Cortical and subcortical structural brain alterations in restless legs syndrome: A whole brain-based MRI approach to in vivo pathoanatomy

Introduction: Various neuroimaging and electro-physiological techniques have demonstrated partly conflicting results of cortical, subcortical, brainstem and spinal alterations in restless legs syndrome (RLS). Thus, up to date, a circumscribed pathoanatomical correlate of this disorder remains still undetermined. By application of the appropriate MRI technique each, i.e. voxel-based morphometry (VBM) for gray matter (GM) and diffusion tensor imaging (DTI) for white matter (WM), the aim of the present study was to address potential alterations in brain anatomy within the same group of patients suffering from idiopathic RLS.

Methods: A large patient sample from the outpatient clinic for movement disorders of the University of Ulm underwent MRI scans including the acquisition of a high-resolution T1w MP-RAGE and a DTI EPI-sequence. Post-processing on 3D data of RLS patients (n=63) was performed using optimized VBM in comparison to an age-matched normal data base (n=40). The post-processing steps of the DTI data sets were based on the compilation of a study-specific template of all single data sets and a whole-brain based statistical analysis of the fractional anisotropy (FA) of RLS patients (n=45) in comparison to a control sample (n=31).

Results: The VBM analysis demonstrated areas of regional decrease in gray matter volume (at p<0.05, svc). The largest voxel clusters were localized in the bilateral somatosensory cortex of the right and left hemisphere almost symmetrically and in the left-hemispheric motor cortex. Large bihemispherically localized voxel clusters of a significant FA reduction at p<0.05 were observed by application of standardized DTI analysis methods. The pattern of the subcortical FA alterations ranged from prefrontal over premotor to motor areas, the latter in close proximity to the cortical areas revealed by VBM.

Conclusion: By use of advanced whole brain-based MRI techniques on group data of patients with idiopathic RLS, the involvement of both cerebral gray matter structures and subcortical white matter areas as structural pathoanatomical correlates in RLS could be identified. With respect to these widespread alterations, the pathoanatomical correlate of RLS symptoms has to be considered as a dysfunction of neuronal networks with a focus on the sensorimotor system, reflecting a new hodological approach to the disease. These results of primary structural brain changes assessable by advanced MRI are in general agreement with a recent genome-wide association study by Winkelmann et al., which pointed at components of a developmental disorder in the pathogenesis of RLS.