In a controlled Virginia Tech feeding trial published in Nature Metabolism, healthy adults consuming an ultraprocessed 300-calorie meal experienced significantly higher insulin spikes and altered energy use compared to a nutritionally matched non-ultraprocessed meal, despite near-identical blood glucose curves and macronutrient profiles.
In Plain English: The Clinical Takeaway
- Insulin Spikes: Meals built from ultraprocessed foods made the body release more insulin between 40 and 120 minutes after eating, even when calories, sugar, and fiber counts were identical.
- Altered Fuel Use: After eating ultraprocessed items, the body burned less carbohydrate and oxidized more fat, while overall metabolic rate temporarily climbed higher.
- Brain Response: Functional MRI brain scans showed that differences in carbohydrate metabolism correlated directly with activity in neural reward circuits when participants viewed food images.
Metabolic Chamber Testing Reveals Divergent Insulin Responses
The randomized crossover study enrolled 57 healthy-weight adults in Roanoke, Virginia, with data collection running from June 2023 through December 2025. A subset of 32 participants consumed nutritionally matched ultraprocessed (UPF) and non-ultraprocessed meals of approximately 300 calories each on separate days. Meals matched on weight, energy, macronutrients, glycemic index, fiber, sodium, and water (all ≤1.6% deviation). During 4 hours of whole-room indirect calorimetry, researchers tracked blood markers and respiratory exchange ratios.
While total glucose exposure measured by area under the curve (AUC) remained similar between conditions, the temporal pattern diverged. Blood glucose rose faster during the first 20 minutes following the non-UPF meal. Conversely, insulin concentrations were higher after the ultra-processed meal at 40, 60, 90 and 120 minutes. By three hours, insulin levels for both test groups had almost returned to baseline. Energy expenditure also shifted: this effect lasted longer after the ultra-processed food, so total post-meal energy expenditure was higher. Following the consumption of the unprocessed meal, the body’s energy selection leaned more heavily on carbohydrates, whereas fat burning predominated after the ultraprocessed food.
| Metabolic Outcome | Non-UPF Meal Response | UPF Meal Response |
|---|---|---|
| Blood Glucose AUC | Rose faster initially (higher at 20 min) | Similar total exposure |
| Blood Insulin AUC | Lower sustained output | Higher concentrations (40–120 min) |
| Metabolic Rate AUC | Standard postprandial rise | Extended elevation in energy expenditure |
| Respiratory Exchange Ratio (RER) | Greater carbohydrate oxidation | Increased fat oxidation |
Neuroimaging Correlates Metabolic Shifts with Brain Reward Hubs
Beyond postprandial metabolic chamber metrics, the investigation expanded to examine neurological processing. All 57 enrolled participants completed a functional MRI session, with 52 included in the final neuroimaging analysis. Subjects evaluated 28 familiar food images—14 ultraprocessed and 14 non-ultraprocessed—during a Becker–DeGroot–Marschak auction task to measure willingness to pay and neural reactivity.
Variations in how the body processed carbohydrates tracked alongside distinct neural reactions to food pictures within regions governing reward, learning, and sensory processing. Furthermore, brain activity tracking food value differed by processing level in the left fusiform gyrus, left lingual gyrus, right putamen, and right caudate. Value was positively associated with activity for non-UPF cues and negatively for UPF cues. However, participants showed no significant difference in willingness to pay for foods from the two categories.
“If you look at that population-level data, people who consume large amounts of ultraprocessed foods have higher rates of poor health outcomes, obesity, cardiac events, type 2 diabetes, and even some metrics of mental health,”
Alex DiFeliceantonio, Ph.D., Associate Professor, Fralin Biomedical Research Institute
Alex DiFeliceantonio and colleagues at Virginia Tech noted that physical and chemical food structures—rather than just macronutrient labels—may alter nutritional availability via the gut-brain axis. Carlos Monteiro, M.D., Ph.D., Professor of Public Health at the University of São Paulo and creator of the NOVA classification system, commented on the study.
Contraindications & When to Consult a Doctor
While this clinical trial illuminates short-term physiological responses in healthy-weight adults (mean age 26.2 years, mean BMI 22.8 kg/m²), the authors caution that the results do not prove that ultra-processed foods cause these conditions. The study did not measure direct insulin sensitivity, nor did it evaluate long-term body weight, eating behavior, or the risk of chronic diseases. Individuals managing diagnosed metabolic disorders, insulin resistance, or type 2 diabetes should rely on comprehensive glycemic monitoring rather than isolated post-meal feeding data. Patients experiencing unexplained glycemic variability, chronic fatigue, or unintended weight shifts should consult a qualified physician or registered dietitian for personalized diagnostic evaluation.
Funding for the trial was provided by grants from the National Institutes of Health and the National Science Foundation.
