Normal aging is associated with changes in both micro-structure of white matter in large due to loss of mayelin and in cerebral vasculature due to development of small aneurysms and weakening of the arterial walls. It has been hypothesized that both processes are related in that age-related microvascular changes contribute to both impaired integrity of neuronal fiber tracts and increased vascular pathology. We sought to test this hypothesis by examining the association between fractional anisotropy (FA, a diffusion tensor imaging (DTI) measure of structural integrity) and pulsitility index (PI, a transcranial Doppler (TCD) ultrasound measure of abnormal arterial flow). White matter hyperintensities had been previously attributed to cerebral microvascular disease and linked to increase in pulsatile flow (1).

This study was approved by Albert Einstein College of Medicine IRB; 144 study participants were drawn from a pool of non-demented adults over the age of 70 years from the Einstein Aging Study (EAS). Eligible participants were non-institutionalized at least 70 years old and free of stroke, dementia and disability at enrollment. Details of the EAS study design, subject enrolment and methods are described in (2).

TCD was performed on a Pioneer TC8080$$$^{\mbox{TM}}$$$ system (Viasys Healthcare Inc.) using a 2MHz transducer through temporal "acoustic window" in the total of 9 (bilateral) cerebral arteries: vertebral arteries (VA), basilar artery (BA), middle cerebral arteries (MCA), posterior cerebral arteries (PCA) and the anterior cerebral arteries (ACA). Peak systolic velocity (PSV), end-diastolic velocity (EDV) and mean flow velocity (MFV) were analyzed off-line using the imaging reading work station (ImagePro, Inc.). The PI index was calculated as $$$(PSV − EDV)/MFV$$$.

MRI was performed using a 3.0T Philips Achieva TX scanner (Philips Medical Systems, Best, The Netherlands) utilizing its 32-channel head coil and the following protocol: T1-weighted 3D MPRAGE with TR/TE/TI=9.9/4.6/900msec, flip angle $$$8^{\circ}$$$, 1mm$$$^{3}$$$ isotropic resolution, 240x188x220 matrix; DTI using 2D single-shot EPI with 32 diffusion encoding directions, b-value=800s/mm$$$^{2}$$$, TR=10.0sec, TE=65msec, 2mm$$$^{3}$$$ isotropic resolution, 128x120 matrix, 70 slices; and an auxiliary 3D $$$B_{0}$$$ field map using a dual echo gradient echo technique with TR/TE/$$$\Delta$$$TE=20/2.4/2.3ms; 4mm$$$^{3}$$$ isotropic resolution, flip angle, $$$20^{\circ}$$$ to correct EPI distortions in DTI and small distortions in T1-weighted images.

After correction for head motion and eddy current effects, FA was derived from the tensor model fit using the FSL (3). FA images were co-registered to the study-specific brain template using a multi-step procedure that incorporates EPI distortion correction and both linear within-subject and nonlinear to template transformations, as described previously (4).

Clusters of voxels with significant correlations of FA and TCD measures were identified using voxel-wise $$$t$$$-test regressing out age, gender and years of education. Correlations were considered significant at the voxel-level threshold of $$$p < 0.005$$$. To control overall type-I error (false positives) rate to below 0.01 we only retained contiguous clusters of such voxels at least 100mm$$$^{3}$$$ in size. These criteria have been previously reported (4). Analysis was restricted to voxels within the white matter as delineated by Free Surfer (5) over T1-weighted template brain.

Demographics of the sample: The mean age of the subjects was 78; 54% were women; average education was 14 years. Clusters of voxels with significant correlations of FA and age are shown in Figure 1. A total of 23 clusters are detected, of which 19 where correlation is negative (older age is associated with lower FA) and 4 where correlation is positive. The total volumes of these sets are 11665mm$$$^3$$$ and 1520mm$$$^3$$$ respectively. Clusters with statistically significant correlations of FA and PI in the left MCA are presented in Figure 2 with the summary for all vessels given in Table 1. Both positive and negative correlations are observed with negative correlations prevailing. Intersection between clusters where FA is significantly related to age and clusters where FA is significantly related to PI is small.To our knowledge this is the first report to show a link between a TCD measure of cerebral arterial pathology and DTI based white matter integrity. This supports the hypothesis that microvascular disease contributes to disruption of white matter integrity in aging.