Brown fat and your metabolism: what a 2026 Nature study of 100,000 scans found
Gómez-Banoy et al. in Nature linked brown fat on 100k+ PET scans to better cardiometabolic markers. What exercise, mild cold and losartan findings mean—and do not.
revns editorial · Published November 22, 2026 · 8 min read

Gómez-Banoy et al. used AI on 100k+ PET/CT scans. Brown fat on imaging matched better glycemic, lipid and echo markers and lower anxiety or depression codes—association, not cause.
Brown fat is real in adults, and a 2026 Nature study of more than 100,000 PET/CT scans links detectable brown adipose tissue with better cardiometabolic markers and lower recorded rates of some mood diagnoses—but it does not prove that activating brown fat will fix your weight or mental health. Nicolás Gómez-Banoy, Paul Cohen at Rockefeller University and colleagues at Memorial Sloan Kettering Cancer Center used machine learning on oncology follow-up imaging to phenotype brown fat across six body regions. People with active brown fat on scans tended to have healthier blood sugar, lipid and heart ultrasound indices in the paper's summary, and lower odds for several cardiometabolic conditions than people without—patterns that echo Cohen's 2021 analysis of roughly 50,000 patients. For people who train, the takeaway is not a new supplement or prescription: keep evidence-based habits—movement, sleep, whole-food meals—and treat this paper as a metabolism map, not a personal prescription.
TL;DR
| Topic | What Gómez-Banoy et al. report in Nature (2026) | What it does not show |
|---|---|---|
| Data | AI-enabled phenotyping on 100,000+ PET/CT scans (MSKCC cancer monitoring cohort) | Not a randomized diet or exercise trial in you |
| Regions | Brown fat scored in six areas (neck, axilla, mediastinum and others) | Location mattered for some associations (e.g. axillary fat vs blood pressure codes) |
| Cardiometabolic | Better average glycemic, lipid, echo and inflammation markers when brown fat was detected | Causation; sick people may scan less often |
| Mood codes | Lower rates of anxiety and depression diagnostic codes when brown fat was present | Not proof exercise or fat treats mental illness |
| Trade-offs | Higher rates of osteoporosis and some skin conditions in the brown-fat group | May reflect confounding or scan indications |
| Losartan | Activated brown-fat cells in culture; raised energy use in healthy mice via pathways separate from blood-pressure lowering | Not a human weight-loss drug study |
| Your week | Train consistently, sleep, eat enough protein, use easy cold after hard sessions only if you tolerate it | Do not change blood-pressure meds for brown fat |

What is brown fat?
Brown adipose tissue is a specialized fat that burns calories to make heat instead of mainly storing them. Its cells hold many mitochondria and a protein called UCP1 that lets energy escape as warmth. That is useful in cold rooms and may add a small share of daily expenditure when the tissue is active.
Adults do not carry large brown-fat pads like newborns, but PET/CT scans can still detect depots in the neck, shoulders, upper back and around the heart when brown fat is switched on—often after a mild cold stimulus during the scan protocol.
White fat stores energy. Brown fat spends it. Searchers asking brown fat vs white fat should think storage versus furnace, not good versus bad: you need both types, and the science asks how much the furnace runs day to day.
What did the 2026 Nature study find?
Rockefeller University's press materials and ScienceDaily summarized the peer-reviewed paper "Artificial intelligence-enabled clinical phenotyping of human brown fat" (Nature, 2026; DOI 10.1038/s41586-026-11162-7).
The team reverse-translated from patients to bench science:
-
Clinical phenotyping at scale. Memorial Sloan Kettering collaborators built a machine-learning read of brown fat on routine PET/CT scans collected during cancer surveillance. The pipeline scored total brown-fat activity and activity in six anatomical regions.
-
Health-record links. Compared with people who had no detectable brown fat, those with brown fat showed better average glycemic parameters, lipid profiles, echocardiographic indices and lower inflammatory markers in the study's descriptive summaries. The scope was wider than Cohen's 2021 paper (~50,000 patients, mainly cardiovascular codes): here the team searched 500+ diagnostic codes.
-
Regional nuance. Activity in the axillary (armpit) region was linked to lower blood pressure code rates than some other depots—suggesting that where brown fat sits may matter, not just whether it appears.
-
Mental health codes. People with detectable brown fat had lower recorded rates of anxiety and depression. The authors speculate that metabolic health and psychiatric diagnoses may correlate; this is the first time Cohen's group highlighted psychiatric associations in their brown-fat program. That is not evidence that activating brown fat treats mood disorders.
-
Downsides in the same database. Brown fat also associated with higher rates of osteoporosis and certain skin diseases. Observational studies can show trade-offs or confounding; they do not assign risk to individuals.
-
Medication signals. Patients taking angiotensin receptor blockers (ARBs)—blood-pressure drugs including losartan—were more likely to show detectable brown fat. Lab work then showed losartan increased metabolic activity in cultured brown-fat cells and in healthy mice (higher oxygen use and carbon dioxide production, stable body weight) through pathways distinct from blood-pressure control. SSRIs, common antidepressants, appeared to suppress brown-fat activity in the imaging analysis—a hypothesis for future work, not a reason to alter psychiatric care.
The authors call the losartan finding a starting point for drugs that might target metabolism without lowering blood pressure. Nothing in this paper approves new prescriptions for gym-goers.
Is brown fat good for you?
In this cohort, detectable brown fat went with healthier average cardiometabolic numbers and fewer anxiety or depression codes—but the design cannot say brown fat caused those outcomes.
People who show brown fat on scans may also exercise more, sleep better, live in cooler climates or differ genetically. Cancer survivors undergoing frequent imaging are not the same as a random gym cohort. Treat the results as population patterns, not a score you should chase on a consumer gadget. Most watches and rings do not measure brown fat.
What you can use is the direction of the association: habits that support metabolic health—regular aerobic and resistance training, adequate protein, fibre, and sleep—align with the same markers this paper ties to brown fat presence. See blood pressure by age for how training and weight relate to hypertension, and walking after eating for a low-risk glucose habit.

