Cerebral blood volume in Alzheimer's disease and correlation with tissue structural integrity

Abstract

A vascular component is increasingly recognized as important in Alzheimer's disease (AD). We measured cerebral blood volume (CBV) in patients with probable AD or Mild Cognitive Impairment (MCI) and in elderly non-demented subjects using a recently developed Vascular-Space-Occupancy (VASO) MRI technique. While both gray and white matters were examined, significant CBV deficit regions were primarily located in white matter, specifically in frontal and parietal lobes, in which CBV was reduced by 20% in the AD/MCI group. The regions with CBV deficit also showed reduced tissue structural integrity as indicated by increased apparent diffusion coefficients, whereas in regions without CBV deficits no such correlation was found. Subjects with lower CBV tended to have more white matter lesions in FLAIR MRI images and showed slower psychomotor speed. These data suggest that the vascular contribution in AD is primarily localized to frontal/parietal white matter and is associated with brain tissue integrity.

Fig. 1

Procedure for the calculation of normalized CBV. The signal differences between pre- and post-contrast VASO images were first calculated, which was realigned to the T1-weighted anatomic image. The image of each individual was then registered to the brain template using an elastic transformation. The CBV value of each voxel was normalized to the cerebellum CBV to reduce the effect of normal variations in global CBV.

Fig. 2

Group averaged CBV maps in AD/MCI and control subjects. (a) Relative CBV maps in AD/MCI patients. (b) Relative CBV maps in controls. The scale of the color bar is from 0 to 2.5 times the cerebellum CBV. (c) An anatomical image corresponding to the slices presented. (d) The locations of coronal slices are marked in a sagittal anatomical image. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

Fig. 3

Results of voxel-based analysis with two sample t-test on the CBV maps of AD/MCI and control groups. The colored regions show significant CBV deficit in AD/MCI patients (p < 0.005 and minimum cluster size = 1250 mm³). The color bar indicates Student's t-statistic value. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

Fig. 4

ROI analysis results of CBV in AD/MCI and control groups. 8-mm (frontal, parietal, and occipital WM, hippocampus) or 6-mm (parietotemporal GM) diameter circles were drawn on each subject's normalized CBV maps to obtain the mean value for each ROI. The asterisks indicate regions with significant differences (two sample Student's t-test, p < 0.05).

Fig. 5

Correlation between CBV and ADC values across subjects (including both AD/MCI patients and controls, n = 25). The values of CBV and ADC for each subject were calculated from the regions showing significant CBV deficits (see Fig. 3).

Fig. 6

Correlation between the normalized CBV and the volume of white matter lesion measured by FLAIR images (including both AD/MCI patients and controls, n = 24). The value of CBV for each subject was calculated from the regions showing significant CBV deficits, and the volume of white matter lesion was from the entire brain.