Association of Intracranial Atherosclerotic Disease With Brain β-Amyloid Deposition: Secondary Analysis of the ARIC Study | Cardiology | JAMA Neurology | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Histograms Showing Distribution of Number of Intracranial Atherosclerotic Plaques
Histograms Showing Distribution of Number of Intracranial Atherosclerotic Plaques

Shown, for the total sample (A) and for the group with elevated florbetapir standardized uptake value ratio (SUVR) (B) and without elevated florbetapir SUVR (C).

Figure 2.  Mean Standardized Uptake Value Ratio (SUVR) Values in Regions Within Certain Vascular Territories for Individuals With and Without Regional Intracranial Atherosclerotic Disease (ICAD) Features in Matching Vascular Territories
Mean Standardized Uptake Value Ratio (SUVR) Values in Regions Within Certain Vascular Territories for Individuals With and Without Regional Intracranial Atherosclerotic Disease (ICAD) Features in Matching Vascular Territories

A, Anterior cerebral artery territory. B, Posterior cerebral artery territory. C, Middle cerebral artery territory.

Table 1.  Participant Demographics for 300 Participants
Participant Demographics for 300 Participants
Table 2.  Association Between Intracranial Plaque and Stenosis Features and Elevated Brain β-Amyloida
Association Between Intracranial Plaque and Stenosis Features and Elevated Brain β-Amyloida
Table 3.  Associations Between Plaque and Vessel Characteristics and Elevated Global Cortical Brain β-Amyloid by APOE ε4 Status
Associations Between Plaque and Vessel Characteristics and Elevated Global Cortical Brain β-Amyloid by APOE ε4 Status
1.
Beach  TG, Wilson  JR, Sue  LI,  et al.  Circle of Willis atherosclerosis: association with Alzheimer’s disease, neuritic plaques and neurofibrillary tangles.  Acta Neuropathol. 2007;113(1):13-21. doi:10.1007/s00401-006-0136-yPubMedGoogle ScholarCrossref
2.
Dearborn  JL, Zhang  Y, Qiao  Y,  et al.  Intracranial atherosclerosis and dementia: the Atherosclerosis Risk in Communities (ARIC) study.  Neurology. 2017;88(16):1556-1563. doi:10.1212/WNL.0000000000003837PubMedGoogle ScholarCrossref
3.
Gottesman  RF, Schneider  ALC, Zhou  Y,  et al.  Association between midlife vascular risk factors and estimated brain amyloid deposition.  JAMA. 2017;317(14):1443-1450. doi:10.1001/jama.2017.3090PubMedGoogle ScholarCrossref
4.
Knopman  DS, Griswold  ME, Lirette  ST,  et al.  Vascular imaging abnormalities and cognition: mediation by cortical volume in nondemented individuals: Atherosclerosis Risk in Communities-Neurocognitive Study.  Stroke. 2015;46(2):433-440. doi:10.1161/STROKEAHA.114.007847PubMedGoogle ScholarCrossref
5.
Qiao  Y, Guallar  E, Suri  FK,  et al.  MR imaging measures of intracranial atherosclerosis in a population-based study.  Radiology. 2016;280(3):860-868. doi:10.1148/radiol.2016151124PubMedGoogle ScholarCrossref
6.
Gottesman  RF, Schneider  ALC, Zhou  Y,  et al.  The ARIC-PET amyloid imaging study: brain amyloid differences by age, race, sex, and APOE.  Neurology. 2016;87(5):473-480. doi:10.1212/WNL.0000000000002914PubMedGoogle ScholarCrossref
7.
Samuels  OB, Joseph  GJ, Lynn  MJ, Smith  HA, Chimowitz  MI.  A standardized method for measuring intracranial arterial stenosis.  AJNR Am J Neuroradiol. 2000;21(4):643-646.PubMedGoogle Scholar
8.
Chimowitz  MI, Lynn  MJ, Howlett-Smith  H,  et al; Warfarin-Aspirin Symptomatic Intracranial Disease Trial Investigators.  Comparison of warfarin and aspirin for symptomatic intracranial arterial stenosis.  N Engl J Med. 2005;352(13):1305-1316. doi:10.1056/NEJMoa043033PubMedGoogle ScholarCrossref
9.
La Joie  R, Perrotin  A, Barré  L,  et al.  Region-specific hierarchy between atrophy, hypometabolism, and β-amyloid (Aβ) load in Alzheimer’s disease dementia.  J Neurosci. 2012;32(46):16265-16273. doi:10.1523/JNEUROSCI.2170-12.2012PubMedGoogle ScholarCrossref
