Cause and Mechanisms of Intracranial Atherosclerosis

HomeCirculationVol. 130, No. 16Cause and Mechanisms of Intracranial Atherosclerosis Free AccessResearch ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissionsDownload Articles + Supplements ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toSupplemental MaterialFree AccessResearch ArticlePDF/EPUBCause and Mechanisms of Intracranial Atherosclerosis Katja Ritz, PhD, Nerissa P. Denswil, MSc, Olga C.G. Stam, MD, Johannes J. van Lieshout, MD, PhD and Mat J.A.P. Daemen, MD, PhD Katja RitzKatja Ritz From the Departments of Pathology (K.R., N.P.D., O.C.G.S., M.J.A.P.D.) and Internal Medicine (J.J.v.L.), Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Laboratory for Clinical Cardiovascular Physiology, AMC Center for Heart Failure Research, Academic Medical Centre, Amsterdam, The Netherlands (J.J.v.L.); and MRC/Arthritis Research UK Centre for Musculoskeletal Ageing Research, School of Life Sciences, University of Nottingham Medical School, Queen’s Medical Centre, Nottingham, UK (J.J.v.L.). Search for more papers by this author , Nerissa P. DenswilNerissa P. Denswil From the Departments of Pathology (K.R., N.P.D., O.C.G.S., M.J.A.P.D.) and Internal Medicine (J.J.v.L.), Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Laboratory for Clinical Cardiovascular Physiology, AMC Center for Heart Failure Research, Academic Medical Centre, Amsterdam, The Netherlands (J.J.v.L.); and MRC/Arthritis Research UK Centre for Musculoskeletal Ageing Research, School of Life Sciences, University of Nottingham Medical School, Queen’s Medical Centre, Nottingham, UK (J.J.v.L.). Search for more papers by this author , Olga C.G. StamOlga C.G. Stam From the Departments of Pathology (K.R., N.P.D., O.C.G.S., M.J.A.P.D.) and Internal Medicine (J.J.v.L.), Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Laboratory for Clinical Cardiovascular Physiology, AMC Center for Heart Failure Research, Academic Medical Centre, Amsterdam, The Netherlands (J.J.v.L.); and MRC/Arthritis Research UK Centre for Musculoskeletal Ageing Research, School of Life Sciences, University of Nottingham Medical School, Queen’s Medical Centre, Nottingham, UK (J.J.v.L.). Search for more papers by this author , Johannes J. van LieshoutJohannes J. van Lieshout From the Departments of Pathology (K.R., N.P.D., O.C.G.S., M.J.A.P.D.) and Internal Medicine (J.J.v.L.), Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Laboratory for Clinical Cardiovascular Physiology, AMC Center for Heart Failure Research, Academic Medical Centre, Amsterdam, The Netherlands (J.J.v.L.); and MRC/Arthritis Research UK Centre for Musculoskeletal Ageing Research, School of Life Sciences, University of Nottingham Medical School, Queen’s Medical Centre, Nottingham, UK (J.J.v.L.). Search for more papers by this author and Mat J.A.P. DaemenMat J.A.P. Daemen From the Departments of Pathology (K.R., N.P.D., O.C.G.S., M.J.A.P.D.) and Internal Medicine (J.J.v.L.), Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; Laboratory for Clinical Cardiovascular Physiology, AMC Center for Heart Failure Research, Academic Medical Centre, Amsterdam, The Netherlands (J.J.v.L.); and MRC/Arthritis Research UK Centre for Musculoskeletal Ageing Research, School of Life Sciences, University of Nottingham Medical School, Queen’s Medical Centre, Nottingham, UK (J.J.v.L.). Search for more papers by this author Originally published14 Oct 2014https://doi.org/10.1161/CIRCULATIONAHA.114.011147Circulation. 2014;130:1407–1414Intracranial atherosclerosis, one of the leading causes of ischemic stroke, is associated with an increased risk for recurrent stroke and dementia.1,2 Individuals of Asian, Hispanic, and African American ancestry are especially affected. Recent European studies revealed a much higher prevalence of intracranial lesions than commonly presumed, suggesting that intracranial atherosclerotic disease is potentially the most common cause of ischemic stroke worldwide.1,3 Ischemic strokes are clinically categorized into 5 subtypes based on their underlying cause: large-artery atherosclerotic stenosis, small-artery disease (lacunes), cryptogenic, major-risk-source cardiogenic embolism, and unusual (eg, dissections, arteritis). Most nonlacunar ischemic strokes are thought to be thromboembolic, which presumably also accounts for most cryptogenic strokes. Embolic sources include minor-risk or covert cardiac sources, veins via paradoxical embolism, and nonocclusive atherosclerotic plaques in the aortic arch or cervical or cerebral arteries.4 Besides embolic strokes, 2 other mechanisms have been associated with intracranial atherosclerosis-related strokes, namely hypoperfusion through a stenotic artery causing watershed or border-zone stroke and plaque overgrowth of perforator artery ostia, which is associated with penetrating artery disease and lacunar infarcts and has been related to cryptogenic strokes.5–7 Even mild stenosis