BMI & Waist-to-Hip Ratio: Gray Matter Associations in 22,075 Healthy Individuals from the UK Biobank

2312 

Abstract Submission 

Emma Gleave¹, Paul Thompson², Priya Rajagopalan¹

¹University of Southern California, Los Angeles, CA, ²Imaging Genetics Center, Keck School of Medicine of University of Southern California, Los Angeles, CA

Emma Gleave  
University of Southern California
Los Angeles, CA

Paul Thompson, PhD  
Imaging Genetics Center, Keck School of Medicine of University of Southern California
Los Angeles, CA

Priya Rajagopalan, MBBS, MPH  
University of Southern California
Los Angeles, CA

Higher body mass index (BMI) and waist-to-hip ratio (WHR) are markers of obesity, and represent important cardiovascular-metabolic risk factors. Prior work has reported associations between lower brain volume and obesity markers such as BMI (Raji et al., 2010) and blood plasma leptin levels (Rajagopalan et al., 2013). Specifically, higher BMI and WHR have been linked to brain structural alterations including lower gray matter volume in numerous subcortical nuclei (caudate, putamen, pallidum, and nucleus accumbens;; Hamer & Batty, 2019; Dekkers et al., 2019; Pflanz et al., 2022). Here, we carried out whole-brain analyses using voxel-based morphometry (VBM) to map brain signatures of BMI and WHR in a large population-based sample of healthy adults.

The Computational Anatomy Toolbox (CAT12; Gaser et al., 2023) was used to perform a large-scale voxelwise segmentation of 3D T1-weighted brain MRI data from 27,075 participants (52.5% female; 63.6 (7.5SD) years old) from the UK Biobank (UKBB; Table 1). All participants were assessed for BMI, body size measurements (to calculate WHR), systolic & diastolic blood pressure (to calculate mean arterial pressure), and pulse wave arterial stiffness, at the time of scan. Subjects without these parameters, or with neurological or neuropsychiatric diagnoses, were excluded. Within this subset of healthy UKBB participants, effects of BMI on gray matter volume were tested using a whole-brain voxelwise VBM analysis adjusting for mean arterial blood pressure (MAP), pulse wave arterial stiffness (PWAS), age, sex, intracranial volume, genetic ancestry and education.

Significant gray matter volume alterations were mapped throughout the brain after correcting for multiple comparisons by controlling the false discovery rate at 5%. Voxel-wise brain analyses demonstrated patterns of significantly lower gray matter volumes (GMV) associated with BMI (standard-FDR critical P-value=0.02; q=0.05), after covarying for age, sex, education, ancestry, mean arterial pressure, pulse wave arterial stiffness and intracranial volumes (Figure 1). Specifically, extensive areas of lower gray matter volume were noted predominantly in the bilateral basal ganglia (caudate, putamen, globus pallidum, nucleus accumbens), insula, opercular cortex, diffuse temporal cortex, as well as cerebellum. Similar association patterns were noted for the voxel-wise brain analyses of GMV and waist-to-hip ratio (standard-FDR critical P-value=0.02; q=0.05). Pearson correlations were as follows: BMI:WHR (r=.42), BMI:PWAS (r=.12), BMI:MAP (r=.27), WHR:PWAS (r=.21), WHR:MAP (r=.26), PWAS:MAP (r=.13).

Building on prior findings in a large population sample, we found that markers for both overall obesity and central obesity are both associated with neurostructural alterations in gray matter architecture in healthy adults, even after adjusting for effects of blood pressure and age. Higher BMI and WHR were associated with significantly lower gray matter volume in brain regions involved in sensorimotor processing, motor coordination, and balance.

Cortical Anatomy and Brain Mapping

Anatomical MRI ¹

Imaging Methods Other

Physiology, Metabolism and Neurotransmission Other ²

ADULTS

Basal Ganglia

MRI

Segmentation

STRUCTURAL MRI

Other - Obesity