10.
Saint-Aubert  L, Nemmi  F, Péran  P,  et al.  Comparison between PET template-based method and MRI-based method for cortical quantification of florbetapir (AV-45) uptake in vivo.  Eur J Nucl Med Mol Imaging. 2014;41(5):836-843. doi:10.1007/s00259-013-2656-8PubMedGoogle ScholarCrossref
11.
Wong  DF, Rosenberg  PB, Zhou  Y,  et al.  In vivo imaging of amyloid deposition in Alzheimer disease using the radioligand 18F-AV-45 (florbetapir [corrected] F 18).  J Nucl Med. 2010;51(6):913-920. doi:10.2967/jnumed.109.069088PubMedGoogle ScholarCrossref
12.
Sharrett  AR, Patsch  W, Sorlie  PD, Heiss  G, Bond  MG, Davis  CE.  Associations of lipoprotein cholesterols, apolipoproteins A-I and B, and triglycerides with carotid atherosclerosis and coronary heart disease: the Atherosclerosis Risk in Communities (ARIC) Study.  Arterioscler Thromb. 1994;14(7):1098-1104. doi:10.1161/01.ATV.14.7.1098PubMedGoogle ScholarCrossref
13.
Siedel  J, Hägele  EO, Ziegenhorn  J, Wahlefeld  AW.  Reagent for the enzymatic determination of serum total cholesterol with improved lipolytic efficiency.  Clin Chem. 1983;29(6):1075-1080.PubMedGoogle ScholarCrossref
14.
Qiao  Y, Suri  FK, Zhang  Y,  et al.  Racial differences in prevalence and risk for intracranial atherosclerosis in a US community-based population.  JAMA Cardiol. 2017;2(12):1341-1348. doi:10.1001/jamacardio.2017.4041PubMedGoogle ScholarCrossref
15.
Gottesman  RF, Albert  MS, Alonso  A,  et al.  Associations between midlife vascular risk factors and 25-year incident dementia in the Atherosclerosis Risk in Communities (ARIC) cohort.  JAMA Neurol. 2017;74(10):1246-1254. doi:10.1001/jamaneurol.2017.1658PubMedGoogle ScholarCrossref
16.
Roher  AE, Tyas  SL, Maarouf  CL,  et al.  Intracranial atherosclerosis as a contributing factor to Alzheimer’s disease dementia.  Alzheimers Dement. 2011;7(4):436-444. doi:10.1016/j.jalz.2010.08.228PubMedGoogle ScholarCrossref
17.
Sacco  RL, Kargman  DE, Gu  Q, Zamanillo  MC.  Race-ethnicity and determinants of intracranial atherosclerotic cerebral infarction: the Northern Manhattan Stroke Study.  Stroke. 1995;26(1):14-20. doi:10.1161/01.STR.26.1.14PubMedGoogle ScholarCrossref
18.
Bannai  T, Mano  T, Chen  X,  et al.  Chronic cerebral hypoperfusion shifts the equilibrium of amyloid β oligomers to aggregation-prone species with higher molecular weight.  Sci Rep. 2019;9(1):2827. doi:10.1038/s41598-019-39494-7PubMedGoogle ScholarCrossref
19.
Zhai  Y, Yamashita  T, Nakano  Y,  et al.  Chronic cerebral hypoperfusion accelerates Alzheimer’s Disease pathology with cerebrovascular remodeling in a novel mouse model.  J Alzheimers Dis. 2016;53(3):893-905. doi:10.3233/JAD-160345PubMedGoogle ScholarCrossref
20.
Gupta  A, Iadecola  C.  Impaired Aβ clearance: a potential link between atherosclerosis and Alzheimer’s disease.  Front Aging Neurosci. 2015;7:115. doi:10.3389/fnagi.2015.00115PubMedGoogle ScholarCrossref
21.
Iadecola  C, Gottesman  RF.  Neurovascular and cognitive dysfunction in hypertension.  Circ Res. 2019;124(7):1025-1044. doi:10.1161/CIRCRESAHA.118.313260PubMedGoogle ScholarCrossref
22.
Thomas  T, Thomas  G, McLendon  C, Sutton  T, Mullan  M.  beta-Amyloid-mediated vasoactivity and vascular endothelial damage.  Nature. 1996;380(6570):168-171. doi:10.1038/380168a0PubMedGoogle ScholarCrossref
23.
Van Nostrand  WE.  The influence of the amyloid ß-protein and its precursor in modulating cerebral hemostasis.  Biochim Biophys Acta. 2016;1862(5):1018-1026. doi:10.1016/j.bbadis.2015.10.020PubMedGoogle ScholarCrossref
24.
Gottesman  RF, Schneider  ALC, Albert  M,  et al.  Midlife hypertension and 20-year cognitive change: the Atherosclerosis Risk in Communities Neurocognitive Study.  JAMA Neurol. 2014;71(10):1218-1227. doi:10.1001/jamaneurol.2014.1646PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    December 20, 2019