of intracranial atherosclerotic arteries (<50%) may therefore be clinically relevant, and high-resolution magnetic resonance imaging studies are needed to identify and determine the degree and location of stenosis in this patient group.5,8 The possibly causal role of nonstenotic plaques in ischemic stroke highlights the need for more insight into the mechanisms and occurrence of intracranial atherosclerosis.In the 1960s and 1970s, large, descriptive autopsy studies were conducted, providing classic morphological features of intracranial arteries. Despite the importance of intracranial atherosclerosis to stroke and dementia, there is a lack of more recent mechanistic studies. Therefore, we intend to draw attention to this neglected research field by providing an overview of available literature and a working model for intracranial atherosclerosis.Search StrategyA MEDLINE search was performed of literature published in English using the MeSH term intracranial atherosclerosis. A total of 8146 search hits were obtained for articles published before October 2013 and screened for relevance to the cause and mechanisms of intracranial atherosclerosis. Other references were extracted by cross-reference. Only studies including ≥10 patients were considered relevant. Because of space restrictions, we were able to include only a subset of references in this study; the remaining references can be found in Table I in the online-only Data Supplement.EpidemiologyIncidence and PrevalenceIn 45% to 62% of patients with ischemic stroke, intracranial plaques or stenoses were identified, which were causal in ≈10% to 20% of cases (reviewed elsewhere9). The estimated prevalence of symptomatic intracranial stenosis in literature ranges from 20% to 53%, depending on the study population, race, and method of choice (Table II in the online-only Data Supplement), and from 3% to 82% in asymptomatic patients. Most studies revealed a higher incidence of intracranial atherosclerosis in Asians and African Americans compared with white Americans. The few studies on white Europeans that are available suggest a high prevalence of intracranial plaques or stenosis. One French autopsy study detected intracranial plaques and stenosis in 62% and 43% of stroke cases, respectively, and in a Dutch study, 82% of asymptomatic patients showed calcification of the intracranial internal carotid artery (ICA) by computed tomography.3,8 Notably, magnetic resonance angiography, computed tomography angiography, and transcranial Doppler, which measure luminal changes, may underestimate the number of intracranial plaques, partly explaining the large differences in the literature.8Onset, Course, and DistributionIn the 1960s and 1970s, several large-scale autopsy studies performed in asymptomatic cohorts from fetuses to patients in their 10th decade of life revealed intracranial atherosclerotic changes from the first to second decade, progressing with age.10,11 Advanced atherosclerotic lesions are almost nonexistent up to the fourth decade.12 Overall, intracranial atherosclerosis develops ≈20 years later in life compared with atherosclerosis in extracranial arterial beds.13 Progression of atherosclerosis was not parallel in different vascular beds. Although aortic atherogenesis progressed linearly, intracranial atherosclerosis increased more slowly initially and paralleled aortic lesions thereafter.12 The steepest gradient for the incidence of intracranial atherosclerosis was reported in the sixth and seventh decades,10 with a steady increase beyond the eighth and ninth decades,11 whereas coronary atherosclerosis progressed more rapidly initially and attenuated between the fifth and eighth decades.12 Three percent to 4% of individuals >80 years of age exhibited only mild intracranial atherosclerotic changes.10,12 Intracranial atherosclerotic stenoses were described as dynamic lesions showing progression and regression, but they were less dynamic than coronary stenoses. Repeated magnetic resonance angiography over a 7-year period in patients with ischemic heart disease showed a 1.1% annual progression of the average intracranial stenosis. Differences among the intracranial vessels were evident with stable atherosclerosis in the intracranial ICA and dynamic lesions in the anterior (ACA), middle (MCA), and posterior (PCA) cerebral arteries, with 2.6% annual progression of average stenosis. Regression was noted in 14% of intracranial ICAs and 28% of ACAs, MCAs, and PCAs.14 A 5-year longitudinal study of 41 Japanese patients with ischemic heart disease reported the progression of cervical carotid artery stenosis in 5 patients (12%) and in only 1 patient with intracranial stenosis.15 During a 2- to 3-year follow-up of 40 stroke patients, 33% of MCA stenoses progressed and 8% regressed.16 Intracranial lesions were identified predominantly in the anterior circulation.17 Overall, American and European studies showed a similar pattern: The ICA was