Hidden body fat may put you at greater brain disease risk

Body fat rarely stays in one place. A new imaging study suggests that the spots where fat settles can line up with changes in your brain.

Researchers at The Affiliated Hospital of Xuzhou Medical University in Xuzhou, China, analyzed MRI scans from nearly 26,000 adults in the UK Biobank. The team focused on how fat spreads across organs and tissues, not just on body mass index. Their results were published today in Radiology, the flagship journal of the Radiological Society of North America.

The study links two underrecognized fat patterns to smaller brain volumes, more signs of tissue damage, slower thinking speed, and higher odds of certain neurologic and psychiatric conditions. Those patterns showed up in both men and women, with some differences by sex.

Body fat is often reduced to a single score, such as BMI. That number can hide major differences in fat placement. MRI lets researchers measure “ectopic” fat stored inside organs, along with fat around the abdomen, heart, and muscles.

The team used nine MRI-derived fat measures from abdominal and cardiac scans. They included liver fat and pancreas fat, along with visceral fat around organs, fat under abdominal skin, pericardial fat, and muscle fat infiltration. They also included ratios tied to abdominal fat and weight-to-muscle balance.

To highlight distribution rather than body size, the researchers adjusted each fat measure for BMI. Then they used latent profile analysis, a statistical grouping method, to sort people into patterns. The analysis was done separately for men and women.

Out of more than 500,000 UK Biobank volunteers, 26,233 had complete imaging and BMI data. After removing 236 people with extreme outlying measurements, the final sample included 25,997 participants. The average age was 55, with ages ranging from 37 to 73. Women made up about 52% of the group.

Six fat distribution profiles emerged in both sexes. Two became the clearest red flags for brain-related outcomes.

One was “pancreatic predominant,” marked by unusually high fat in the pancreas. In that group, pancreatic fat was far higher than in other categories. Study coauthor Kai Liu, M.D., Ph.D., an associate professor in The Affiliated Hospital’s Department of Radiology, said some participants had a striking level of pancreas fat on MRI.

“Individuals with ‘pancreatic-predominant’ distribution patterns showed a proton density fat fraction; an MRI marker that provides a precise estimation of fat concentration in tissue; of around 30 percent in the pancreas,” Dr. Liu said.

“This level is about two to three times higher than that of other fat distribution categories, and it can be up to six times higher than that of lean individuals with low overall fat,” Dr. Liu said. “Additionally, this group tends to have a higher BMI and overall body fat load.”

The other high-risk pattern was “skinny fat.” This group carried heavy fat burdens across many depots, even though BMI was not the highest. In men, the “skinny fat” pattern often paired high fat stores with a higher weight-to-muscle ratio.

“Most notably, this type does not fit the traditional image of a very obese person, as its actual average BMI ranks only fourth among all categories,” explained Dr. Liu. “The increase is perhaps more in fat proportion. Therefore, if one feature best summarizes this profile, I think, it would be an elevated weight-to-muscle ratio, especially in male individuals.”

The study also identified a liver-heavy pattern, two “balanced” patterns, and a lean pattern with the lowest fat across depots. The lean group served as a benchmark for many comparisons.

Brain imaging came from a separate MRI protocol used in the UK Biobank. Scans were performed on a 3.0-T Magnetom Skyra MRI scanner with a 32-channel head coil. Researchers compared total brain volume, gray matter volume, and white matter hyperintensity volume.

"Across both sexes, the higher-fat patterns generally tracked with lower total brain volume and lower gray matter volume. White matter hyperintensities were also higher outside the lean group. These bright spots on MRI are often treated as signs of injury or disease burden in brain wiring," Dr. Liu told The Brighter Side of News.

"The pancreatic-predominant pattern showed some of the most pronounced differences. In men, total brain volume averaged about 1.46 million mm3 in that group, compared with about 1.50 million mm3 in the lean group. Gray matter volume was also lower in the pancreas-heavy group," he added.

In women, total brain volume averaged about 1.48 million mm3 in the pancreatic-predominant group, compared with about 1.52 million mm3 in the lean group. Gray matter volume followed a similar pattern.

The study also looked at white matter microstructure using neurite orientation dispersion and density imaging, known as NODDI. Differences appeared more strongly in men than women. Men in higher-risk patterns tended to show signals consistent with lower neurite density and more free-water fraction.

Researchers also estimated “brain age gap,” the difference between MRI-predicted brain age and actual age. That gap varied more clearly across profiles in men than in women.

Cognitive results showed the sharpest differences in measures tied to speed and memory. One key measure was psychomotor speed, based on the time needed to correctly identify matches.

In men, the lean profile averaged about 517 milliseconds. The pancreatic-predominant group averaged about 541 milliseconds. The “skinny fat” group averaged about 542 milliseconds. Women showed a similar pattern, with slower average times in higher-risk profiles.

Memory measures also differed. In men, prospective memory was poorer in several higher-risk profiles compared with lean. Visual memory declines appeared in specific groups too. A global cognitive score was lower than lean only in women in the pancreatic-predominant profile.

The analysis also examined neurologic and psychiatric conditions using ICD codes and self-reported diagnoses. Models adjusted for many factors, including age, BMI, education, smoking, alcohol use, hypertension, diabetes, lipids, and physical activity.

In men, the “skinny fat” pattern stood out for elevated odds of anxiety, depressive episodes, stroke, and other outcomes. In women, profiles including pancreatic-predominant, liver-heavy, and “skinny fat” showed higher risks across several conditions. In the pancreatic-predominant profile, women showed higher odds for stroke and epilepsy compared with the lean group.

This work argues for a shift in how risk is discussed in clinics and research. BMI alone can miss people whose fat distribution signals higher brain vulnerability, especially those who do not look “typically” obese.

The findings may guide more personalized prevention strategies. MRI-based fat measurements could help identify patients who might benefit from earlier brain-health screening, stronger cardiovascular risk control, and tailored exercise plans that build muscle as well as reduce harmful fat stores.

The study also points to new targets for research. Scientists can test whether reducing pancreatic fat, improving muscle mass, or shifting abdominal fat patterns changes brain outcomes over time. Longer, follow-up studies could also clarify cause and effect, since this analysis shows associations, not proof of causation.

As Dr. Liu explained, “Brain health is not just a matter of how much fat you have, but also where it goes.”

Research findings are available online in the journal Radiology.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.