Regional changes in thalamic shape and volume with increasing age

Abstract

The thalamus undergoes significant volume loss and microstructural change with increasing age. Alterations in thalamo-cortical connectivity may contribute to the decline in cognitive ability associated with aging. The aim of this study was to assess changes in thalamic shape and in the volume and diffusivity of thalamic regions parcellated by their connectivity to specific cortical regions in order to test the hypothesis age related thalamic change primarily affects thalamic nuclei connecting to the frontal cortex. Using structural magnetic resonance imaging (MRI) and diffusion tensor imaging (DTI), we assessed thalamic volume and diffusivity in 86 healthy volunteers, median (range) age 44 (20-74) years. Regional thalamic micro and macro structural changes were assessed by segmenting the thalamus based on connectivity to the frontal, parietal, temporal and occipital cortices and determining the volumes and mean diffusivity of the thalamic projections. Linear regression analysis was performed to test the relationship between increasing age and (i) normalised thalamic volume, (ii) whole thalamus diffusion measures, (iii) mean diffusivity (MD) of the thalamo-cortical projections, and (iv) volumes of the thalamo-cortical projections. We also assessed thalamic shape change using vertex analysis. We observed a significant reduction in the volume and a significant increase in MD of the whole thalamus with increasing age. The volume of the thalamo-frontal projections decreased significantly with increasing age, however there was no significant relationship between the volumes of the thalamo-cortical projections to the parietal, temporal, and occipital cortex and age. Thalamic shape analysis showed that the greatest shape change was in the anterior thalamus, incorporating regions containing the anterior nucleus, the ventroanterior nucleus and the dorsomedial nucleus. To explore these results further we studied two additional groups of subjects (a younger and an older aged group, n=20), which showed that the volume of the thalamo-frontal projections was correlated to executive functions scores, as assessed by the Stroop test. These data suggest that atrophy of the frontal thalamo-cortical unit may explain, at least in part, disorders of attention, working memory and executive function associated with increasing age.

Supplementary Fig. 1

Thalamo-cortical tracts in a 55 year old subject. a. Thalamo-frontal tracts shown in the axial plane b. Thalamo-parietal tracts shown in the sagittal plane c. Thalamo-temporal tracts shown in the sagittal plane d. Thalamo-occipital tracts shown in the axial plane.

Fig. 1

Mean thalamo cortical projections in MNI space. Columns a, b, c, and d show axial and coronal views of thalamic regions that were connected to the frontal cortex (red), parietal cortex (blue), temporal cortex (yellow) and occipital cortex (green) respectively.

Fig. 2

Linear regression plots for mean MD (mm²/s) of the left and right thalamo-frontal, thalamo-parietal, thalamo-temporal and thalamo-occipital projections with age. A regression line is shown for those regions where a significant robust linear regression with age was found (Bonferroni correction; p < 0.05/22).

Fig. 3

Linear regression plots for normalised volume (voxels) in the left and right thalamo-frontal, thalamo-parietal, thalamo-temporal and thalamo-occipital projections with age. A regression line is shown for those regions where a significant robust linear regression with age was found (Bonferroni correction; p < 0.05/22).

Fig. 4

Main picture shows the results of the shape analysis for the left and right thalami in MNI space. All regions showed a significant correlational change in shape with age (F = 4.5–12), Blue represents the regions of the thalamus that showed the highest significant correlated changes in shape with age (F = 12). Red regions (F = 2) represent the areas where no significant correlation was found between changes in shape and age. Panels A and B show a magnified view of the shape analysis for the left thalamus, and panels C and D show magnified views for the right thalamus with vectors representing the direction of the shape change (most vector arrow heads cannot be visualised as they point inwards). B. Rotated to view the lateral surface of the left thalamus. D. rotated to view the lateral surface of the right thalamus.