most commonly affected, followed by the MCA, basilar artery (BA), intracranial vertebral artery (VA), PCA, and ACA (Figure 1).10,18 The MCA appeared to be most commonly involved in Asians, followed by the ICA, BA, VA, PCA, and ACA.17 In all cohorts, cerebellar and communicating arteries were barely affected. Atherosclerosis in the intracranial ICA was observed mainly in the cavernous but also in the supraclinoid segment.18,19 The BA was commonly affected in the upper and lower parts and less affected in the middle part.20 MCA lesions were mainly found in the M2 segment.18Download figureDownload PowerPointFigure 1. Circle of Willis of a 90-year-old subject. Macroscopically, atherosclerotic lesions can be identified by the white vessels, whereas nondiseased arteries appear largely transparent. This case shows prominent atherosclerosis mainly in the internal carotid artery, vertebral artery, basilar artery, left middle cerebral artery, and posterior cerebral artery.MorphologyBasic StructureOne major characteristic that distinguishes healthy intracranial from extracranial arteries is that extracranial arteries such as the aorta and carotid arteries are elastic arteries rich in elastin filaments in the tunica media. In contrast, intracranial arteries are muscular arteries with few elastic fibers.21 The transition from elastic to muscular artery is at the level of the carotid bifurcation22 and embryological junctions between segments of the VA and ICA, which has been attributed to different embryological sites of origin of their primordial mesodermal cells.23 Compared with extracranial arteries of a similar size, a thinner media, less abundant adventitia, and only a few elastic fibers have been reported for intracranial vessels13 with a denser internal elastic lamina and without an external elastic lamina.13,21 The external elastic lamina is still present in the petrous portion of the ICA but disappears within the cavernous portion, which forms a hotspot of stenosis.24 A distinct vessel wall metabolism was suggested for intracranial arteries. Intima-media preparations of unaffected intracranial arteries showed lower contents of hexosamine, uronic acid, and sulfur; a lower proportion of hyaluronic acid and chondroitin sulfates in the total glycosaminoglycans; and a lower ratio of ester to total cholesterol, whereas the percentage of heparin sulfate was higher compared with normal aorta and coronary arteries.25 In addition, unaffected intracranial arteries of all ages revealed elevated antioxidant enzyme activity (manganese superoxide dismutase, copper-zinc superoxide dismutase, catalase) compared with extracranial arteries. Animal studies in rats and rabbits reported that the BA had less vesicles and caveolae and exhibited tight junctions between endothelial cells, leading to a reduced intimal permeability compared with the aorta and suggesting the presence of a barrier function.26,27 In monkeys and rabbits, a distinct composition of the glycocalyx on luminal endothelial cells was suggested in cerebral arteries; the carbohydrate-binding protein concanavalin A reacted with aorta, coronary, and carotid arteries but not with cerebral arteries. It has been speculated that a specific glycocalyx composition inhibits trapping of chylomicrons and very-low-density lipoprotein, resulting in a reduced deposition of apolipoproteins in the intima of intracranial vessels.28 In cats, histamine stimulation showed a 3-fold-stronger contraction of extracranial compared with intracranial arteries.29 Smooth muscle cells of rabbit intracranial arteries were relatively insensitive to sympathomimetic stimulation compared with systemic vessels.Vasa VasorumVasa vasorum are an essential component of extracranial vessels that facilitate nourishing and elimination of wastes; their existence in intracranial vessels is controversial. Most animal studies claim that vasa vasorum are absent beyond the first 1 to 2 cm of dural penetration in studies on Wistar rats, cats, and dogs. In cats, the presence of rete vasorum in the adventitia was suggested, being in continuity with the subarachnoid space, permeable to large proteins, and allowing diffusion between the cerebrospinal fluid and medial layer.30 Rete vasorum were not observed in humans. Vasa vasorum are found mainly in proximal intracranial segments; are associated with aging and vascular pathological processes such as vasculitis, atherosclerosis, aneurysm, and thrombosis; and may represent a response to vascular damage (reviewed elsewhere31). Vasa vasorum were reported in the MCA (M1 and A1 segments) and proximal parts of the ICA, VA, and BA.32,33 They were not seen beyond the first 1.5 cm proximal segments of the intracranial arteries after dural penetration, which corresponds to arteries with a thickness of <250 μm and to arteries surrounded by nutrient-rich cerebrospinal fluid.31,32 It is our experience that adventitial stripping, especially of smaller arteries, which