Thalamic white matter (WM) injury in multiple sclerosis (MS) remains relatively poorly understood. Combining multiple imaging modalities, sensitive to different tissue properties, may aid in further characterizing thalamic damage. Forty-five MS patients and 17 demographically-matched healthy controls (HC) were scanned with 3T MRI to obtain quantitative measures of diffusivity and magnetic susceptibility. Participants underwent cognitive evaluation with the Brief International Cognitive Assessment for Multiple Sclerosis battery. Tract-based spatial statistics identified thalamic WM. Non-parametric combination (NPC) analysis was used to perform joint inference on fractional anisotropy (FA), mean diffusivity (MD) and magnetic susceptibility measures. The association of surrounding WM lesions and thalamic WM pathology was investigated with lesion probability mapping. Compared to HCs, the greatest extent of thalamic WM damage was reflected by the combination of increased MD and decreased magnetic susceptibility (63.0% of thalamic WM, peak p = .001). Controlling for thalamic volume resulted in decreased FA and magnetic susceptibility (34.1%, peak p = .004) as showing the greatest extent. In MS patients, the most widespread association with information processing speed was found with the combination of MD and magnetic susceptibility (67.6%, peak p = .0005), although this was not evident after controlling for thalamic volume. For memory measures, MD alone yielded the most widespread associations (45.9%, peak p = .012 or 76.7%, peak p = .001), even after considering thalamic volume, albeit with smaller percentages. White matter lesions were related to decreased FA (peak p = .0063) and increased MD (peak p = .007), but not magnetic susceptibility, of thalamic WM. Our study highlights the complex nature of thalamic pathology in MS.
Keywords: MRI; diffusion tensor imaging; multiple sclerosis; quantitative susceptibility mapping; thalamus.
© 2018 Wiley Periodicals, Inc.