Ultra-processed foods linked to silent brain injury on MRI in healthy adults
A high intake of ultra-processed foods was associated with early MRI signs of brain injury in otherwise healthy middle-aged adults, raising concern that diet quality may influence cerebrovascular health long before memory complaints appear. In a Canadian cohort of nearly 9,000 adults aged 35 to 69, people with the highest intake of these foods were 40% more likely to show small areas of brain damage than those with the lowest intake.
The MRI findings matter because they are not cosmetic abnormalities. White matter hyperintensities and silent brain infarcts are established markers of small-vessel disease, and both have been linked in prior research to higher risks of stroke, cognitive decline, and dementia. These lesions often develop quietly, without a dramatic clinical event, which makes them especially important in preventive neurology.
The study appeared in JAMA Network Open and drew on participants from the Canadian Alliance for Healthy Hearts and Minds and the Prospective Urban and Rural Epidemiology-MIND study. Each participant completed dietary questionnaires and underwent brain imaging. Neurologists reviewed the scans without knowing the dietary data, reducing the chance that expectations influenced image interpretation.
MRI lesions emerged before memory problems did
A striking feature of the analysis is the disconnect between imaging and symptoms. The cohort was relatively healthy, and no measurable differences in memory or thinking were seen, even though MRI markers of vascular brain injury were already more common in those who ate the most ultra-processed food. In practical terms, this places dietary exposure in the same conversation as silent vascular injury rather than late-stage neurologic decline.
Russell de Souza, ScD, RD, the study’s first author and an associate scientist at the Population Health Research Institute of McMaster University and Hamilton Health Sciences, said the association was concerning because the MRI abnormalities are not considered benign. He also drew a distinction between relative and absolute risk. A 40% relative increase does not mean 40% of people will develop brain injury; it means the likelihood was higher in the top intake group, while the underlying MRI findings remained relatively uncommon in middle-aged adults.
That distinction is clinically important. A modest risk shift can still matter when the exposure is common, and ultra-processed foods are widely consumed across many countries. If the signal is real, small increases in risk could accumulate across large populations and translate into a meaningful burden of stroke-prone brain disease over time.
Women showed a stronger association than men
The sex-specific result was one of the more unexpected findings. Women in the highest ultra-processed food intake group were nearly twice as likely to show signs of brain damage as women in the lowest intake group. The reasons remain uncertain, and the analysis does not prove that women are biologically more vulnerable, but the pattern is difficult to ignore.
Several plausible explanations exist. Midlife hormonal changes, especially around the menopausal transition, may alter vascular risk. Men and women also differ in body fat distribution, cardiometabolic risk, and vascular biology, all of which can shape cerebrovascular health. A dietary pattern rich in ultra-processed foods may also displace different foods in women’s diets than in men’s diets, producing different nutrient and fiber deficits.
Residual confounding remains possible, particularly because subgroup analyses are based on smaller numbers than the full cohort. Chance cannot be excluded. Even so, the sex difference provides a useful hypothesis for future work rather than a reason to dismiss the signal.
Why ultra-processed foods may affect the brain
The most biologically plausible pathway runs through metabolic and vascular injury. Diets high in ultra-processed foods are often dense in salt, sugar, and unhealthy fats, and they tend to track with obesity, diabetes, and cardiovascular disease. In this study, higher consumers also carried more body fat and more deep abdominal fat, or visceral fat, which is strongly tied to inflammation and cardiometabolic disease.
Small vessels in the brain are especially vulnerable to this kind of systemic stress. When those vessels become impaired, oxygen and nutrient delivery to brain tissue can suffer, and over time the result may be white matter damage or small silent infarcts. Those lesions may not produce immediate symptoms, but they can set the stage for future stroke or cognitive problems.
One sentence captures the bedside meaning: silent vascular injury can accumulate long before a patient notices a change in memory.
The researchers did not prove that body fat, inflammation, or vascular damage explained the association. Because diet and MRI were assessed at a single point in time, the study cannot determine whether higher ultra-processed food intake came first or whether another factor drove both the diet pattern and the imaging findings. The analysis also cannot show whether reducing ultra-processed food intake would slow the development of these lesions.
Ultra-processed foods may be replacing healthier staples
The dietary pattern itself was revealing. Ultra-processed foods accounted for about 25% of daily food servings overall, but among the highest consumers they made up 45% of daily servings, compared with 16% in the lowest intake group. At the same time, whole and minimally processed foods made up 41% of daily servings in the highest intake group, versus 68% in the lowest.
This suggests substitution rather than simple addition. The more ultra-processed foods people ate, the fewer fruits, vegetables, whole grains, and other minimally processed foods they consumed. From a nutritional standpoint, that matters because fiber, micronutrients, and food structure all support vascular and metabolic health. A diet built around convenience foods can quietly crowd out the foods most likely to protect the brain.
Affordability, convenience, availability, and marketing likely shape those patterns. For many people, ultra-processed foods are easier to buy and store than fresh alternatives. That reality complicates any public health message aimed at reducing intake, because the issue is not only individual choice but also access, price, and routine.
What the findings can and cannot support
The study does not show that ultra-processed foods cause brain damage, and it does not mean that everyone who eats them will develop stroke or dementia. It does, however, add another layer to the evidence that long-term dietary quality may influence cerebrovascular injury well before overt neurologic disease becomes apparent.
The most defensible takeaway is narrow but meaningful: middle age may be the window when vascular brain injury begins to appear on imaging, and eating patterns during this period could influence that process. Future studies will need to determine whether the association persists over time, whether specific types of ultra-processed foods carry different levels of risk, and whether lowering intake changes MRI outcomes.
For now, the study strengthens the rationale for diets built around minimally processed foods, especially fruits, vegetables, beans, whole grains, and other nutrient-dense staples. It also leaves open a harder question: whether the risk comes from the processing itself, the nutrients displaced, or both.