidance on how to maintain brain health throughout life.69 This definition of optimal brain health (nonsmoking status, physical activity at goals level, body mass index <25 kg/m2, healthy diets, untreated blood pressure <120/80 mm Hg, untreated total cholesterol <200 mg/dL, fasting blood glucose <100 mg/dL) emphasizes the importance of a favorable or ideal cardiovascular risk profile. Because many cardiovascular risks are modifiable, it may be possible to slow SVD progression by their optimal control. Maintaining such a low-risk factor profile earlier in life showed promising results with better cognition in midlife.70Future research effort and randomized control trials are warranted to evaluate interventions that may prevent or mitigate the wide spectrum of clinical symptoms related to SVD.DisclosuresDr Cordonnier is the principal investigator of the A3ICH study (Avoiding Anticoagulation After Intracerebral Haemorrhage) funded by the French ministry of health (programme hospitalier de recherche clinique [PHRC]). Dr Cordonnier is a member of the Institut Universitaire de France. The other author reports no conflicts.FootnotesCorrespondence to Charlotte Cordonnier, MD, PhD, Department of Neurology, Univervité de Lille, Inserm U1171, Degenerative and Vascular Cognitive Disorders, CHU Lille, Lille, France. Email charlotte.[email protected]fr