TY - JOUR
T1 - Age-related differences in memory-encoding fMRI responses after accounting for decline in vascular reactivity
AU - Liu, Peiying
AU - Hebrank, Andrew C.
AU - Rodrigue, Karen M.
AU - Kennedy, Kristen M.
AU - Section, Jarren
AU - Park, Denise C.
AU - Lu, Hanzhang
N1 - Funding Information:
This study was supported in part by NIH R21 AG034318 (to H.L.), NIH R37 AG006265 (to D.C.P.), NIH R01 AG042753 (to H.L.), NIH R01 MH084021 (to H.L.), NIH R01 NS067015 (to H.L.), and NIH R21 NS078656 (to H.L.). The authors are grateful to Bela Bhatia and Blair Flicker for assistance with the experiments.
PY - 2013/9
Y1 - 2013/9
N2 - BOLD fMRI has provided a wealth of information about the aging brain. A common finding is that posterior regions of the brain manifest an age-related decrease in activation while the anterior regions show an age-related increase. Several neurocognitive models have been proposed to interpret these findings. However, one issue that has not been sufficiently considered to date is that the BOLD signal is based on vascular responses secondary to neural activity. Thus the above findings could be in part due to a vascular change, especially in view of the expected decline of vascular health with age. In the present study, we aim to examine age-related differences in memory-encoding fMRI response in the context of vascular aging. One hundred and thirty healthy subjects ranging from 20 to 89years old underwent a scene-viewing fMRI task and, in the same session, cerebrovascular reactivity (CVR) was measured in each subject using a CO2-inhalation task. Without accounting for the influence of vascular changes, the task-activated fMRI signal showed the typical age-related decrease in visual cortex and medial temporal lobe (MTL), but manifested an increase in the right inferior frontal gyrus (IFG). In the same individuals, an age-related CVR reduction was observed in all of these regions. We then used a previously proposed normalization approach to calculate a CVR-corrected fMRI signal, which was defined as the uncorrected signal divided by CVR. Based on the CVR-corrected fMRI signal, an age-related increase is now seen in both the left and right sides of IFG; and no brain regions showed a signal decrease with age. We additionally used a model-based approach to examine the fMRI data in the context of CVR, which again suggested an age-related change in the two frontal regions, but not in the visual and MTL regions.
AB - BOLD fMRI has provided a wealth of information about the aging brain. A common finding is that posterior regions of the brain manifest an age-related decrease in activation while the anterior regions show an age-related increase. Several neurocognitive models have been proposed to interpret these findings. However, one issue that has not been sufficiently considered to date is that the BOLD signal is based on vascular responses secondary to neural activity. Thus the above findings could be in part due to a vascular change, especially in view of the expected decline of vascular health with age. In the present study, we aim to examine age-related differences in memory-encoding fMRI response in the context of vascular aging. One hundred and thirty healthy subjects ranging from 20 to 89years old underwent a scene-viewing fMRI task and, in the same session, cerebrovascular reactivity (CVR) was measured in each subject using a CO2-inhalation task. Without accounting for the influence of vascular changes, the task-activated fMRI signal showed the typical age-related decrease in visual cortex and medial temporal lobe (MTL), but manifested an increase in the right inferior frontal gyrus (IFG). In the same individuals, an age-related CVR reduction was observed in all of these regions. We then used a previously proposed normalization approach to calculate a CVR-corrected fMRI signal, which was defined as the uncorrected signal divided by CVR. Based on the CVR-corrected fMRI signal, an age-related increase is now seen in both the left and right sides of IFG; and no brain regions showed a signal decrease with age. We additionally used a model-based approach to examine the fMRI data in the context of CVR, which again suggested an age-related change in the two frontal regions, but not in the visual and MTL regions.
KW - Age-related decrease
KW - Age-related increase
KW - Cerebral vascular reactivity (CVR)
KW - Cognitive aging
KW - FMRI normalization
KW - Functional magnetic resonance imaging (fMRI)
UR - http://www.scopus.com/inward/record.url?scp=84877821093&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84877821093&partnerID=8YFLogxK
U2 - 10.1016/j.neuroimage.2013.04.053
DO - 10.1016/j.neuroimage.2013.04.053
M3 - Article
C2 - 23624491
AN - SCOPUS:84877821093
SN - 1053-8119
VL - 78
SP - 415
EP - 425
JO - NeuroImage
JF - NeuroImage
ER -