Abstract

Cognitive reserve (CR) explains interindividual differences in the ability to maintain cognitive function in the presence of neuropathology. We developed a neuroimaging approach including a measure of brain atrophy and cognition to capture this construct. In a group of 511 Alzheimer's disease (AD) biomarker-positive subjects in different stages across the disease spectrum, we performed 3T magnetic resonance imaging and predicted gray matter (GM) volume in each voxel based on cognitive performance (i.e. a global cognitive composite score), adjusted for age, sex, disease stage, premorbid brain size (i.e. intracranial volume) and scanner type. We used standardized individual differences between predicted and observed GM volume (i.e. W-scores) as an operational measure of CR. To validate this method, we showed that education correlated with mean W-scores in whole-brain (r = −0.090, P < 0.05) and temporoparietal (r = −0.122, P < 0.01) masks, indicating that higher education was associated with more CR (i.e. greater atrophy than predicted from cognitive performance). In a voxel-wise analysis, this effect was most prominent in the right inferior and middle temporal and right superior lateral occipital cortex (P < 0.05, corrected for multiple comparisons). Furthermore, survival analyses among subjects in the pre-dementia stage revealed that the W-scores predicted conversion to more advanced disease stages (whole-brain: hazard ratio [HR] = 0.464, P < 0.05; temporoparietal: HR = 0.397, P < 0.001). Our neuroimaging approach captures CR with high anatomical detail and at an individual level. This standardized method is applicable to various brain diseases or CR proxies and can flexibly incorporate different neuroimaging modalities and cognitive parameters, making it a promising tool for scientific and clinical purposes. Hum Brain Mapp 38:4703–4715, 2017.

Original languageEnglish
Pages (from-to)4703-4715
Number of pages13
JournalHuman Brain Mapping
Volume38
Issue number9
DOIs
Publication statusPublished - 1 Sep 2017

Cite this

@article{e23a61d536254be09a38f08a26af2734,
title = "A neuroimaging approach to capture cognitive reserve: Application to Alzheimer's disease",
abstract = "Cognitive reserve (CR) explains interindividual differences in the ability to maintain cognitive function in the presence of neuropathology. We developed a neuroimaging approach including a measure of brain atrophy and cognition to capture this construct. In a group of 511 Alzheimer's disease (AD) biomarker-positive subjects in different stages across the disease spectrum, we performed 3T magnetic resonance imaging and predicted gray matter (GM) volume in each voxel based on cognitive performance (i.e. a global cognitive composite score), adjusted for age, sex, disease stage, premorbid brain size (i.e. intracranial volume) and scanner type. We used standardized individual differences between predicted and observed GM volume (i.e. W-scores) as an operational measure of CR. To validate this method, we showed that education correlated with mean W-scores in whole-brain (r = −0.090, P < 0.05) and temporoparietal (r = −0.122, P < 0.01) masks, indicating that higher education was associated with more CR (i.e. greater atrophy than predicted from cognitive performance). In a voxel-wise analysis, this effect was most prominent in the right inferior and middle temporal and right superior lateral occipital cortex (P < 0.05, corrected for multiple comparisons). Furthermore, survival analyses among subjects in the pre-dementia stage revealed that the W-scores predicted conversion to more advanced disease stages (whole-brain: hazard ratio [HR] = 0.464, P < 0.05; temporoparietal: HR = 0.397, P < 0.001). Our neuroimaging approach captures CR with high anatomical detail and at an individual level. This standardized method is applicable to various brain diseases or CR proxies and can flexibly incorporate different neuroimaging modalities and cognitive parameters, making it a promising tool for scientific and clinical purposes. Hum Brain Mapp 38:4703–4715, 2017.",
keywords = "Alzheimer's disease, cognitive reserve, education, global cognition, magnetic resonance imaging (MRI), voxel-based morphometry",
author = "{van Loenhoud}, {Anna C.} and Wink, {Alle Meije} and Colin Groot and Verfaillie, {Sander C.J.} and Jos Twisk and Frederik Barkhof and {van Berckel}, Bart and Philip Scheltens and {van der Flier}, {Wiesje M.} and Rik Ossenkoppele",
year = "2017",
month = "9",
day = "1",
doi = "10.1002/hbm.23695",
language = "English",
volume = "38",
pages = "4703--4715",
journal = "Human Brain Mapping",
issn = "1065-9471",
publisher = "Wiley-Liss Inc.",
number = "9",

}

TY - JOUR

T1 - A neuroimaging approach to capture cognitive reserve

T2 - Application to Alzheimer's disease

AU - van Loenhoud, Anna C.