    Association of Intracranial Atherosclerotic Disease With Brain β-Amyloid Deposition: Secondary Analysis of the ARIC Study

    Author Affiliations
    • 1Department of Neurology, Johns Hopkins University, Baltimore, Maryland
    • 2Department of Epidemiology, Johns Hopkins University, Baltimore, Maryland
    • 3Department of Medicine, University of Mississippi Medical Center, Jackson
    • 4Department of Neurology, Mayo Clinic, Rochester, Minnesota
    • 5Department of Neurology, Mount Sinai Medical Center, New York, New York
    • 6Department of Radiology, Johns Hopkins University, Baltimore, Maryland
    • 7Division of Public Health Sciences, Wake Forest School of Medicine, Winston-Salem, North Carolina
    • 8Department of Internal Medicine, Wake Forest School of Medicine, Winston-Salem, North Carolina
    • 9Department of Neurology, Beth Israel Deaconness Medical Center, Harvard Medical School, Boston, Massachusetts
    JAMA Neurol. 2020;77(3):350-357. doi:10.1001/jamaneurol.2019.4339
    Key Points

    Question  Is intracranial atherosclerotic disease (ICAD) cross-sectionally associated with brain β-amyloid in individuals without dementia from US communities?

    Findings  In this cross-sectional study, 300 participants of the Atherosclerosis Risk in Communities positron emission tomography (PET) study completed florbetapir PET scans and magnetic resonance imaging high-resolution vessel wall imaging. In this sample, ICAD, including atherosclerotic plaque characteristics and measures of stenosis, was not associated with brain β-amyloid by PET.

    Meaning  In individuals without dementia from US communities, neither ICAD features nor the degree of stenosis are associated with brain β-amyloid when studied cross-sectionally.

    Abstract

    Importance  Intracranial atherosclerotic disease (ICAD) is an important cause of stroke and has also been recently identified as an important risk factor for all-cause dementia, but the mechanism of its association with cognitive performance is not fully understood.

    Objective  To test the hypothesis that ICAD is associated with cerebral β-amyloid deposition as a marker of Alzheimer disease.

    Design, Setting, and Participants  This cross-sectional analysis of data collected from August 2011 through November 2014 was a community-based cohort study conducted in 3 US communities. Of 346 adults without dementia aged 70 to 90 years who were sequentially recruited from 3 of 4 sites of the larger Atherosclerosis Risk in Communities study into a study of brain florbetapir positron emission tomography (ARIC-PET), 300 met inclusion criteria. A total of 589 were approached about recruitment, of whom 346 (58.7%) consented (the remainder either met exclusion criteria for ARIC-PET or refused to participate). Data were analyzed from July 2017 through October 2019.

    Exposures  Intracranial atherosclerotic disease presence, frequency, and extent of stenosis, by high-resolution vessel wall magnetic resonance imaging.

    Main Outcomes and Measures  Global cortical standardized uptake value ratio (SUVR) of greater than 1.2 as measured by florbetapir PET. Models were conducted using logistic regression methods. In secondary analyses, we tested effect modifications by apolipoprotein E ε4 genotype with interaction terms and in stratified models and evaluated regional patterns of associations.

    Results  In 300 participants (mean [SD] age, 76 [5] years; 132 African American individuals [44%], 167 women [56%], and 94 carriers of at least 1 apolipoprotein E ε4 allele [31%]), ICAD was found in 105 participants (35%) and mean (SD) SUVR was higher in individuals with vs without intracranial plaques (1.34 [0.29] vs 1.27 [0.23]; P = .03). In adjusted models, ICAD presence (plaque presence [adjusted odds ratio (aOR), 1.20; 95% CI, 0.69-2.07] and frequency [aOR, 1.10; 95% CI, 0.96-1.26]) was not associated significantly with elevated SUVR in the total sample. Furthermore, modest stenosis of the intracranial vessels (defined as >50% stenosis) was not associated with elevated SUVR (aOR, 2.33; 95% CI, 0.82-6.60).

    Conclusions and Relevance  In this community-based cohort of adults without dementia, intracranial atherosclerotic plaque or stenosis was not associated with brain β-amyloid deposition.

    Introduction

    Intracranial atherosclerotic disease (ICAD) is an important cause of stroke, but its association with cognitive impairment and dementia is less well established. Although autopsy series have demonstrated associations between Circle of Willis atherosclerosis and dementia, including neuropathologic changes specific for Alzheimer disease (AD),1 the ability to evaluate intracranial plaque in vivo, before death, has been limited until recently.

    High-resolution vessel wall imaging with black-blood magnetic resonance imaging (MRI)2 allows the visualization of vessel wall characteristics. In the Atherosclerosis Risk in Communities Neurocognitive Study (ARIC-NCS), the presence and location of ICAD using this technology was associated with adjudicated mild cognitive impairment (MCI) and dementia. Whether this association reflects AD neuropathology or exclusively the vascular components of MCI and dementia is less clear. Vascular risk factors, when measured in midlife, were associated with elevated β-amyloid deposition (which, by leading hypotheses, is a major factor in the development of AD), measured by positron emission tomography (PET), which indicates a potential direct association between vascular risk factors and AD neuropathology.3

    In this study, in individuals from the ARIC-PET substudy of ARIC-NCS, in which individuals without dementia from 3 US communities completed florbetapir PET scans, we tested the cross-sectional associations between black-blood MRI characteristics of ICAD and β-amyloid deposition. We hypothesized that ICAD, indicated by plaque presence and number, would be associated with more brain β-amyloid independent of other vascular risk factors and these associations would follow regional patterns. Furthermore, in a secondary analysis, we hypothesized that the association between the two would be the strongest among carriers of the apolipoprotein E (APOE) ε4 gene.

    Methods

    ARIC-PET is an ancillary to ARIC and its substudy, the ARIC-NCS study. The ARIC-NCS study included brain MRI, including high-resolution vessel wall imaging, in a subset of participants and ARIC-PET recruited from this group. Each affiliated institution’s institutional review board approved the study and participants signed informed consent documents for each visit.

    Participant Inclusion

    ARIC was initiated in 1987 to 1989 when 15 792 individuals from 4 US communities (Washington County, Maryland; Jackson, Mississippi; Forsyth County, North Carolina; and the suburbs of Minneapolis, Minnesota) were recruited. As of article preparation, they have completed 6 in-person visits, including at baseline, and most recently in 2016 to 2017; for this analysis, only data from visit 5 (the ARIC-NCS visit [2011-2013]) and the ARIC-PET visit (2012-2014) were used. In addition to these visits, annual (and now semiannual) telephone calls are made, which include surveillance for hospitalizations and dementia.

