Perivascular and parenchymal fluid characteristics are related to age and cognitive performance across the lifespan

Figures

• 

  Figure 1.. The free water diffusion metric framework used in the present study. (A) The neurite orientation dispersion and density imaging (NODDI) and tissue tensor imaging (TTI) models were run on the diffusion MRI data, yielding the tissue features (ficvf, tFA, tAD, tRD and tMD) and the complementary free water features (fiso, fFA, fAD, fRD and fMD). The free water diffusion features were then averaged within segmented PVS, the PVS-subtracted brain parenchyma, and within 1-mm thick concentric rings surrounding the PVS. (B) The presence of PVS can be observed in an enhanced PVS contrast (EPC) axial slice. These PVS are clearly visible on the same axial slices of fiso, fFA and fRD maps.

• 

  Figure 2.. Free water diffusion measures are differentially associated with age across the lifespan. Free water diffusion measures (fiso, fFA, fAD, fMD and fRD) are correlated with age in the developing (8–21 years) and aging (35–90 years) cohorts separately. Individual data points within the PVS (blue) and brain parenchyma (red) are shown, and the best fit line is provided for each tissue compartment.

• 

  Figure 3.. The relationship between free water diffusion measures with distance from PVS differ with age. Free water diffusion measures were averaged within 1-mm concentric circles emanating from PVS and analyzed in HCP-D (green line) and HCP-A (orange line) separately. Significant differences indicate rings around the PVS where diffusion metrics differ between HCP-A and HCP-D. Significant differences in parenchymal fluid characteristics between developing and aging cohorts were observed across diffusion metrics in regions furthest from the PVS (8–10 mm).

• 

  Figure 4.. The relationship between free water diffusion and cognition differs across ages. The relationship between fFA and the NIH total cognition composite score (Total Cog) is shown for the (A) PVS and (B) parenchyma. A significant interaction between fFA and age is observed in HCP-D, but not HCP-A. (Left) Individual subjects are shown as points color-coded by age, and representative regression lines are shown for the youngest, middle and oldest ages (from light to dark) for each cohort. Visualization of the three-dimensional nature of the continuous-by-continuous interactions using (middle) matrix plots with contour lines and (right) perspective plots. The contour lines of the matric plots describe the joint distribution of fFA and age, where the area of each binned rectangle in the plot is proportional the frequency of the corresponding combination of variables. The curved, non-parallel nature of the contour lines in the interaction plots for HCP-D (Top A, B) indicates an interaction is observed, while the parallel contour lines for HCP-Aging (Bottom A, B) demonstrates a simple additive relationship. The perspective plots provides a 3D surface-based representation where the height of…

• 

  Figure 5.. Free water diffusion measures undergo greater alterations with distance from PVS in subjects with lower cognitive performance. The interaction between NIH Toolbox total cognition composite scores and distance from PVS on parenchymal free water diffusion measures (fFA and fiso) are shown for (A) developing children (8–21 years) and (B) aging adults (35–90 years). The estimated regression lines are derived from linear mixed effects models that account for age and sex in the model. Estimated regression lines are shown for 1 standard deviation below (score of 76) and 1 standard deviation above (score of 136) the mean (score of 100).