Show simple item record

dc.contributor.authorLambert, Christian
dc.contributor.authorZeestraten, Eva
dc.contributor.authorWilliams, Owen
dc.contributor.authorBenjamin, Philip
dc.contributor.authorLawrence, Andrew J
dc.contributor.authorMorris, Robin G
dc.contributor.authorMackinnon, Andrew D
dc.contributor.authorBarrick, Thomas R
dc.contributor.authorMarkus, Hugh S
dc.date.accessioned2018-10-18T10:20:47Z
dc.date.available2018-10-18T10:20:47Z
dc.date.issued2018
dc.identifier.issn2213-1582
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/284105
dc.description.abstractSporadic cerebral small vessel disease is an important cause of vascular dementia, a syndrome of cognitive impairment together with vascular brain damage. At post-mortem pure vascular dementia is rare, with evidence of co-existing Alzheimer's disease pathology in 95% of cases. This work used MRI to characterize structural abnormalities during the preclinical phase of vascular dementia in symptomatic small vessel disease. 121 subjects were recruited into the St George's Cognition and Neuroimaging in Stroke study and followed up longitudinally for five years. Over this period 22 individuals converted to dementia. Using voxel-based morphometry, we found structural abnormalities present at baseline in those with preclinical dementia, with reduced grey matter density in the left striatum and hippocampus, and more white matter hyperintensities in the frontal white-matter. The lacunar data revealed that some of these abnormalities may be due to lesions within the striatum and centrum semiovale. Using support vector machines, future dementia could be best predicted using hippocampal and striatal Jacobian determinant data, achieving a balanced classification accuracy of 73%. Using cluster ward linkage we identified four anatomical subtypes. Successful predictions were restricted to groups with lower levels of vascular damage. The subgroup that could not be predicted were younger, further from conversion, had the highest levels of vascular damage, with milder cognitive impairment at baseline but more rapid deterioration in processing speed and executive function, consistent with a primary vascular dementia. In contrast, the remaining groups had decreasing levels of vascular damage and increasing memory impairment consistent with progressively more Alzheimer's-like pathology. Voxel-wise rates of hippocampal atrophy supported these distinctions, with the vascular group closely resembling the non-dementing cohort, whereas the Alzheimer's like group demonstrated global hippocampal atrophy. This work reveals distinct anatomical endophenotypes in preclinical vascular dementia, forming a spectrum between vascular and Alzheimer's like pathology. The latter group can be identified using baseline MRI, with 73% converting within 5 years. It was not possible to predict the vascular dominant dementia subgroup, however 19% of negative predictions with high levels of vascular disease would ultimately develop dementia. It may be that techniques more sensitive to white matter damage, such as diffusion weighted imaging, may prove more useful for this vascular dominant subgroup in the future. This work provides a way to accurately stratify patients using a baseline MRI scan, and has utility in future clinical trials designed to slow or prevent the onset of dementia in these high-risk cohorts.
dc.format.mediumElectronic-eCollection
dc.languageeng
dc.publisherElsevier BV
dc.rightsAttribution 4.0 International
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectBrain
dc.subjectHumans
dc.subjectDementia, Vascular
dc.subjectMagnetic Resonance Imaging
dc.subjectEarly Diagnosis
dc.subjectNeuropsychological Tests
dc.subjectAged
dc.subjectAged, 80 and over
dc.subjectMiddle Aged
dc.subjectFemale
dc.subjectMale
dc.subjectNeuroimaging
dc.subjectCerebral Small Vessel Diseases
dc.subjectWhite Matter
dc.titleIdentifying preclinical vascular dementia in symptomatic small vessel disease using MRI.
dc.typeArticle
prism.endingPage938
prism.publicationDate2018
prism.publicationNameNeuroimage Clin
prism.startingPage925
prism.volume19
dc.identifier.doi10.17863/CAM.31476
dcterms.dateAccepted2018-06-17
rioxxterms.versionofrecord10.1016/j.nicl.2018.06.023
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserved
rioxxterms.licenseref.startdate2018-01
dc.contributor.orcidMarkus, Hugh [0000-0002-9794-5996]
dc.identifier.eissn2213-1582
rioxxterms.typeJournal Article/Review
cam.issuedOnline2018-06-20


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

Attribution 4.0 International
Except where otherwise noted, this item's licence is described as Attribution 4.0 International