1. Department of Radiology, University of Manitoba, Winnipeg, MB, Canada; 2. Department of Psychological & Brain Sciences, Johns Hopkins University, Baltimore, MD, USA
While it is well established that obesity decreases overall life expectancy and increases the risk of several adverse health conditions, the effects of excess body fat on brain development and neural function are relatively poorly understood. Previous investigations suggest that individual differences in body composition are related to structural and functional brain variations, but the findings have been inconclusive; and since no single study has ever examined both structural and functional neural correlates of obesity, no direct links between the two have been established. Therefore, the purpose of the current study was to expand on this work by assessing how two different measures of body composition – body mass index (BMI) and body fat percentage (BFP) – relate to individual differences in: 1) regional gray and white matter volumes, 2) white matter microstructure, and 3) intrinsic functional connectivity.
For each of the 32 healthy volunteers enrolled in the study (balanced for gender), BFP was measured using a bioelectric impedance scale, and BMI was calculated based on height and weight. All MRI data were collected on a 3T Philips Achieva system, equipped with a 32-channel head coil. Voxel-based morphometry (VBM) was performed on high-resolution T1-weighted images to assess volumetric changes in gray and white matter that were correlated with either BMI or BFP (corrected for age and gender). All diffusion tensor imaging (DTI) data were normalized using a high-dimensional, non-linear warping algorithm, before performing voxel-wise regression to identify regions where fractional anisotropy (FA) and mean diffusivity (MD) – putative markers of microstructural integrity – were correlated with either BMI or BFP (corrected for age and gender). Finally, resting state fMRI was also performed to assess intrinsic functional connectivity within three well-known brain networks (i.e., the default mode network, executive control network, and salience network) in odder to determine how strongly within-network functional connectivity strength was associated with either BMI or BFP (corrected for age and gender).
VBM analyses revealed that gray matter volumes were negatively correlated with BMI and BFP throughout the bilateral basal ganglia and occipital lobe, while white matter volumes were decreased in the basal ganglia, corpus callosum (CC), inferior longitudinal fasciculus (ILF), superior longitudinal fasciculus (SLF), and regions of the fronto-occipital fasciculus (FOF). Neither gray matter nor white matter analyses revealed obesity-related volumetric increases. DTI analyses revealed increased FA (i.e., greater microstructural integrity) with higher BMI and BFP in the left cerebellum, and several significant clusters of decreased FA (i.e., lower microstructural integrity) were found in and around: the basal ganglia, thalamic radiation, genu of the corpus callosum, orbitofrontal white matter, temporal white matter, and regions of the ILF. Several regions also exhibited positive correlations between either BMI or BFP and MD, indicating lower microstructural integrity throughout regions of the olfactory bulb, orbitofrontal cortex; the splenium of the corpus callosum; bilateral cuneus/precuneus; regions of the cingulum proximal to the hippocampus, as well as the parahippocampal gyri. Finally, analyses of resting state fMRI data revealed that there was no significant relationship between default mode network connectivity and either BMI or BFP (p = 0.35 and p = 0.32); no significant relationship between executive control network connectivity and BFP, but a marginally significant trend toward increased connectivity and higher BMI (p = 0.54 and p = 0.09); and significantly increased salience network connectivity among individuals with higher BMI and BFP (p = 0.04 and p = 0.06).
Therefore, in addition to other well-known health effects, our results indicate that increased BMI and BFP are associated with widespread decreases in cerebral gray and white matter volumes, as well as reduced white matter microstructure and altered functional connectivity within a cortical network involved in salience processing, the coordination of behavioral responses, and the initiation of cognitive control. These results are consistent with a wide body of literature reporting obesity-related cognitive decline, and provide a clear neurophysiological explanation for this phenomenon.