Affection of basal ganglia in spinocerebellar ataxia type 3– a transcrainal sonography study

Spinocerebellar ataxia type 3 (SCA3) is the most frequent form of autosomal dominant cerebellar ataxia. Autopsy findings in spinocerebellar ataxia type 3 (SCA3) frequently reveal severe degeneration of the basal ganglia especially the substantia nigra in SCA3. In contrast, clinical signs of basal ganglia involvement are mild and occur only in a minority of SCA3 patients. Transcranial sonography (TCS) is a valuable tool in the detection of basal ganglia alterations in extrapyramidal disorders. To define basal ganglia affection in SCA3 we performed a TCS study in 15 patients with the molecular diagnosis of SCA3 and determined morphological abnormalities in comparison to 15 age- and sex-matched healthy control individuals. Echogenicity of the substantia nigra was measured quantitatively as area of increased echogenicity. Caudate nucleus, pallidum, putamen, dentate nucleus and cerebellar white matter were assessed semiquantitatively (0=none, 1=mild, 2=marked echogenicity) on both sides in comparison to the echotexture of defined brain structures.

Substantia nigra was hyperechogenic in 40% of SCA3 patients. In addition, pallidum (40%) and putamen (40%) but not the caudate exhibited hyperechogenicity. The most frequent alterations in TCS were found in the dentate nucleus (54%) and cerebellar white matter (54%). Hyperechogenicity was marked in at least one region in 53% of SCA3 subjects but in none of the healthy controls. Echogenicity in control individuals was mildly increased in cerebellar white matter (20%), dentate nucleus (9%) and substantia nigra (13%) but not in pallidum or putamen. Hyperechogenicity of the substantia nigra did not correlate with clinical signs of Parkinsonism in SCA3 patients.

We conclude that TCS is suitable to detect basal ganglia abnormalities in SCA3. The most striking finding was hyperechogenicity of pallidum and putamen which was not observed in control individuals. Signal alterations did not correspond well to clinical findings but are confirmed by autopsy studies and therefore most likely reflect distribution and morphology of the neurodegenerative process in vivo.