Probing tissue microstructural changes in neurodegenerative processes using non-gaussian diffusion MR imaging

Abstract

Development of non-invasive imaging biomarkers sensitive to microstructural organization is crucial for deepening our understanding of mechanisms underlying neurodegenerative processes such as aging and further improving early diagnosis and monitoring of neurodegenerative disease such as Alzheimer’s disease (AD) and amnestic mild cognitive impairment (MCI). The diffusional kurtosis imaging (DKI) is an extension of conventional diffusion tensor imaging. It is hypothesized that DKI will provide complementary information to conventional diffusivity metrics in a new dimension that will more comprehensively capture microstructural changes in anisotropic white matter tracts and particularly in relatively isotropic tissues such as gray matter during neurodegenerative processing of aging, MCI and AD and probably improve the early diagnosis of the diseases.

Firstly, DKI method and a white-matter model that provided metrics of explicit neurobiological interpretations were applied on healthy participants. In white matter tracts, age-related degenerations appeared to be broadly driven by axonal loss. Demyelination may also be a major driving mechanism, although confined to the anterior brain. In terms of deep gray matter, higher mean kurtosis (MK) and fractional anisotropy (FA) in the globus pallidus, substantia nigra, and red nucleus reflected higher microstructural complexity and directionality compared with the putamen, caudate nucleus, and thalamus. In particular, unique age-related positive correlations for FA, MK, and radial kurtosis (KR) in the putamen opposite to those in other regions were observed.

Secondly, to verify the speculation that iron deposition could be one probable underlying mechanism driving changes in microstructure, another advance MRI technique of quantitative susceptibility mapping (QSM) was also used in healthy participants. Significant age-related increases of iron were observed in the putamen, red nucleus, substantia nigra, and caudate nucleus. Putamen exhibited the highest rate of iron accumulation with aging, which was nearly twice of the rates in substantia nigra and caudate nucleus. Significant positive correlations between susceptibility value and diffusion measurements were observed for FA and MK in the putamen as well as FA in the red nucleus.

Thirdly, whether DKI metrics could serve as imaging biomarkers to indicate the severity of cognitive deficiency for AD and MCI was investigated. In AD, significantly increased diffusivity and decreased kurtosis parameters were observed in both white and gray matter of the parietal and occipital lobes as compared to MCI. Significantly decreased FA was also observed in the white matter of these lobes in AD. With the exception of FA and KR, all the other five DKI metrics exhibited significant correlations with mini-mental state examination score in both white and gray matter.

Lastly, DKI metrics were compared against volumetry for diagnosis of AD and MCI. In AD vs. aMCI, although no significant difference of either FA or MD was observed in white matter tracts, it is encouraging to note that MK captured loss of microstructural complexity in the superior longitudinal fasciculus and internal capsule. MK in the putamen showed the highest power that outperformed volume of the hippocampus for discriminating AD from normal. Besides, FA in the putamen showed the second highest power for discriminating aMCI from normal.