    As part of ARIC-NCS, 1980 participants underwent a brain MRI as described.4,5 In brief, individuals with a prior ARIC brain MRI, with evidence of cognitive decline (by change in cognitive score or low performance on a series of cognitive tests), or from an age-stratified sample of cognitively normal individuals were invited to receive brain MRI. ARIC-PET recruited, from the Washington County, Forsyth County, and Jackson sites, 346 of these individuals after further excluding for dementia (by expert adjudication, Clinical Dementia Rating sum-of-boxes score of >3, Functional Activities Questionnaire score of >5, or Mini-Mental State Examination of <19 for African American participants or <21 for white participants), heavy alcohol use, renal dysfunction, or prolonged corrected QT interval.6 These 346 resulted from screening of 589, of whom 110 met these exclusionary criteria and the remaining 132 refused to participate.

    Neuroimaging Protocol

    Brain MRI scans were obtained on 3-T scanners at facilities near each field center, with structural sequences read centrally at the Mayo Clinic (Rochester, Minnesota) and vessel images read at the Johns Hopkins MRI Reading Center. Magnetization-prepared 180° radio frequency pulses and rapid-gradient echo sequences were used to coregister the PET images. The 3-dimensional time-of-flight magnetic resonance angiogram (MRA) (acquired resolution, 0.50 × 0.50 × 0.55 mm3) and 3-dimensional high-resolution isotropic vessel wall imaging (acquired resolution, 0.5 mm3) were obtained. Individual vessels and their territories were reviewed,2,5 with atherosclerotic plaque in a vessel defined as eccentric wall thickening (with or without luminal stenosis by MRA), with the presence and number of ICAD plaques recorded along with markers of ICAD plaque burden (normalized wall index, defined for the most stenotic atherosclerotic plaque within each vascular territory). The degree of stenosis (none, <50%, 50%-70%, 71%-99%, and occlusion7) was measured on MRA based on published criteria.8 For this analysis, less than 50% was compared with 50% or more stenosis, with any stenosis of 50% or more referring to any vessel in any territory on either side; all individuals with stenosis had at least some ICAD plaque by definition.

    For regional comparisons of vascular territories, anterior circulation plaque and stenosis were defined as any of the following: internal carotid artery (ICA), middle cerebral artery (MCA), and anterior cerebral artery (ACA). Posterior circulation involvement included stenosis or plaques in the posterior cerebral artery, the basilar artery, or either vertebral artery.

    Florbetapir PET scans were obtained within 1 year of MRI scans at PET facilities near the field centers. After coregistration with MRI, PET scans were read centrally at the Johns Hopkins PET reading center. Standardized uptake value ratios (SUVRs) were calculated for cortical regions of interest, with a global cortical measurement calculating a weighted (based on region of interest size) average of the orbitofrontal, prefrontal, and superior frontal cortices; lateral temporal, parietal, and occipital lobes; and precuneus, anterior cingulate, and posterior cingulate. Cerebellar gray matter was used as an automated reference region.6,9-11 A scan was positive for β-amyloid if the global cortical SUVR was more than 1.2 (the sample median).3,6 Other regional SUVRs were considered, each dichotomized at that region’s median as described later in this article.

    Covariates

    Covariates were considered at visit 5 unless otherwise specified. Sex, race, and educational level (<high school (HS)/HS or graduate equivalency degree/>HS) were self-reported at the ARIC baseline visit, as was date of birth, with age calculated at subsequent visits. Race was explored because of prior reports of disparities in dementia rates and amyloid positivity, as well as intracranial stenosis frequency. Hypertension was defined as a measured systolic blood pressure of more than 140 or more than 90 mm Hg (diastolic) or use of an antihypertensive medication. Diabetes was defined as a measured fasting glucose level of 126 mg/dL or greater (to convert to millimoles per liter, multiply by 0.0555), nonfasting glucose levels of 200 mg/dL or more, or a self-report of physician-diagnosed diabetes or use of oral diabetes medications or insulin. Smoking status (current/former/never) was self-reported. Body mass index was calculated as weight in kilograms divided by height in meters squared; hypercholesterolemia was defined as a measured12,13 total cholesterol level of more than 200 mg/dL (to convert to millimoles per liter, multiply by 0.0259) and statin use was also considered (yes/no). Apolipoprotein E (APOE), previously genotyped in ARIC, was considered as 0 vs 1 or 2 ε4 alleles.

    Statistical Analysis

    Group comparisons were evaluated using descriptive statistics. Groups defined by ICAD features were compared using 2-sample t tests (for continuous variables) and χ2 tests (for categorical variables). Logistic regression models considered global cortical SUVR positivity (>1.2) as the primary dependent variable, with measures of ICAD in separate models that were adjusted first for demographics (age, sex, educational level, and race) and additionally for vascular risk factors and APOE genotype. In secondary analyses, regional SUVR measurements were considered with regional vessel wall abnormalities. For example, anterior circulation ICAD (defined previously, plus ACA specifically) was considered in association with elevated SUVR in frontal cortices (ie, superior frontal, prefrontal, and orbitofrontal cortices) and ICAD in the middle cerebral artery territory (anterior circulation and MCA specifically) was considered with elevated SUVR in MCA territory regions (ie, medial temporal cortex, lateral temporal cortex, posterior precuneus, parietal lobe, and the anterior and posterior cingulate) without separate consideration of laterality (eg, MCA involvement on either side was analyzed with the mean average medial temporal cortical SUVR). Posterior circulation and posterior cerebral artery ICAD were each considered in association with elevated occipital lobe SUVR. Positivity in each region was defined as an SUVR greater than each region’s median; continuous SUVR value was also compared between ICAD-defined groups.

