White matter myelin profiles linked to clinical subtypes of Parkinson's disease

Background White matter (WM) microstructural integrity is important for effective brain functioning and alterations have been shown in many neurodegenerative diseases. Purpose To investigate WM myelin profiles and their relation to clinical features of Parkinson's disease (PD). Study Type Retrospective cross‐sectional. Population In all, 29 PD subjects and 15 healthy controls. Field Strength/Sequence Multiecho GRASE with 10 msec echo spacing and echo planar imaging (EPI) diffusion‐weighted (b‐value = 700 with 32 gradient directions) on a 3T scanner. Assessment Myelin water fraction (MWF) and fractional anisotropy (FA) across 20 WM regions of interest (ROIs) were compared between groups. Partial least squares (PLS) was used to associate MWF and FA with clinical and behavioral measures. Statistical Tests Group comparisons were done using two‐sample t ‐tests. PLS was assessed with permutation tests. Bootstrapping was used to investigate the robustness of imaging features. Results No group differences in myelin content could be detected with univariate tests. A three‐component PLS model linked MWF profiles to clinical phenotypes but no FA profiles. The three components appeared to follow along broad motor/nonmotor subtypes of “akinetic‐rigid,” “tremor‐predominant,” and “depression/apathy” subtypes, respectively. The first component showed associations between overall motor scores ( r = –0.43, P = 0.0196) and cognitive performance ( r = 0.44, P = 0.0171) with interhemispheric and long‐range association fibers. A second component linked overall motor scores ( r = –0.58, P = 0.0009) and tremor scores ( r = –0.48, P = 0.0091) to predominantly projection fibers. The last component related depression ( r = –0.60, P = 0.0006) and apathy scores ( r = –0.66, P = 0.0001 and r = –49, P = 0.0072) to a mixture of association and projection fibers. Data Conclusion MWF was robustly linked to distinct clinical subtypes of PD and may serve as an additional tool to characterize the disease. Level of Evidence: 4 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2019;50:164–174.