is barely avoidable in autopsy studies, may lead to an underestimation of the presence of vasa vasorum. It has been suggested that vasa vasorum are required only beyond a critical vessel thickness to supplement luminal diffusion.32 The thin media and adventitia, which may facilitate luminal diffusion, and the absence of the external elastic lamina, which may allow greater diffusion from the cerebrospinal fluid, support this theory.31 Recent evidence suggests a critical role for the adventitia, including vasa vasorum in atherogenesis by initiating an inflammatory cascade. Along this line, later onset of more stable intracranial lesions may be explained partly by the initial absence of adventitial vasa vasorum, at least in normal intracranial arteries.31Plaque Characteristics and Age-Related ChangesIn intracranial arteries, aging was associated with a gradual loss of elastic fibers and muscular elements in the media and an increase in collagen tissue replacing medial muscle fibers.34 From the second to third decade, reduplication and splitting of the thick internal elastic lamina were observed frequently in combination with intimal thickening, which was most prominent from the fifth to sixth decade. In the same age group, fibrosis and hyalinization of media and adventitia prevailed.20 In the aorta and coronary arteries, fragmentation and reduplication of the internal elastic lamina were common in fetuses, infants, and young juveniles. In contrast to the aorta, lipids were rarely observed in intracranial arteries in patients <15 years of age.34Intracranial lesions not only developed later in life compared with extracranial vessels but developed mainly as fibrous plaques with fewer fatty streaks and complicated lesions. Complicated lesions, which contain calcifications or a plaque rupture, appeared after the fifth decade with a degree of involvement and lipid content similar to that observed in coronary arteries and were limited mainly to proximal segments of the ICA, VA, and BA.12,13 Hoff35,36 reported no major differences in chemical and enzymatic plaque characteristics in intracranial compared with extracranial arteries and no qualitative differences in apolipoproteins (apolipoprotein A1 from high-density lipoprotein, apolipoprotein B from low-density lipoprotein [LDL], apolipoprotein CIII from very-low-density lipoprotein).Intracranial arteries of human fetuses from hypercholesterolemic mothers showed fewer intimal macrophages and less intimal LDL, and oxidized LDL compared with extracranial arteries, which is suggestive of divergent atherogenic responses.37,38 Evidence for intracranial protective mechanisms comes from animal studies in rabbits. Hypercholesterolemia alone evoked a reduced intimal permeability and foam cell accumulation in the aorta but not in the BA, whereas the combination of hypercholesterolemia and hypertension also affected the permeability of intracranial arteries.27Our Histological DataOur review of the literature made clear that histological data in a recent cohort are lacking. Therefore, we screened 283 circle of Willis segments from 18 asymptomatic patients (mean age, 70.2±10.9 years; range, 51–90 years; male, 9; causes of death: cardiovascular disease, 8; malignancy, 3; subarachnoid hemorrhage, 2; Alzheimer disease, 2; sickle cell disease, 1; acute stroke, 1; HIV/hepatitis, 1) for basic structural features (Methods in the online-only Data Supplement). In accordance with previous literature, we identified mainly early lesions (63%) and a few advanced atherosclerotic lesions (15%). Calcifications were rare (6%). Two patients presented with complicated lesions (chronic total occlusion and intraplaque hemorrhage). Intracranial arteries, especially the smaller arteries (ACA, PCA, and cerebellar and communicating arteries), show a distinct structure such as the lack of vasa vasorum and an external elastic lamina and only a few medial elastic fibers compared with extracranial arteries. The larger intracranial arteries such as the ICA, MCA, VA, and BA, however, show an intermediate phenotype sharing structural features of both the larger extracranial and the smaller intracranial arteries (Table 1), which may explain the conflicting results in literature. Macrophage load was low in our series, which is in line which previous observations38 (0.9±0.7% CD68 positivity per plaque area compared with 1.8±2.4% in coronaries).39 In general, intracranial arteries exhibited fewer and more stable lesions, and the few advanced atherosclerotic lesions were identified predominantly in the large intracranial arteries such as the ICA, MCA, and VA.Table 1. Data From This Review on the Basic Characteristics of the Large Intracranial Arteries of 18 Asymptomatic PatientsICAVAMCABAType of lesion, % (n/N) Early57 (17/30)54 (15/28)68 (19/28)75 (24/32) Advanced33 (10/30)25 (7/28)25 (7/28)16 (5/32)High content of elastin fibers, % (n/N)37 (11/30)43 (12/28)29 (8/28)13 (4/32)Continuous