Macro- and micro-structural insights into primary dystonia: a UK Biobank study

Abstract

Background: Dystonia is a hyperkinetic movement disorder with key motor network dysfunction implicated in pathophysiology. The UK Biobank encompasses > 500,000 participants, of whom 42,565 underwent brain MRI scanning. This study applied an optimized pre-processing pipeline, aimed at better accounting for artifact and improving data reliability, to assess for grey and white matter structural MRI changes between individuals diagnosed with primary dystonia and an unaffected control cohort.

Methods: Individuals with dystonia (n = 76) were identified from the UK Biobank using published algorithms, alongside an age- and sex-matched unaffected control cohort (n = 311). Grey matter morphometric and diffusion measures were assessed, together with white matter diffusion tensor and diffusion kurtosis metrics using tractography and tractometry. Post-hoc Neurite Orientation and Density Distribution Imaging (NODDI) was also undertaken for tracts in which significant differences were observed.

Results: Grey matter tremor-specific striatal differences were observed, with higher radial kurtosis. Tractography identified no white matter differences, however segmental tractometry identified localised differences, particularly in the superior cerebellar peduncles and anterior thalamic radiations, including higher fractional anisotropy and lower orientation distribution index in dystonia, compared to controls. Additional tremor-specific changes included lower neurite density index in the anterior thalamic radiations.

Conclusions: Analysis of imaging data from one of the largest dystonia cohorts to date demonstrates microstructural differences in cerebellar and thalamic white matter connections, with architectural differences such as less orientation dispersion potentially being a component of the morphological structural changes implicated in dystonia. Distinct tremor-related imaging features are also implicated in both grey and white matter.

Keywords: Diffusion MRI; Dystonia; MRI; Structural MRI.

Fig. 1

Schematic of methodology. a outline of analysis methodology; b grey matter regions of interest; c white matter tracts assessed with tractography; those shown in bold are used in the tractometry analysis; d recruitment flowchart showing final inclusions and exclusions. dMRI diffusion magnetic resonance imaging; DKI diffusion kurtosis imaging; ROI region of interest; FA fractional anisotropy; MD mean diffusivity; MK mean kurtosis; AK axial kurtosis; RK radial kurtosis

Fig. 2

Box plots showing median values for dystonia and controls. a regional volume; b cortical thickness; c Diffusion measures. Significant differences on linear model following multiple comparison correction are highlighted with *

Fig. 3

Along tract profiles (tractometry) for the whole dystonia cohort compared to controls. Green shaded regions represent statistically significant linear model results following multiple comparison correction. Yellow shaded regions are significant (p < 0.050) prior to correction

Fig. 4

Along tract profiles (tractometry) for the cervical dystonia cohort compared to controls. Green shaded regions represent statistically significant linear model results following multiple comparison correction. Yellow shaded regions are significant (p < 0.050) prior to correction

Fig. 5

Along tract profiles (tractometry) for the tremor cohort compared to controls. Green shaded regions represent statistically significant linear model results following multiple comparison correction. Yellow shaded regions are significant (p < 0.050) prior to correction

Fig. 6

Summary of key findings. FA fractional anisotropy; MD mean diffusivity; MK mean kurtosis; AK axial kurtosis; RK radial kurtosis; ODI orientation distribution index; NDI neurite density index