    In additional secondary analyses, the effect modification of the primary associations (β-amyloid PET on ICAD) by APOE, race, and sex was tested, with presentation of stratified models when an interaction term was significant at P < .10; otherwise, P < .05 was considered statistically significant. Stata IC, version 15.1 for Macintosh (StataCorp; Apple) was used.

    Results

    Evaluable high-resolution vessel wall imaging was available for 300 ARIC-PET participants (Table 1). In general, participants with ICAD of any vessel on intracranial imaging, which constituted 105 (35%) of the sample, were older, more likely to be men, and had nonsignificantly higher rates of hypertension, diabetes, and former (but not current) smoking. The median number of intracranial atherosclerotic plaques was 0 (range, 0-15); for individuals with at least 1 ICAD plaque, the median number was 1 (range, 1-15) atherosclerotic plaque, but with 3 individuals having more than 10 total plaques (Figure 1). Of the 105 participants with at least 1 ICAD plaque, 22 (21.0%) had least 1 vessel with more than 50% stenosis (of whom 9 [40.9%] had >50% stenosis in the anterior circulation) and only 4 (3.8%) had at least 1 vessel with 70% or more stenosis. Eighty-five participants (81%) with an ICAD plaque had at least some measurable stenosis, leaving 20 (19%) of the cohort with ICAD but no stenosis by MRA.

    Elevated Global β-Amyloid and ICAD

    The mean global cortical SUVR was greater in individuals with at least 1 intracranial atherosclerotic plaque (1.34) compared with those without (1.27) (P = .03). However, no significant association was found in minimally or more fully adjusted models considering several ICAD features and elevation of β-amyloid SUVR (Table 2). The ICAD plaque burden, characterized as normalized wall index in the most stenotic vessel in a vascular territory, was not significantly associated with β-amyloid SUVR (adjusted odds ratio [OR], 1.02; 95% CI, 0.98-1.07).

    Secondary Analyses: Stratification and Effect Modification by APOE

    When models were considered with stratification by APOE status (Table 3), based on a priori hypotheses about a potential interaction, a significant (P < .05) effect modification was identified, with stronger associations for APOE ε4 carriers than noncarriers; yet, no individual stratified model reached statistical significance. For the presence of plaques in a posterior circulation artery, the odds of elevated global cortical β-amyloid among ε4 carriers was markedly higher, although imprecise (adjusted OR, 7.14; 95% CI, 0.81-63.08), and among noncarriers no evidence of an association remained (adjusted OR, 0.85; 95% CI, 0.41-1.76; P for interaction = .03).

    Localization of Plaques and β-Amyloid Deposition

    Regional SUVR values were generally higher for individuals with any intracranial vessel plaque compared with those without (Figure 2), but regional plaque and regional SUVR associations were mostly limited to those involving the anterior circulation (anterior circulation plaque or ACA plaque involvement) and the posterior circulation, with fewer associations seen for regions within the MCA territory. When these associations were explored in demographic-adjusted models, with a binary classification of positive vs negative, several associations remained statistically significant. The presence of stenosis in a vessel supplying the MCA (ICA or MCA) was associated with a 2.53-times higher odds of β-amyloid positivity in the lateral temporal lobe (95% CI, 1.12, 5.73) in a fully adjusted model. Similarly, stenosis in the ACA territory (ACA or ICA) was associated with more than a doubling of elevated odds of β-amyloid positivity in the superior frontal cortex, a region supplied by these vessels (adjusted OR, 2.20; 95% CI, 1.10, 4.43).

    Effect Modification by Race and Sex

    In general, associations did not differ between ICAD and β-amyloid SUVR by racial group with the exception of presence of stenosis, which, when present, was generally a stronger risk factor for elevated β-amyloid SUVR in white individuals vs African American individuals, but without statistical evidence of effect modification. The presence of any stenosis of 50% or more was nonsignificantly associated with higher odds of elevated SUVR in white individuals (demographic-adjusted OR, 5.15; 95% CI, 0.98-27.14) but less so in African American individuals (OR, 1.43; 95% CI, 0.39-5.27; P for interaction = .32). Associations between ICAD characteristics and presence and elevated SUVR did not differ by sex.

    Discussion

    Vascular risk factors may act on cognitive decline and dementia through various mechanisms. Understanding these mechanisms may help identify targets and potential points for intervention and prevention. In ARIC, ICAD was identified as a risk factor for dementia,2 and many risk factors for ICAD14 have been associated not only with dementia15 but also with β-amyloid deposition,3 a likely key component in the development of AD. However, in this study, we found no direct evidence of an association between ICAD and brain β-amyloid in AD-prone regions in individuals without dementia from US communities.