EEL, % (n/N)17 (5/30)79 (22/28)21 (6/28)9 (3/32)Calcification, % (n/N)20 (6/30)18 (5/28)14 (4/28)9 (3/32)Vasa vasorum, % (n/N)53 (16/30)43 (12/28)11 (3/28)16 (5/32)Macrophages(mean±SD), %0.9±0.70.4±0.50.8±0.70.9±0.9BA indicates basilar artery; EEL, external elastic lamina; ICA, intracranial internal carotid artery; MCA, middle cerebral artery; and VA, intracranial vertebral artery. See Methods in the online-only Data Supplement for more information.Risk FactorsNonmodifiable Risk FactorsAgeAge is one of the most important independent risk factors for both intracranial and extracranial atherosclerosis. Several autopsy and imaging studies showed that aging is associated with increasing prevalence and severity of intracranial atherosclerosis among all investigated races with, as mentioned above, a distinct disease progression compared with extracranial arteries.3,10,19,40RaceAfrican Americans show a comparable or more severe degree of atherosclerosis compared with white Americans.41 In Asians, intracranial atherosclerosis developed earlier and more extensively compared with white Americans and Europeans.42 In symptomatic patients with transient ischemic attack or ischemic stroke, studies consistently reported the highest incidence and severity of intracranial atherosclerosis in Asians and Hispanics, followed by African Americans and whites. The reverse order was found for extracranial lesions.43,44 Differences in incidence and location of atherosclerosis were also observed within countries. Intracranial lesions in symptomatic patients were more common in North than in South China, which has been attributed to a more Westernized lifestyle in North China.45 Within African populations, Nigerians had lower atherosclerotic scores compared with Senegalese, Ugandans, and African Americans, with highest scores in the last group.46,47 Therefore, observed differences among races cannot be attributed to genetic factors only but are highly influenced by lifestyle and other risk factors. In addition, most studies do not take into account differences in the prevalence of vascular risk factors, being higher in Hispanic and African American populations and maybe resulting in an overrepresentation of specific races in intracranial atherosclerosis.48SexThe incidence of extracranial atherosclerosis is higher in men; this correlation is less evident in patients with intracranial lesions.19,40,49 This discrepancy may be attributed to the different disease course in men and women. Men showed a high increase in intracranial lesions in the fourth and fifth decades, which steadily progressed with age, whereas women exhibited relatively mild atherosclerotic lesions until the sixth decade, with rapidly increasing lesion formation thereafter. In the eighth and ninth decades, the degree of intracranial atherosclerosis was comparable between sexes, whereas women showed higher atherosclerotic scores in the ninth and tenth decades.10,45 It has been speculated that the observed sex differences can be explained by a distinct risk factor profile resulting from the influence of sex hormones such as the known hypocholesterolemic effect of estrogens.50OthersIt has been suggested recently that the circle of Willis and its communicating arteries protect the cerebral artery and blood-brain barrier from hemodynamic stress.51 In line with this hypothesis, variations in the circle of Willis were shown to influence the volume flow rates of the bilateral ICA and BA in healthy individuals and the development of atherosclerosis.52,53 Racial differences in atherogenesis that have been discussed before could not be linked to anatomic variations of the circle of Willis.54 No clear genetic risk factors such as 9p21 for coronary heart disease have been identified for intracranial atherosclerosis. Notably, most studies have been conducted in Asian cohorts; data for other countries, especially European countries, are lacking.Modifiable Risk FactorsHypertensionOne of the most important risk factors for atherosclerosis, especially intracranial lesions, is hypertension. Hypertension has been correlated to the degree of atherosclerosis in intracranial arteries in different ethnic cohorts.17,19,55 Some studies reported a higher incidence of hypertension in populations of African and Asian ancestry, which may explain their higher prevalence of intracranial atherosclerosis.43,46,56Diabetes MellitusDiabetes mellitus is a specific risk factor for intracranial lesions regardless of race in symptomatic and asymptomatic cohorts.3,8,19,40,44 In Koreans, diabetes mellitus was an independent risk factor for intracranial lesions only after 50 years of age50 and only in posterior, not anterior, circulation diseases in a prospective study.17 As for hypertension, the aforementioned higher incidence of intracranial lesions in patients of African and Asian ancestry may be partly attributed to an increased prevalence of diabetes mellitus.56Metabolic