Localized Microstructural Alterations of the Dentatorubrothalamic Tract in Dystonia

1621 

Abstract Submission 

Elizabeth Haddad¹, Xenos Mason¹, Iyad Ba Gari¹, Siddharth Narula¹, Neda Jahanshad¹

¹University of Southern California, Los Angeles, CA

Elizabeth Haddad, MS  
University of Southern California
Los Angeles, CA

Xenos Mason, MD  
University of Southern California
Los Angeles, CA

Iyad Ba Gari  
University of Southern California
Los Angeles, CA

Siddharth Narula  
University of Southern California
Los Angeles, CA

Neda Jahanshad, PhD  
University of Southern California
Los Angeles, CA

Dystonia is a movement disorder characterized by sustained or intermittent muscle contractions which presents as abnormal movements and/or postures. Dystonia may involve widespread physiological, structural, and functional changes in the cerebello-thalamic networks, beyond well-documented neurophysiologic alterations in the basal ganglia [1]. Nonspecific DTI measures, including FA, have been implicated in Dystonia [2], yet NODDI may better inform specific microstructural tissue changes [3]. Tractography methods have been developed to characterize along-tract changes; these could help guide the targeting of surgical interventions including deep brain stimulation (DBS). Here, we used the UK Biobank [4] to interrogate location-specific white-matter properties along the dentatorubrothalamic tract (DRTT), in individuals with dystonia compared to controls, using advanced dMRI methods.

19 subjects with dystonia (63.8±7.5 years old; 14F) and 65 controls (64.3±7.3; 46F) matched on age, sex, medical comorbidities and scan site were included (Figure 1a). Raw dMRI were downloaded, corrected for noise, Gibbs ringing, EPI distortion, eddy currents, and bias field inhomogeneity. The following NODDI maps were generated: orientation dispersion (OD), extracellular (ECVF), intracellular (ICVF), and isotropic (ISOVF) volume fraction. ISOVF maps were thresholded with an upper limit of 0.5 to ensure CSF cancellation. Coregistration was performed between subject T1w MRI and dMRI, as well as between subject T1w MRI and the MNI152 template in order to propagate labels to diffusion space. We used a multi-shell multi-tissue constrained spherical deconvolution reconstruction for probabilistic tractography in MRTrix [5]. The DRTT was generated by reconstructing the cerebello-thalamic (CTT) and the thalamo-cortical (TCT) components separately. First, the dentate, fastigial, and interposed nuclei were combined and dilated and then used as a seed ROI to track to either the ipsilateral or contralateral thalamus. Second, seed termination points of both the ipsilateral and contralateral segments were combined and used as a probabilistic seed map to track motor cortex targets. The resulting TCT tract was parcelleted into either the premotor, primary, or supplementary motor cortices. See details in Figure 1b. NODDI metrics were sampled across all tracts. Robust means were calculated for all bilaterally averaged tracts. Along tract analysis was performed using Automated Fiber Quantification [6], where each individual tract component was divided into 10 segments. Linear regression models were run with dystonia, age, and sex as independent variables on all bilaterally averaged measures. The bilateral averages with the strongest associations then underwent along tract analysis on lateralized tracts to localize differences. Multiple comparisons correction was performed on (1) all bilateral averages across all measures and (2) all segments across all tracts which were significant in 1.

For all bilateral tracts assessed, we find lower ISOVF to be significantly associated with dystonia. The strongest component was found in the ipsilateral CTT component (standardized β=-0.67; p=0.008; q=0.14) although none of these associations survived multiple testing correction (Figure 2a). Several segments along this tract and others were significantly associated with dystonia, although these did not survive multiple testing correction (Figure 2b).

We find trends of lower ISOVF along the DRTT in individuals with dystonia. These findings support the hypothesis that changes in cerebellar networks contribute to dystonia, a rare condition with otherwise limited characterization of white matter microstructure. Neuroimaging can be leveraged to inform underlying microstructural tissue properties contributing to disease pathogenesis and possibly to guide treatment. Larger studies are needed for better powered analyses to capture such effects.

Deep Brain Stimulation ²

Diffusion MRI Modeling and Analysis ¹

Basal Ganglia

Cerebellum

DISORDERS

Motor

Movement Disorder

Tractography

White Matter