Objective: We examined the usefulness of voxel-based morphometry (VBM) analysis in auxiliary diagnosis of cases with spinocerebellar degeneration, including those with MSA-C.

Background: In Japan, one-third of SCD cases are hereditary; two-thirds of sporadic cerebellar ataxia cases are MSA-C. MRI is not suitable for diagnosing hereditary diseases.

Method: This study included cerebellar ataxia patients who visited either Tokyo Medical University Hachioji Medical Center between 2007 and 2014 or Tokyo Medical University Hospital between 2015 and 2017. They were diagnosed with hereditary spinocerebellar degeneration (SCA6, SCA31, DRPLA, MJD/SCA3) or MSA-C based on the second consensus criterion. We performed brain MRI to obtain 3D-T1WI. VBM analysis was conducted by VSRAD advance (SPM8+DARTEL), and voxel-based stereotactic extraction estimation (vbSEE). We calculated the percentage of abnormal voxels (extent) in the region of interest (cerebellar/brainstem) at the hemisphere level (level 1) in both gray and white matter using vbSEE. Patients with hereditary spinocerebellar degeneration were classified into 2 groups: pure cerebellar type (PC type: SCA6 and SCA31) and non-pure cerebellar type (non-PC type: MJD/SCA3 and DRPLA). We compared the extent (%) of the three cerebellar degeneration types (PC type, non-PC type, MSA-C) in the region of interest.

Results: Six patients had DRPLA; 11, MJD/SCA3; 12, SCA6; 5, SCA31; and 34, MSA-C. No difference was present with respect to sex or gait in Scale for Assessment and Rating of Ataxia scores for each disease. MSA-C had a shorter disease duration than the other disease types. PC type showed lower value in the brainstem white matter extent (BSWEx) than did the other types (cutoff value, 21.6). Non-PC type showed lower value in the cerebellar gray matter extent (CGEx) than did the other types (cutoff value, 26.8). We calculated the brainstem white matter to cerebellar gray matter (BSWCG) ratio by dividing BSWEx by CGEx. This ratio was high in non-PC type patients, low in PC type patients, and intermediate in MSA-C patients; this ratio showed high accuracy (95.6%) for disease classification.

Conclusion: Results of the VBM analysis in the three cerebellar ataxias were consistent with pathological changes. In the classification of cerebellar ataxia, the BSWCG ratio showed high accuracy and can be used as a powerful diagnostic imaging tool.

« Back to 2019 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/classification-of-spinocerebellar-degeneration-based-on-ratio-of-brainstem-white-matter-to-cerebellar-gray-matter-by-voxel-based-morphometry/