SyndromeRecent magnetic resonance angiography studies showed an independent association of the metabolic syndrome with intracranial atherosclerosis.17,44 In a prospective Korean study, the metabolic syndrome was related more to intracranial than to extracranial lesions and to posterior and not anterior circulation strokes.17DyslipidemiaDyslipidemia is a known risk factor for coronary atherosclerosis and myocardial infarction, but its role in intracranial atherosclerosis is less clear.48,57 High LDL cholesterol was associated mainly with extracranial lesions, whereas a high ratio of apolipoprotein B to apolipoprotein I and low levels of apolipoprotein AI, the major protein component of high-density lipoprotein, correlated with intracranial lesions.49,58 In China, low high-density lipoprotein cholesterol is one of the most common types of dyslipidemia and was associated with the development of intracranial artery stenosis in a cohort of acute ischemic stroke.59 Sex-specific differences were reported in 2 Asian studies. Hypercholesterolemia was an independent risk factor for intracranial atherosclerosis only in asymptomatic men, whereas elderly symptomatic women >63 years of age had significantly more intracranial atherosclerotic lesions and hyperlipidemia than men.45,50 Race and environmental factors may influence the effect of dyslipidemia on atherogenesis. Generally, individuals of Asian and African ancestry exhibited lower serum lipid levels than whites, which may be one factor explaining the lower incidence of extracranial and coronary atherosclerosis in both populations.11,43OthersExtracranial atherosclerosis was suggested as a risk factor for intracranial lesions. Extensive coronary atherosclerotic disease correlated with intracranial lesions,8,11,60 and patients with concurrent lesions had a higher risk of suffering further (fatal) vascular events.61 In a cohort of symptomatic intracranial atherosclerosis, 52% of cases were diagnosed with silent myocardial ischemia caused by coronary artery disease.62 The American Heart Association Stroke Council recommends testing for asymptomatic coronary artery disease in patients who have had ischemic events associated with intracranial atherosclerosis.9,63 In contrast, other studies reported correlations between coronary and carotid but not intracranial lesions64 or failed to show a correlation.15,18 A few studies suggested that smoking, especially duration of smoking, is a risk factor for intracranial lesions.3,49 However, large-scale studies of the effects of smoking on intracranial atherosclerosis are scarce. As for extracranial lesions, moderate hyperhomocysteinemia was a predictor for severity of intracranial atherosclerosis in Asian patients with cerebral infarction.65 A few reports associated Alzheimer disease, sickle cell disease, systemic lupus erythematosus, radiotherapy, bacterial meningitis, and Herpes zoster infection with intracranial atherosclerosis, but their contribution needs further research.66–71In conclusion, age, hypertension, diabetes mellitus, and probably the metabolic syndrome are the most consistent risk factors for intracranial atherosclerosis. Race may represent a predisposing factor, which is unfavorable in combination with other risk factors and especially lifestyle. Sex appears to influence intracranial atherosclerosis, and its effects are age dependent. Genetics may predispose to intracranial atherosclerosis, but large-scale association studies in different ethnic groups are lacking.MechanismsTwo major characteristics that distinguish intracranial and extracranial atherosclerosis are the later onset and the more stable plaque phenotype in intracranial arteries, which may be explained by the distinct characteristics of the intracranial arteries (Table 2 and Figure 2). These characteristics may also be linked to the role of intracranial arteries in regulating the cerebrovascular resistance. The control mechanisms of cerebral blood flow (CBF) encompass cerebrovascular responsiveness to O2 and CO2, cerebral autoregulation and neurogenic control of the cerebral vasculature, endothelium-mediated signaling, and neurovascular coupling meeting local cerebral metabolic demand.75 Cerebral artery endothelial cells and pericytes produce nitric oxide in direct proportion to the arterial CO2 partial pressure76 and contribute to the resting tone of cerebral arteries and arterioles.77 Impairment of cerebral autoregulation results in pressure-passive CBF; that is, CBF increases and decreases together with cerebral perfusion pressure, whereas reduced CO2 responsiveness affects the vasodilatory reserve of the brain (Figure I in the online-only Data Supplement). The traditional concept in peripheral circulations is that arterioles rather than large arteries are the main site of vascular resistance. However, for the brain, large extracranial vessels and surface vessels contribute ≈50% of cerebral vascular resistance,78 and both CBF