    Beyond traditional risk factors, ICAD provides evidence of specific vascular damage. Furthermore, it represents one of the few areas where, on pathology, an association has been demonstrated between a vascular marker and brain β-amyloid: individuals with pathologically proven AD have more ICAD on autopsy than non-AD decedents,16 and this association is independent of demographics and APOE genotype.1 Evidence of an association between ICAD markers and brain β-amyloid in individuals without dementia might point to a more specific mechanism by which vascular risk factors might act. In addition to the importance of studying ICAD and brain β-amyloid more broadly, this study allowed us to study a potential mechanism for the observed racial/ethnic differences in β-amyloid deposition. In ARIC14 and in clinical populations,17 ICAD is observed at higher frequencies in African American individuals than in white individuals. Thus, it represents a potential mechanism for the previously reported racial/ethnic disparities in amyloid rates,6 where African American ARIC participants without dementia had higher SUVR levels even after adjustment for other covariates and demographics than white ARIC participants without dementia, and perhaps even ultimately in dementia rates (observed in our own study15 and elsewhere). However, in this study, we found no evidence of differential associations between ICAD and β-amyloid by race/ethnicity.

    Potential mechanisms for an association between ICAD and β-amyloid may depend on a greater degree of stenosis than what was found in our generally healthy cohort or even with a more cognitively impaired population if relevant at later stages in disease. Hemodynamically, stenosis could lead to hypoperfusion, which might not only result from ICAD but also contribute to ICAD and β-amyloid; hypoperfusion contributes to increased aggregation of Aβ oligomers18 and, chronically, to cerebrovascular remodeling in amyloid precursor protein transgenic mice.19 As stated previously, few individuals in our population had a significant amount of stenosis present, with nearly 20% having ICAD but no detectable stenosis,14 and thus would not be expected to have significant hypoperfusion from the ICAD itself. It is possible that this lack of hemodynamically significant stenosis might explain the lack of an association in our cohort. However, other mechanisms would still be relevant even in the absence of this degree of stenosis. Our use of high-resolution vessel wall MRI allows highly sensitive measures of plaque presence and burden even in the absence of luminal narrowing. Atherosclerosis in the intracranial vessels might also act on β-amyloid deposition via reduced clearance,20 perhaps via changes in arterial stiffness or other microvascular dysfunction, or might increase inflammation or oxidative stress, which can contribute to increased Aβ deposition or reduced clearance.21 In general, the co-occurrence of ICAD and elevated β-amyloid and the cross-sectional nature of our study make it difficult to evaluate the direction of a possible association, but by studying individuals without dementia, we have reduced the chances that the observed findings are due to reverse causation. However, it is recognized that β-amyloid is toxic to the endothelium, leading to changes in vasoactivity22,23 and potentially priming the vasculature for the development of ICAD. The chronology of these changes may be better elucidated in the future, with continued longitudinal follow-up in ARIC-PET. Our population was also mostly individuals with normal cognition, with fewer than a quarter with MCI; we may have failed to find an association between ICAD and amyloid if ICAD is relevant later in the disease course and thus is not relevant for amyloid deposition in the cognitively normal population.

    Limitations

    This study is primarily limited by its cross-sectional nature, with evaluation of ICAD only at the time of a single measurement of β-amyloid deposition. Because clinical vascular risk factors are most important for cognitive trajectories and β-amyloid deposition in middle age,3,15,24 with less to no role in late life, it is plausible that ICAD is a more important risk factor from earlier in life, which we could not evaluate. In addition, ARIC-PET is a study of survivors: individuals with dementia were excluded, and thus individuals with the most severe atherosclerosis might have died of other associated disease earlier in the study or may have developed dementia and thus been excluded, particularly in APOE ε4 carriers. Other limitations include the inability to fully adjust for potential confounders. Additionally, we defined β-amyloid positivity using the sample median, globally and for each region of interest. A clearly standardized cut point has not been defined in the field, but our prior work has shown that results in other studies were similar when we considered different cut points.3,6 Although it is standard in the field to dichotomize β-amyloid because of the skewness of these data and a likely clinical threshold effect, we acknowledge that an association with our ICAD measures might have been found across the range of SUVR values were they considered in a nonbinary fashion. Finally, although we found several interesting observations and associations in APOE-stratified subgroups and when we evaluated regional patterns, we interpret these with caution, given the lack of an overall finding in the total population.

    Conclusions

    We found no evidence of an association between plaque presence and global cerebral β-amyloid in individuals without dementia broadly. Furthermore, our data suggest that there may be regional associations between ICAD and brain β-amyloid. Further evaluating the potential role of ICAD as a risk factor or marker of elevated risk of β-amyloid deposition, itself a marker of AD neuropathology, is needed in further studies.

    Back to top
    Article Information

    Accepted for Publication: October 17, 2019.

    Published Online: December 20, 2019. doi:10.1001/jamaneurol.2019.4339

    Correction: This article was corrected on March 9, 2020, to fix a data error in the Results section.

    Corresponding Author: Rebecca F. Gottesman, MD, PhD, The Johns Hopkins Hospital, 600 N Wolfe St, Phipps 446D, Baltimore, MD 21287 (rgottesm@jhmi.edu).

    Author Contributions: Dr Gottesman had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Gottesman, Wagenknecht, Wasserman.

