The Association of Aging with White Matter Integrity and Functional Connectivity Hubs

Abstract

Normal aging is associated with reduced cerebral structural integrity and altered functional brain activity, yet the association of aging with the relationship between structural and functional brain changes remains unclear. Using combined diffusion tensor imaging (DTI) and functional magnetic resonance imaging (fMRI) modalities, we hypothesized that aging-related changes in white matter integrity (i.e., fractional anisotropy) was associated with the short- or long-range functional connectivity density (FCD) in hub regions. We tested this hypothesis by using a healthy aging cohort comprised of 140 younger adults aged 20-39 years and 109 older adults aged 60-79 years. Compared with the younger group, older adults exhibited widespread reductions in white matter integrity with selective preservation in brain stem tracts and the cingulum connected to the hippocampus and cingulate cortex, whereas FCD mapping in older adults showed a reduced FCD in the visual, somatosensory, and motor functional networks and an increased FCD in the default mode network. The older adults exhibited significantly increased short- or long-range FCD in functional hubs of the precuneus, posterior, and middle cingulate, and thalamus, hippocampus, fusiform, and inferior temporal cortex. Furthermore, DTI-fMRI relationship were predominantly identified in older adults in whom short- and long-range FCD in the left precuneus was negatively correlated to structural integrity of adjacent and nonadjacent white matter tracts, respectively. We also found that long-range FCD in the left precuneus was positively correlated to cognitive function. These results support the compensatory hypothesis of neurocognitive aging theory and reveal the DTI-fMRI relationship associated with normal aging.

Keywords: aging; compensatory hypothesis; diffusion tensor imaging; functional connectivity density mapping.

Figure 1

Analysis flow of resting-state blood oxygen level dependent (BOLD) signals and diffusion tensor images (DTI). Both data were preprocessed according to the standard procedure. Short- and long-range functional connectivity density (FCD) was calculated from resting-state BOLD signals, and fractional anisotropy (FA) was determined from DTI data. We analyzed the between age-group difference in FCD and FA, using ROI analysis over 42 functional hubs and 48 white matter tracts. Furthermore, we determined the correlation between FA of 48 white matter tracts and FCD value at 42 functional hubs in younger and older adults.

Figure 2

Regional differences between age-groups in short- and long-range FCD, as well as white matter integrity based on fractional anisotropy value.

Figure 3

Patterns of correlation between white matter integrity (green nodes; labeled with W) and FCD in hub regions (red nodes; labeled with G). The number shown in the node labels corresponds to the brain region order listed in Table 2 (G) and Table 3 (W). For example, G2, left precuneus; G11, right angular, G15: right inferior frontal. W27, right posterior corona radiata; W30, left posterior thalamic radiation.

Figure 4

Correlation graph between white matter integrity and long-range FCD in left precuneus in older adults.