Age‐related ventricular expansion is not spatially concordant with MR‐visible periventricular white matter disease

Abstract

AbstractBackgroundVentricular expansion (VE) is recognized as one of the earliest predictors of Alzheimer’s disease (AD); however, VE is also ubiquitous in normal aging. VE is reportedly caused by an ex vacuo mechanism, whereby tissue that appears as periventricular white matter hyperintensity visible on FLAIR MRI (PVH) experiences white matter (WM) atrophy and is subsumed by the lateral ventricles. This longitudinal study sought to test the degree to which the locations into which the ventricles expand were previously PVH as opposed to other tissues.MethodSixty‐five cognitively intact aged subjects each received 2‐6 T1‐weighted and FLAIR MRIs (Table 1). Each volume was rigidly coregistered to the baseline volume and gray matter (GM), lateral ventricular and PVH masks were generated for each timepoint. Ventricular growth masks (VG) were generated for each subsequent timepoint, each VG voxel was classified into either GM, PVH, or normal appearing white matter (NAWM) per overlap with previous timepoint tissue masks (Figure 1A).ResultOverall, average VE between visits was 5.1cm3. Over all subjects and visits, 4.2% of VE occurred where tissue had been previously marked as PVH, 16.1% of VE occurred where tissue had previously been marked as GM, including subcortical GM and the hippocampus, and 75.4% of VE occurred in tissue that had been previously marked as NAWM (Figure 1B).ConclusionThe results support an alternative explanation for VE; that is, that ex vacuo expansion of the lateral ventricles may not be the primary mechanism by which ventriculomegaly occurs in aging. One possible explanation is that unaltered CSF production in the face of dysfunctional CSF egress dynamics causes fluid stasis and reflux which increases pressure in the lateral ventricles.