Middle-aged adults who ate the most ultra-processed food were more likely to show signs of vascular injury on brain scans, and carried more body fat, than those who ate the least, according to a study of 8,783 Canadians published Oct. 15 in JAMA Network Open. The same participants performed no differently on two standard cognitive tests, which the authors say suggests the brain changes may be detectable before any measurable decline in thinking.
The peer-reviewed study is a cross-sectional analysis of two existing cohorts, the Canadian Alliance for Healthy Hearts and Minds (CAHHM) and PURE-MIND. It was led by Russell J. de Souza of McMaster University and the Population Health Research Institute (PHRI) in Hamilton, Ontario, with Sonia S. Anand of McMaster as corresponding author. Participants were aged 35 to 69, had no history of cardiovascular disease and had a mean age of 57.7 years; 56.6% were women. Diet was measured once with a validated food frequency questionnaire, and ultra-processed food intake was expressed as servings per 2,000 kilocalories a day.
Vascular brain injury was defined on magnetic resonance imaging (MRI) as covert brain infarcts, sometimes called silent strokes, or a high burden of white matter hyperintensities. Among participants in the lowest quarter of ultra-processed food intake, 8.1% had such injury; in the highest quarter, 11.3% did. After adjustment, the highest-intake group had 40% higher odds of vascular brain injury than the lowest (adjusted odds ratio 1.40; 95% confidence interval 1.10 to 1.77), with a significant trend across intake levels.
The highest consumers also had 1.7 percentage points more body fat and 5.7 mL more visceral fat on MRI than the lowest, but no significant difference in liver fat. Adjusting for adiposity did not materially change the brain findings, the authors report, so body fat alone did not appear to explain the link. Scores on the Montreal Cognitive Assessment and the Digit Symbol Substitution Test did not differ significantly across intake groups.
In an exploratory analysis, the association with vascular brain injury was present in women (odds ratio 1.78; 95% CI 1.27 to 2.48) but not in men (1.08; 0.76 to 1.54). The paper also notes that the brain finding depended on how exposure was measured: when ultra-processed food was expressed as a share of total energy or of all food servings, rather than servings per 2,000 kcal, the association was not consistent.
“Usually, this kind of damage is associated with conditions we hear about and treat at a much older age,” de Souza said, according to a HealthDay report. Anand said the results suggest “the link between diet and brain health may be detectable before people experience noticeable problems.” The team told HealthDay the study does not prove cause and effect and that longer-term research is needed.
Why it matters
Covert infarcts and white matter hyperintensities are markers of small vessel disease that are usually studied in older adults, where they precede clinical stroke and cognitive decline. Finding them more often in otherwise healthy people in their 40s, 50s and 60s who eat more ultra-processed food extends to the brain a body of evidence that has so far linked these foods mainly to obesity and cardiometabolic disease. Ultra-processed foods make up a large and growing share of Canadian diets, the authors write.
For clinicians, the absence of a cognitive signal is itself informative. The authors suggest that imaging markers may precede detectable cognitive deficits in midlife, when brief cognitive tests have limited sensitivity. Independent experts asked to comment by the UK Science Media Centre said the findings justify further research into diet and brain blood vessel health, while cautioning that the design cannot show that diet caused the injury.
The caveats
The study is cross-sectional: diet, body composition and MRI were assessed at a single point in time, so it cannot establish which came first, and the authors acknowledge that reverse causation is possible. A single questionnaire captured about one year of habitual intake, while adiposity and vascular brain injury reflect much longer exposure. Applying the NOVA processing classification to questionnaire data may misclassify some foods, and ultra-processed food is a heterogeneous category. The sex-stratified results were exploratory. The highest-intake group also contained more participants of white European background, which the authors say may limit generalizability.
Jules Griffin, director of the Rowett Institute at the University of Aberdeen, told the Science Media Centre the study was well conducted but that the authors did not adjust for the higher salt, sugar and saturated fat intake typical of heavy ultra-processed food consumers. Derek Hill, professor of medical imaging at University College London, said the MRI scoring approach is widely used and reliable, but that proving causation would require following people over time.
CAHHM was funded by the Canadian Partnership Against Cancer, the Heart and Stroke Foundation of Canada and the Canadian Institutes of Health Research. PURE received support from PHRI, Hamilton Health Sciences, public funders and unrestricted grants from pharmaceutical and dairy-industry sources. The funders had no role in the study, the paper states. Several authors reported grants or personal fees from pharmaceutical companies unrelated to the work.
What to watch
The authors say longitudinal studies are needed to establish whether high ultra-processed food intake precedes vascular brain injury, and whether changing diet alters the course of these early brain changes.
This article is for general information and does not replace advice from a qualified clinician. Corrections: email the editor using the address on the Contact page.
