Ultra-Processed Food Intake Linked to Brain Structure Changes and Adiposity-Related Mechanisms

A population-based study using data from over 33,000 participants in the UK Biobank investigated how consumption of ultra-processed foods (UPFs) is associated with structural brain measures and metabolic markers. The analysis included neuroimaging data, blood biomarkers, dietary records, and anthropometric measures to assess whether observed brain changes were mediated by obesity-related factors or occurred independently.

UPF consumption was measured using 24-hour dietary recalls, categorized according to the NOVA food classification system. On average, 46.6% of total energy intake in the cohort was derived from UPFs. General linear models (GLMs) examined associations between UPF consumption and various metabolic, anthropometric, and neuroimaging variables, while mediation analyses were used to explore underlying mechanisms.

Higher UPF intake was significantly associated with multiple adverse health indicators. These included higher body mass index (BMI), waist-to-hip ratio, visceral adiposity, blood pressure, HbA1c, triglycerides, and C-reactive protein (CRP), along with lower levels of high-density lipoprotein (HDL) cholesterol. UPF consumption also correlated with increased intake of saturated fat, sugar, and sodium.

Neuroimaging results indicated associations between UPF consumption and changes in brain microstructure. Specifically, UPF intake was positively associated with mean diffusivity (MD) and negatively associated with intracellular volume fraction (ICVF) in regions including the nucleus accumbens, putamen, and pallidum, suggesting lower cellular density in these areas. Additionally, UPF intake was associated with higher cortical thickness in the lateral occipital cortex and reduced ICVF in the hypothalamus, pallidum, and putamen.

Mediation analyses revealed that certain metabolic factors partially explained the relationship between UPF intake and brain characteristics. For instance, elevated CRP levels mediated the association between UPF intake and MD in the right nucleus accumbens. HDL cholesterol mediated associations between UPF intake and reduced cortical thickness in the occipital cortex and lower fractional anisotropy (FA) in the fornix. Additionally, BMI mediated associations with changes in the hypothalamus, fornix, and occipital cortex.

The study also found that UPF-related structural differences in brain areas associated with feeding behavior—such as the hypothalamus, amygdala, and nucleus accumbens—may be influenced by both adiposity and systemic metabolic alterations. While some changes appeared to be mediated by BMI or biomarkers like CRP and HDL, others persisted after adjusting for these factors, suggesting potential non-obesity-related mechanisms.

While the authors note that causality cannot be established due to the observational nature of the study, the findings align with existing literature suggesting that UPFs may impact health beyond their nutritional content. The study included adjustments for multiple covariates such as total caloric intake, nutrient composition, physical activity, education, and income to account for confounding variables.

The authors emphasize the need for further research to clarify causality and to examine potential pathways through which UPFs influence brain structure, including inflammation and metabolic dysregulation.