Extending the ALPS index: Associating whole-brain patterns with glymphatic-related diffusion metrics in multiple...

The glymphatic system (GS) in the perivascular spaces is key in the clearance of metabolic waste products from the brain and is impaired in neuroinflammatory conditions. Unfortunately, its in-vivo non-invasive examination is currently spatially limited to the periventricular spaces. Therefore, we present a new approach that aims to extend the spatial significance of the non-invasive evaluation method.

T1- and diffusion weighted image data of 165 MS patients (30 with clinically isolated syndrome, 17 of which female; and 135 relapsing-remitting MS, 91 female) and 57 healthy age- and sex-matched subjects (HC) were included in this retrospective study. The periventricular DTI-ALPS index was calculated automatically for each individual. Then, based on the voxel-level associations between the DTI-ALPS index with the fractional anisotropy (FA) and apparent diffusion coefficient (ADC) we determined the Glymphatic Association Mapping (GAM). Further associations between GAM and clinical parameters were evaluated.

The distribution of GAM-FA and GAM-ADC voxels was different between groups. In HC, GAM was mostly concentrated in the periventricular space, while in RRMS GAM was significantly more distributed throughout the brain. GAM-FA showed correlations of impaired local glymphatic function and tissue damage in the basal ganglia and thalamus for RRMS patients. RRMS patients demonstrated reduced differences in GS integrity compared to HC. Lower DTI-ALPS index values associated with age, lesion volume and disease duration. Females showed higher values.

The proposed GAM method evidenced widespread associations of GS impairment and tissue damage in MS. The overlap of GAM in basal ganglia and thalamus in MS can serve as mechanistic corroboration of previous pathological findings in these areas.