How does brown fat generate heat?
Brown fat mitochondria burn glucose and fat to make warmth instead of ATP for movement. Cold triggers nerves and hormones (such as norepinephrine) that switch brown fat on. That is non-shivering thermogenesis: you stay comfortable without visible shaking.
In the lab, Cohen's group showed losartan can bump brown-fat metabolic activity through signaling paths that do not mirror its blood-pressure effect. In mice, that raised energy expenditure without changing weight in the experiments ScienceDaily described—mice are not humans, and short lab runs do not predict years of lifestyle.
For trainees, the more relevant question is how brown fat activation fits a week you can repeat:
| Habit | Role for brown fat and metabolism |
|---|---|
| Strength + cardio | Mostly works muscle; improves insulin sensitivity and TDEE regardless of scan results |
| Mild cold exposure | Can increase brown-fat signaling in short trials; respect heart and skin limits |
| Sleep | Poor sleep raises appetite hormones and blunts recovery |
| Enough calories | Chronic large deficits suppress hormones that support training |
Use the TDEE calculator for maintenance context, not to micromanage brown fat.
Does cold exposure after workouts help?
Short, tolerable cool-downs may nudge thermogenesis, but recovery evidence is mixed—and safety comes first. Interest spiked after influencers paired ice baths with fat-loss claims. A controlled Journal of Thermal Biology trial in 40 active adults (summarized in our hot or cold after exercise guide) found cold-water immersion did not restore strength faster than heat or rest after damaging eccentric leg work, even if soreness felt different.
If you experiment:
- Prefer cool showers or brief outdoor walks in cool weather over extreme ice when you are new.
- Skip aggressive cold with uncontrolled high blood pressure, Raynaud phenomenon, pregnancy, fever, or recent concussion.
- Stop for chest pain, fainting, numbness that will not warm up, or confusion.
Brown-fat PET scans use controlled cooling in hospitals. Copying that intensity at home is unnecessary and risky.
What should you do this week?
-
Keep your training log honest. Log hard, easy and rest days. If loads climb while sleep falls, deload before you hunt metabolism hacks. Stress, sleep and training load walks through the checks.
-
Eat for your sport. Protein across vegetarian, vegan and omnivorous plates supports muscle; fibre supports gut health. A modest calorie deficit for fat loss should still leave fuel for sessions.
-
Use steps after big meals. A 10-minute easy walk after eating can blunt glucose spikes in multiple trials—orthogonal to brown fat, but practical.
-
Track mood next to load. The Nature paper's anxiety and depression code links are intriguing for researchers, not self-diagnosis. If low mood lasts two weeks or more, or you have thoughts of self-harm, contact a clinician or crisis line (U.S. 988; UK and Ireland Samaritans 116 123). revns can log mood beside training; it does not treat conditions.
-
Connect wearables you already use. On iPhone, revns can read Apple Health workouts, sleep, resting heart rate, HRV and weight into one day view—useful for spotting overreach even though no consumer device reports brown fat.

When should you see a clinician?
Ask for medical review—not more cold exposure—if you notice:
- Blood pressure that stays high despite treatment, or dizziness when standing.
- Chest pain, palpitations or fainting during exercise or heat/cold exposure.
- Unexplained weight loss, night sweats or persistent fever.
- Bone pain or fracture with minimal trauma (the study linked brown fat with higher osteoporosis codes at population level; that does not predict your bones, but it is a reminder that weight cycling and low energy availability hurt bone health in athletes).
- Low mood, panic or insomnia most days for two weeks or longer.
This is general training guidance, not medical advice.
Related
- Hot or cold after exercise: what a 2026 study found about ice baths
- What is a calorie deficit? Meaning, a safe size and how to keep training
- Epigenetic clocks, diet and exercise: what a 2026 Nature Medicine study says
- Blood pressure by age: what is normal, why it rises and what to do
Sources
- Gómez-Banoy N, Barrero JA, Pinheiro de Carvalho LR, et al. Artificial intelligence-enabled clinical phenotyping of human brown fat. Nature. 2026. doi:10.1038/s41586-026-11162-7
- Rockefeller University. Press summary via ScienceDaily, 21 November 2026. A common blood pressure drug has a surprising effect on calorie-burning brown fat
- Becher T, Palanisamy S, Kramer DJ, et al. Brown adipose tissue is associated with cardiometabolic health. Nature Medicine. 2021. doi:10.1038/s41591-020-1126-7 (MSKCC cohort; cited by the 2026 authors for prior cardiovascular associations.)
Terms in this post
revns publishes training and nutrition guidance, not medical advice. See a clinician for symptoms, injury or a medical condition. Accurate as of November 22, 2026.