    Acquisition, analysis, or interpretation of data: Gottesman, Mosley, Knopman, Hao, Wong, Hughes, Qiao, Dearborn-Tomazos, Wasserman.

    Drafting of the manuscript: Gottesman, Wong.

    Critical revision of the manuscript for important intellectual content: Mosley, Knopman, Hao, Wagenknecht, Hughes, Qiao, Dearborn-Tomazos, Wasserman.

    Statistical analysis: Gottesman, Hughes.

    Obtained funding: Wagenknecht, Wasserman.

    Administrative, technical, or material support: Wong, Hughes, Qiao, Wasserman.

    Supervision: Mosley, Wagenknecht, Wasserman.

    Conflict of Interest Disclosures: Dr Gottesman reported grants from the National Institutes of Health (NIH)/National Institute on Aging (NIA) and serves as an associate editor for Neurology, which is published by the American Academy of Neurology. Drs Mosley and Wagenknecht reported grants from NIH during the conduct of the study. Dr Knopman reported personal fees from DIAN TU and grants from Biogen and Lilly. Dr Wong reported grants and nonfinancial support from AVID, Lilly, and Roche Neuroscience and grants from Five Eleven Pharma and Lundbeck. Dr Dearborn-Tomazos reported personal fees from Bristol-Myers. Dr Wasserman reported grants from the NIH (National Heart, Lung, and Blood Institute [NHLBI]) during the conduct of the study and had a patent to 13/922,111 issued. No other disclosures were reported.

    Funding/Support: The Atherosclerosis Risk in Communities (ARIC) study has been funded in whole or in part with federal funds from the NHLBI, NIH, and the US Department of Health and Human Services, under contracts HHSN268201700001I, HHSN268201700002I, HHSN268201700003I, HHSN268201700005I, and HHSN268201700004I. The ARIC-PET study is funded by the NIA (R01AG040282) and Dr Gottesman is supported by NIH/NIA grant K24 AG052573. Neurocognitive data were collected by grants U01 HL096812, HL096814, HL096899, HL096902, HL096917, and the National Institute of Neurological Disorders and Stroke, with previous brain magnetic resonance imaging examinations funded by grants R01-HL70825 (from the NHLBI). Intracranial vessel imaging was funded by NIH grants RO1HL105930, RO1HL105626, and K99HL106232.

    Role of the Funder/Sponsor: Avid Radiopharmaceuticals provided the florbetapir isotope for the study but had no role in the study design or interpretation of results. The other funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Additional Contributions: We thank the staff and participants of the ARIC study for their important contributions. Participants received a nominal reimbursement fee for participating in parts of the study and staff were compensated as study employees.

