White Matter Tracts in Bipolar Disorder Patients: A Comparative Study Based on Diffusion Kurtosis and Tensor Imaging.

Jian Yang,Xiaowei Jiang,Shengnan Wei,Xin Deng,Yue Zhu,Miao Chang,Zhiyang Yin,Haiyang Geng,Yanqing Tang,Xu Dai
DOI: https://doi.org/10.1016/j.jad.2021.05.030
IF: 6.533
2021-01-01
Journal of Affective Disorders
Abstract:BACKGROUND:Diffusion kurtosis imaging (DKI), an extension of diffusion tensor imaging (DTI), is a powerful tool for studying human brain.The purpose is to investigate differences between DKI and DTI by comparing parameters in same analysis methods with bipolar disorder (BD) patients.METHODS:In this study, we attained in 47 BD patients and 49 age-, sex-, and education-matched healthy controls, complimented DTI and DKI scanning and got Fractional Anisotropy (FA), Mean Diffusion (MD) and Mean Kurtosis (MK). Voxel-wise statistical analysis was performed by the tract-based spatial statistics (TBSS) analysis and atlas-based regional data analysis.RESULTS:TBSS analysis showed more widespread regions and higher fidelity in DKI parameters than DTI parameters with the same p-value threshold, and DKI parameters showed significant alterations after Family-Wise Error correction. The DKI-FA value in the corpus callosum, bilateral cingulum (cingulate gyrus), bilateral superior corona radiata, left anterior corona radiata and left posterior corona radiata of BD patients was negatively correlated with the duration of illness. In the atlas-based regional data analysis, the effect size of DTI-FA, DTI-MD, DKI-FA and DKI-MD were quantified using Cohen's d value. DKI-FA and DKI-MD demonstrated more between-group different regions and the higher (p < 0.001) absolute Cohen's d value than DTI-FA.LIMITATIONS:This study did not consider the difference between sub-types of BD.CONCLUSIONS:Compared to DTI parameters, DKI parameters were more sensitive and stable to probe the local microstructure, and particularly powerful to exploit cerebral alterations in BD patients.
What problem does this paper attempt to address?