AU - Wink, Alle Meije

AU - Groot, Colin

AU - Verfaillie, Sander C.J.

AU - Twisk, Jos

AU - Barkhof, Frederik

AU - van Berckel, Bart

AU - Scheltens, Philip

AU - van der Flier, Wiesje M.

AU - Ossenkoppele, Rik

PY - 2017/9/1

Y1 - 2017/9/1

N2 - Cognitive reserve (CR) explains interindividual differences in the ability to maintain cognitive function in the presence of neuropathology. We developed a neuroimaging approach including a measure of brain atrophy and cognition to capture this construct. In a group of 511 Alzheimer's disease (AD) biomarker-positive subjects in different stages across the disease spectrum, we performed 3T magnetic resonance imaging and predicted gray matter (GM) volume in each voxel based on cognitive performance (i.e. a global cognitive composite score), adjusted for age, sex, disease stage, premorbid brain size (i.e. intracranial volume) and scanner type. We used standardized individual differences between predicted and observed GM volume (i.e. W-scores) as an operational measure of CR. To validate this method, we showed that education correlated with mean W-scores in whole-brain (r = −0.090, P < 0.05) and temporoparietal (r = −0.122, P < 0.01) masks, indicating that higher education was associated with more CR (i.e. greater atrophy than predicted from cognitive performance). In a voxel-wise analysis, this effect was most prominent in the right inferior and middle temporal and right superior lateral occipital cortex (P < 0.05, corrected for multiple comparisons). Furthermore, survival analyses among subjects in the pre-dementia stage revealed that the W-scores predicted conversion to more advanced disease stages (whole-brain: hazard ratio [HR] = 0.464, P < 0.05; temporoparietal: HR = 0.397, P < 0.001). Our neuroimaging approach captures CR with high anatomical detail and at an individual level. This standardized method is applicable to various brain diseases or CR proxies and can flexibly incorporate different neuroimaging modalities and cognitive parameters, making it a promising tool for scientific and clinical purposes. Hum Brain Mapp 38:4703–4715, 2017.

AB - Cognitive reserve (CR) explains interindividual differences in the ability to maintain cognitive function in the presence of neuropathology. We developed a neuroimaging approach including a measure of brain atrophy and cognition to capture this construct. In a group of 511 Alzheimer's disease (AD) biomarker-positive subjects in different stages across the disease spectrum, we performed 3T magnetic resonance imaging and predicted gray matter (GM) volume in each voxel based on cognitive performance (i.e. a global cognitive composite score), adjusted for age, sex, disease stage, premorbid brain size (i.e. intracranial volume) and scanner type. We used standardized individual differences between predicted and observed GM volume (i.e. W-scores) as an operational measure of CR. To validate this method, we showed that education correlated with mean W-scores in whole-brain (r = −0.090, P < 0.05) and temporoparietal (r = −0.122, P < 0.01) masks, indicating that higher education was associated with more CR (i.e. greater atrophy than predicted from cognitive performance). In a voxel-wise analysis, this effect was most prominent in the right inferior and middle temporal and right superior lateral occipital cortex (P < 0.05, corrected for multiple comparisons). Furthermore, survival analyses among subjects in the pre-dementia stage revealed that the W-scores predicted conversion to more advanced disease stages (whole-brain: hazard ratio [HR] = 0.464, P < 0.05; temporoparietal: HR = 0.397, P < 0.001). Our neuroimaging approach captures CR with high anatomical detail and at an individual level. This standardized method is applicable to various brain diseases or CR proxies and can flexibly incorporate different neuroimaging modalities and cognitive parameters, making it a promising tool for scientific and clinical purposes. Hum Brain Mapp 38:4703–4715, 2017.

KW - Alzheimer's disease

KW - cognitive reserve

KW - education

KW - global cognition

KW - magnetic resonance imaging (MRI)

KW - voxel-based morphometry

UR - http://www.scopus.com/inward/record.url?scp=85026843328&partnerID=8YFLogxK

U2 - 10.1002/hbm.23695

DO - 10.1002/hbm.23695

M3 - Article

VL - 38

SP - 4703

EP - 4715

JO - Human Brain Mapping

JF - Human Brain Mapping

SN - 1065-9471

IS - 9

ER -