    References
    1.
    Beach  TG, Wilson  JR, Sue  LI,  et al.  Circle of Willis atherosclerosis: association with Alzheimer’s disease, neuritic plaques and neurofibrillary tangles.  Acta Neuropathol. 2007;113(1):13-21. doi:10.1007/s00401-006-0136-yPubMedGoogle ScholarCrossref
    2.
    Dearborn  JL, Zhang  Y, Qiao  Y,  et al.  Intracranial atherosclerosis and dementia: the Atherosclerosis Risk in Communities (ARIC) study.  Neurology. 2017;88(16):1556-1563. doi:10.1212/WNL.0000000000003837PubMedGoogle ScholarCrossref
    3.
    Gottesman  RF, Schneider  ALC, Zhou  Y,  et al.  Association between midlife vascular risk factors and estimated brain amyloid deposition.  JAMA. 2017;317(14):1443-1450. doi:10.1001/jama.2017.3090PubMedGoogle ScholarCrossref
    4.
    Knopman  DS, Griswold  ME, Lirette  ST,  et al.  Vascular imaging abnormalities and cognition: mediation by cortical volume in nondemented individuals: Atherosclerosis Risk in Communities-Neurocognitive Study.  Stroke. 2015;46(2):433-440. doi:10.1161/STROKEAHA.114.007847PubMedGoogle ScholarCrossref
    5.
    Qiao  Y, Guallar  E, Suri  FK,  et al.  MR imaging measures of intracranial atherosclerosis in a population-based study.  Radiology. 2016;280(3):860-868. doi:10.1148/radiol.2016151124PubMedGoogle ScholarCrossref
    6.
    Gottesman  RF, Schneider  ALC, Zhou  Y,  et al.  The ARIC-PET amyloid imaging study: brain amyloid differences by age, race, sex, and APOE.  Neurology. 2016;87(5):473-480. doi:10.1212/WNL.0000000000002914PubMedGoogle ScholarCrossref
    7.
    Samuels  OB, Joseph  GJ, Lynn  MJ, Smith  HA, Chimowitz  MI.  A standardized method for measuring intracranial arterial stenosis.  AJNR Am J Neuroradiol. 2000;21(4):643-646.PubMedGoogle Scholar
    8.
    Chimowitz  MI, Lynn  MJ, Howlett-Smith  H,  et al; Warfarin-Aspirin Symptomatic Intracranial Disease Trial Investigators.  Comparison of warfarin and aspirin for symptomatic intracranial arterial stenosis.  N Engl J Med. 2005;352(13):1305-1316. doi:10.1056/NEJMoa043033PubMedGoogle ScholarCrossref
    9.
    La Joie  R, Perrotin  A, Barré  L,  et al.  Region-specific hierarchy between atrophy, hypometabolism, and β-amyloid (Aβ) load in Alzheimer’s disease dementia.  J Neurosci. 2012;32(46):16265-16273. doi:10.1523/JNEUROSCI.2170-12.2012PubMedGoogle ScholarCrossref
    10.
    Saint-Aubert  L, Nemmi  F, Péran  P,  et al.  Comparison between PET template-based method and MRI-based method for cortical quantification of florbetapir (AV-45) uptake in vivo.  Eur J Nucl Med Mol Imaging. 2014;41(5):836-843. doi:10.1007/s00259-013-2656-8PubMedGoogle ScholarCrossref
    11.
    Wong  DF, Rosenberg  PB, Zhou  Y,  et al.  In vivo imaging of amyloid deposition in Alzheimer disease using the radioligand 18F-AV-45 (florbetapir [corrected] F 18).  J Nucl Med. 2010;51(6):913-920. doi:10.2967/jnumed.109.069088PubMedGoogle ScholarCrossref
    12.
    Sharrett  AR, Patsch  W, Sorlie  PD, Heiss  G, Bond  MG, Davis  CE.  Associations of lipoprotein cholesterols, apolipoproteins A-I and B, and triglycerides with carotid atherosclerosis and coronary heart disease: the Atherosclerosis Risk in Communities (ARIC) Study.  Arterioscler Thromb. 1994;14(7):1098-1104. doi:10.1161/01.ATV.14.7.1098PubMedGoogle ScholarCrossref
    13.
    Siedel  J, Hägele  EO, Ziegenhorn  J, Wahlefeld  AW.  Reagent for the enzymatic determination of serum total cholesterol with improved lipolytic efficiency.  Clin Chem. 1983;29(6):1075-1080.PubMedGoogle ScholarCrossref
    14.
    Qiao  Y, Suri  FK, Zhang  Y,  et al.  Racial differences in prevalence and risk for intracranial atherosclerosis in a US community-based population.  JAMA Cardiol. 2017;2(12):1341-1348. doi:10.1001/jamacardio.2017.4041PubMedGoogle ScholarCrossref
    15.
    Gottesman  RF, Albert  MS, Alonso  A,  et al.  Associations between midlife vascular risk factors and 25-year incident dementia in the Atherosclerosis Risk in Communities (ARIC) cohort.  JAMA Neurol. 2017;74(10):1246-1254. doi:10.1001/jamaneurol.2017.1658PubMedGoogle ScholarCrossref
    16.
    Roher  AE, Tyas  SL, Maarouf  CL,  et al.  Intracranial atherosclerosis as a contributing factor to Alzheimer’s disease dementia.  Alzheimers Dement. 2011;7(4):436-444. doi:10.1016/j.jalz.2010.08.228PubMedGoogle ScholarCrossref
    17.
    Sacco  RL, Kargman  DE, Gu  Q, Zamanillo  MC.  Race-ethnicity and determinants of intracranial atherosclerotic cerebral infarction: the Northern Manhattan Stroke Study.  Stroke. 1995;26(1):14-20. doi:10.1161/01.STR.26.1.14PubMedGoogle ScholarCrossref
    18.
    Bannai  T, Mano  T, Chen  X,  et al.  Chronic cerebral hypoperfusion shifts the equilibrium of amyloid β oligomers to aggregation-prone species with higher molecular weight.  Sci Rep. 2019;9(1):2827. doi:10.1038/s41598-019-39494-7PubMedGoogle ScholarCrossref
    19.
    Zhai  Y, Yamashita  T, Nakano  Y,  et al.  Chronic cerebral hypoperfusion accelerates Alzheimer’s Disease pathology with cerebrovascular remodeling in a novel mouse model.  J Alzheimers Dis. 2016;53(3):893-905. doi:10.3233/JAD-160345PubMedGoogle ScholarCrossref
    20.
    Gupta  A, Iadecola  C.  Impaired Aβ clearance: a potential link between atherosclerosis and Alzheimer’s disease.  Front Aging Neurosci. 2015;7:115. doi:10.3389/fnagi.2015.00115PubMedGoogle ScholarCrossref
    21.
    Iadecola  C, Gottesman  RF.  Neurovascular and cognitive dysfunction in hypertension.  Circ Res. 2019;124(7):1025-1044. doi:10.1161/CIRCRESAHA.118.313260PubMedGoogle ScholarCrossref
    22.
    Thomas  T, Thomas  G, McLendon  C, Sutton  T, Mullan  M.  beta-Amyloid-mediated vasoactivity and vascular endothelial damage.  Nature. 1996;380(6570):168-171. doi:10.1038/380168a0PubMedGoogle ScholarCrossref
    23.
    Van Nostrand  WE.  The influence of the amyloid ß-protein and its precursor in modulating cerebral hemostasis.  Biochim Biophys Acta. 2016;1862(5):1018-1026. doi:10.1016/j.bbadis.2015.10.020PubMedGoogle ScholarCrossref
    24.
    Gottesman  RF, Schneider  ALC, Albert  M,  et al.  Midlife hypertension and 20-year cognitive change: the Atherosclerosis Risk in Communities Neurocognitive Study.  JAMA Neurol. 2014;71(10):1218-1227. doi:10.1001/jamaneurol.2014.1646PubMedGoogle ScholarCrossref
    ×