Multimodal MRI Study Of Deep Grey Matter Alterations In Patients With Relapsing-Remitting Multiple Sclerosis Related Fatigue

Objective:Multimodal MRI method were used to assess the deep gray matter alterations in patients with relapsing-remitting multiple sclerosis（RRMS）presenting fatigue and to explore the underlying pathophysiological mechanisms of multiple sclerosis-related fatigue.Methods:Fifty patients diagnosed with clinically definite RRMS were recruited.MS-related fatigue was assessed using the self-reported fatigue severity scale（FSS）.Patients were divided into RRMS presenting fatigue（F-MS,n=25）and RRMS not presenting fatigue（NF-MS,n=25）groups.We also enrolled 40 healthy controls（HCs）age-,sex-,and education-matched to the patients.The Free Surfer toolbox was used to automatically segment deep grey matter and quantify volumes,and the FSL-FIRST toolbox was used to perform Vertex-based morphological analysis on deep gray matter.The thalamic nuclei,which had been divided into 25 subregions in each hemisphere,acted as nodes of the intrinsic thalamic network.Graph theory was used to analyze and quantify intrinsic thalamic network.The accumbens,amygdala,putamen,thalamus,hippocampus,pallidum,caudate nucleus and 15 pairs of thalamic subregions were extracted from Automated Anatomical Labeling version 3 as seed for functional connectivity（FC）analysis.Statistical analyses were performed for different grouping patterns and index characteristics,and the results were corrected for multiple comparisons.Results:1.Volumetric analysis of deep gray matter showed that the F-MS group had significantly lower right thalamic volume than the HCs group（p=0.0027）,and the right thalamic volume of the NF-MS group had a tendency to be lower than that of the HCs group.2.Surface morphological analysis revealed regional morphological contractions on the surfaces of the bilateral thalamus,right caudate nucleus,left accumbens and right hippocampus in patients with RRMS compared to HCs（p<0.05,TFCE FWE corrected）.Compared with the NF-MS group,the F-MS group had morphological contraction on the surface of the left caudate nucleus,which was negatively correlated with the FSS score（rho=-0.485,p=0.002）.3.Volume analysis of thalamic subregions showed that in the F-MS group,the right anteroventral（AV）,bilateral laterodorsal（LD）,right central lateral（CL）,right lateral geniculate（LGN）,right pulvinar anterior（Pu A）,bilateral pulvinar medial（Pu M）,and bilateral pulvinar inferior（Pu I）nuclei volumes were decreased compared to that of the HCs.In the NF-MS group,the bilateral LD,right Pu M,and right Pu I nuclei volumes were decreased compared to that of the HCs（p<0.001）.4.Thalamic structural covariance network analysis revealed that the F-MS group had significantly increased characteristic path length(Lp,p ^AUC=0.038)and decreased nodal efficiency and betweenness centrality of the right MDm nucleus than the NF-MS group（p<0.05,FDR corrected）.5.FC analysis with seven pairs of deep gray matter structures as seeds showed that the FC between the left putamen and the left inferior frontal gurus triangular part was decreased,the FC between the right amygdala and the right insula was enhanced,and the FC between the right hippocampus and the right superior temporal gurus was enhanced in patients with RRMS compared with HCs（voxel level p<0.001,cluster level p<0.05）.6.FC analysis with fifteen pairs of thalamic subregion as seeds showed that compared with the NF-MS group,the FC between the right Thal＿LP and the right inferior frontal gurus triangular part was decreased,the FC between the left Thal＿VL and the right dorsolateral superior frontal gyrus was decreased,the FC between the left Thal＿VPL and right inferior frontal gurus triangular part was decreased in the F-MS group and had a negative correlation with the FSS score.Compared with the NF-MS group,the FC between the right Thal＿Pu I and right dorsolateral superior frontal gyrus was increased and had a positive correlation with the FSS score.Conclusion:In RRMS patients,the thalamus is the most susceptible deep gray matter structure,with damage occurring mainly in the area closest to the ventricles and with right laterality.More severe thalamic volume loss,poorer functional integration of thalamic structural covariance networks,and disrupted thalamic network information flow in the F-MS group suggested that the thalamus may be a potential MRI marker for MS-related fatigue.In addition,FC abnormalities between multiple thalamic subregions and prefrontal cortex,as well as morphological contraction of the left caudate nucleus,were significantly associated with FSS scores,providing evidence to support the hypothesis that both reduced activity and compensatory mechanisms of neural networks in the thalamus,basal ganglia,and frontal cortex are involved in the pathophysiological mechanisms of MS-related fatigue.