Independent value added by diffusion MRI for prediction of cognitive function in older adults

Area of Science:

• Neuroimaging
• Cognitive Neuroscience
• Gerontology

Background:

• White matter microstructure changes are implicated in cognitive decline.
• Diffusion magnetic resonance imaging (dMRI) assesses white matter integrity.
• Executive function (EF) and memory (MEM) are critical cognitive domains affected by aging and impairment.

Purpose of the Study:

• To investigate if dMRI-derived white matter microstructure provides independent information on baseline and changes in EF and MEM.
• To assess these associations in older adults with varying cognitive statuses, including normal cognition, mild cognitive impairment (MCI), and dementia.
• To compare the predictive value of white matter integrity against established biomarkers like hippocampal volume and PET metabolism.

Main Methods:

• Longitudinal data from Alzheimer's Disease Neuroimaging Initiative (ADNI) phases GO and 2 (2009-2015).
• Analysis of participants diagnosed with cognitively normal, early MCI, late MCI, and dementia (n=162 total).
• Modeling the association between global white matter mean diffusivity (MD) and EF/MEM composite scores, controlling for hippocampal volume and PET imaging data.

Main Results:

• Greater global MD was linked to lower baseline EF in late MCI and dementia groups.
• White matter integrity (MD) was not significantly associated with changes in EF or MEM.
• Hippocampal volume and FDG PET metabolism were predictive of greater EF and MEM decline, as expected.

Conclusions:

• White matter microstructure assessed by dMRI provides independent information about current executive function in older adults with cognitive impairment.
• However, dMRI-based white matter integrity was not sensitive to future cognitive decline in EF or MEM.
• There is a critical need for novel biomarkers that can predict future executive function decline, given its rapid progression to dementia.