ausblenden:
Schlagwörter:
-
Zusammenfassung:
Introduction:
Obesity-related structural brain alterations point to a consistent reduction in gray matter (GM) with increasing body mass index (BMI). However, changes in white matter (WM) are more complex and less conclusive. Hence, more recently diffusion tensor imaging (DTI) has been employed to investigate microstructural WM changes. Altogether, these studies have mostly shown a loss of WM integrity with obesity-related factors. However, the complexity of obesity-related factors resulted in interacting influences on the DTI indices. Hence to increase the specificity of the results, we explored specific brain tissue properties, such as myelin, water and iron content by combining measures of GM and DTI with quantitative multi-parameter mapping (MPM).
Methods:
We performed magnetic resonance imaging (3T scanner, Siemens, Tim Trio) using a high-resolution T1-weighted anatomical image (1x1x1 mm3) and diffusion weighted images in 48 lean and obese young adults (BMI range 19.5 to 46.4 kg/m2; age range 21 to 37 years; 23 female). In 33 out of the 48 participants, we additionally applied a whole-brain MPM protocol based on multi-echo 3D FLASH to quantitatively map longitudinal relaxation rate (R1=1/T1), effective transverse relaxation rate (R2*=1/T2*) and magnetization transfer saturation (MT).
Voxel-wise statistical analysis were performed for DTI parameters (fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD); calculated by FSL) and GM and WM volumes acquired from voxel-based morphometry (using VBM8 with default parameters). Quantitative parameter maps from R1, R2*, and MT were obtained using voxel-based quantification (Draganski, et al., 2011). For statistical analysis, we used a multiple linear regression model including four regressors: gender, age, total intracranial volume and BMI as the regressor of interest. All results were thresholded at p < 0.05 family wise error corrected within explicit masks defining GM and WM voxels separately. To compute differences in cortical thickness vertex-by-vertex analyses were performed using Freesurfer.
Results:
Using VBM, we identified significant BMI-related GM reductions in frontal and temporal regions as well as the cerebellum. WM reductions were only found in the right inferior fronto-occipital fasciculus.
The cortical thickness analysis identified significant BMI-related cortical thinning mainly in frontal regions and right cuneus (Figure 1).
Quantitative MPM
Gray matter: BMI was negatively associated with R1 in the left insula and R2* in the bilateral thalamus. A positive correlation was observed between BMI and R2* in the right posterior cingulate gyrus.
White matter: BMI was associated with significant lower R1 in the right anterior thalamic radiation, and within a cluster ranging from the left superior longitudinal fasciculus, superior corona radiata and external capsule (Figure 2). R2* showed a negative correlation with BMI in the bilateral anterior thalamic radiation (Figure 3).
The DTI data revealed that BMI was negatively associated with FA and RD in the right middle cerebellar peduncle and MD and AD in the bilateral corticospinal tract and anterior thalamic radiation (Figure 4). BMI was positively associated with MD and AD in the right superior longitudinal fasciculus.
