A diffusion kurtosis imaging analysis of white matter microstructural differences in bipolar disorder
Goghari, V., De Luca, A., Shakeel, M., Beasley, C., David, S., Leemans, A., Emsell, L.
DOI: https://doi.org/10.1111/bdi.12934
Bipolar Disorders 22 (S1), p. 17-18
Abstract
Background: White matter (WM) microstructural alterations based on diffusion tensor imaging (DTI) metrics have been reported in bipolar disorder (BD). However clinical and methodological heterogeneity confound detection of microstructural alterations in small samples. Here we apply the potentially more sensitive technique of diffusion kurtosis imaging (DKI) to investigate WM differences in BD.Methods: Multi-shell diffusion MRI (b = 300s/mm2 x 8 dir, b = 700s/mm2 x 30 dir, b = 2000s/mm2 x 60 dir) and 3D structural MRI data were acquired in 26 participants with BD and 25 age and gender-matched controls. Group differences in DTI/DKI parameters: mean diffusivity, fractional anisotropy, axial diffusivity, radial diffusivity, mean kurtosis (MK), axial kurtosis (AK), radial kurtosis, and kurtosis anisotropy (KA) were assessed using a WM voxel-based analysis (VBA) and connectivity analysis based on CSD-tractography. Multiple comparisons correction was applied using TFCE and false-discovery-rate respectively.
Results: BD was associated with significantly lower MK than controls in multiple brain regions, including the corona radiata and posterior association bundles. Regional differences in connectivity based on lower mean MK and KA were detected in connections traversing the temporal and occipital lobes, and lower AK was detected in right cerebellar, thalamo-subcortical pathways. There were no statistically significant group differences in DTI metrics.
Conclusion: Bipolar disorder is associated with differences in DKI metrics in multiple brain regions, which may reflect altered connectivity across cerebellar, subcortical, and cortical substructures. DKI may be more sensitive than DTI to detect such differences and hence may represent a useful quantitative biomarker to investigate subtle microstructural differences in BD.