Associations between Obesity Trajectories, Brain health, and Cognitive Function in Aging Population

2092 

Abstract Submission 

Die Zhang¹, Chenye Shen², Yingji Fu¹, Anqi Qiu^1,2,3

¹Department of Health Technology and Informatics, Hong Kong Polytechnic University, Kowloon, Hong Kong, ²Department of Biomedical Engineering, National University of Singapore, Singapore, Singapore, ³Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD

Die Zhang  
Department of Health Technology and Informatics, Hong Kong Polytechnic University
Kowloon, Hong Kong

Chenye Shen  
Department of Biomedical Engineering, National University of Singapore
Singapore, Singapore

Yingji Fu  
Department of Health Technology and Informatics, Hong Kong Polytechnic University
Kowloon, Hong Kong

Anqi Qiu  
Department of Health Technology and Informatics, Hong Kong Polytechnic University|Department of Biomedical Engineering, National University of Singapore|Department of Biomedical Engineering, Johns Hopkins University
Kowloon, Hong Kong|Singapore, Singapore|Baltimore, MD

Obesity in aging population is a public health challenge [1]. Recent studies have increasingly focused on the long-term changes in obesity, such as longitudinal obesity trajectories [2-5]. However, these studies are either limited by small sample sizes or have typically relied on a single obesity indicator like body mass index (BMI) to identify obesity trajectories. Furthermore, whether the brain effects related to obesity trajectories are involved in modulating cognitive function in the elderly remains unclear. To address these gaps, this study integrated multiple obesity measures simultaneously to identify obesity trajectories. It also investigated their associations with brain morphology, functional connectivity (FC), and cognitive function, and determined whether the effects of obesity trajectories on brain mediate cognitive function.

A total of 502,411 UK Biobank participants over 40 were initially considered at baseline, with 54,243 undergoing 1-2 follow-ups. Of these, 50,538 had full obesity measures, including BMI, waist circumference, waist-hip ratio, and trunk, arm, leg, and body fat percentages, for trajectory analysis. For each participant, age at every timepoint was used as the independent variable and obesity measures at each corresponding timepoint as the dependent variable in linear regressions to calculate slopes of age. The z-scored slopes and baseline measures were analyzed using principal component analysis with varimax rotation and Gaussian Mixture Modeling to identify types of obesity trajectories (Fig. 1a), including the low stable, moderate stable, high stable, increasing, and decreasing trajectories. Subsequently, 33,467 exemplary health participants, without a history of major physical, neurological, or psychiatric diseases, remained. Among them, those with complete data and high-quality MRI images were selected to assess their brain cortical thickness and subcortical volumes (n = 24,663), whole-brain resting-state FC (n = 24,025), and cognitive scores across five domains (n = 14,666~16,950) for group comparison. Specifically, using the low stable group as the reference, separate comparative analyses were conducted with each of the other trajectory groups to assess differences in brain morphology and cognition via one-tailed t-tests, and in FC using two-tailed t-tests, while controlling for sociodemographic factors, lifestyle, etc covariates. Structural equation modeling was applied to examine the mediation effects of brain effects on the relationship between obesity trajectories and cognitive functions.

Five obesity trajectories were identified (Fig. 1a-b). Compared to the low stable group, the high stable, moderate stable, and increasing groups had significantly thinner cortices in the temporal-frontal regions, reduced volumes in several subcortical structures, and extensive alterations in both within-network and between-network FC (Fig. 2a-i). However, the decreasing group exhibited limited cortical thinning in the frontal-temporal regions, no significant subcortical volume reduction, and minimal changes in FC (Fig. 2 j-l). The high stable and increasing groups exhibited the poorest cognitive performance. Mediation analyses suggested that the increasing obesity trajectory mediated cognitive decline in three domains (fluid intelligence, executive function, and processing speed) via alterations in cortical thickness, predominantly in the temporal-frontal regions, and FC, especially within DMN and between the DMN and other functional networks (Fig. 1c).

This study reveals that in aging population, higher stable and increasing obesity trajectories are associated with more severe brain function and structural impairments, as well as cognitive deficits, indirectly suggesting that the maintenance and progression of obesity have detrimental effects on brain health and cognitive function. Importantly, this work highlights the regulatory role of brain effects related to obesity trajectory in cognitive decline.

Aging ²

Connectivity (eg. functional, effective, structural)

Anatomy and Functional Systems ¹

Anatomical MRI

BOLD fMRI

Aging

Cognition

Data analysis

FUNCTIONAL MRI

Statistical Methods

Structures

Other - Obesity