Multimodal MRI Longitudinal Assessment of White and Gray Matter in Different SPG Types of Hereditary Spastic...

Hereditary spastic paraplegias (HSP) are a group of genetically and clinically heterogeneous neurologic disorders. Hereby we describe a relatively large group of patients (pts) affected by HSP studied at baseline (31 pts) and at follow-up (mean period 28.9 ± 8.4 months; 23 pts) with multimodal advanced MRI: high-resolution T1 images for voxel-based morphometry (VBM) analysis, magnetic resonance spectroscopy (MRS), and diffusion tensor imaging (DTI). An age-matched healthy control (HC) group underwent the same neuroimaging protocol in a time schedule matched with the HSP patients. At baseline, VBM showed gray matter (GM) reduction in HSP in the right pre-frontal cortex and bilaterally in the thalami. MRS at baseline depicted in HSP patients compared to the HC group reduction of NAA/Cr ratio in the right pre-frontal region, increase of Cho/Cr ratio in the right pre-central regions, and increase of mI/Cr ratio on the left pre-central area. At cross-sectional follow-up analysis and longitudinal evaluation, no VBM and MRS statistically significant results were obtained. Tract-based spatial statistics (TBSS) analysis showed widespread DTI brain white matter (WM) alterations in patients compared to HC at baseline, which are characterized by reduction of fractional anisotropy (FA) and increase of mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity, as confirmed on cross-analysis of the follow-up dataset. A longitudinal analysis with TBSS in HSP patients did not show significant variations, while upon applying region-based analysis we found increased FA and decreased MD and AD in specific brain WM fiber complex during follow-up. The changes were not correlated with the clinical presentation (pure vs complicated HSP), motor function, and motility indexes or history of specific treatments (botulinum toxin). In conclusion, the cross-sectional analysis of the multiparametric MRI data in our HSP patients confirmed the non-prominent involvement of the cortex in the primary motor regions but rather of other more associative areas. On the contrary, DTI demonstrated a widespread involvement of the brain WM, including the primary motor regions, which was confirmed at follow-up. The longitudinal analysis revealed an apparent inversion of tendency when considering the expected evolution of a neurodegenerative process: we detected an increase of FA and a decrease of MD and AD. These time-related modifications may suggest a repair attempt by the residual central WM fibers, which requires confirmation with a larger group of patients and with a longer time interval.