Urolithin A, the pomegranate-derived 'mitochondria' supplement, was given to 50 adults with obesity for four weeks. Four tests of their blood vessels found nothing. The one signal that moved came from a brain scan, and memory scores stayed where they were

Urolithin A isn't in any food. Gut bacteria make it from ellagitannins, the tannins in pomegranates, walnuts and some berries, and only some people's bacteria do the job well. In animal and early human studies it switches on mitophagy, the cell's way of clearing out worn mitochondria, which is why it has become one of the better-known longevity supplements. Because so many people make little of it themselves, it is sold ready-made in capsules.

Researchers at the University of Oklahoma asked a specific question. Obesity in midlife stiffens and irritates the lining of blood vessels, partly, it is thought, through tired mitochondria. Would urolithin A help? Their trial was published on 9 October 2026 in JAMA Network Open. They enrolled 50 adults aged 40 to 64 with a BMI of 30 or more (the average was 38), three quarters of them women. Half took 1,000 mg of urolithin A a day, two 250 mg capsules before breakfast and two before dinner, and half took matching placebo capsules. Neither the participants nor the people measuring them knew who had which. It lasted four weeks.

The main test was flow-mediated dilation. A cuff squeezes the forearm for five minutes, and when it is released an ultrasound measures how much the artery in the upper arm widens as blood rushes back. A healthy vessel lining widens more. Over the four weeks the figure rose by 1.3 percentage points on urolithin A and 0.3 on placebo, a difference well within chance (p = 0.33). Three other tests of the small vessels in the hand and fingertips showed nothing either. Nor did a panel of blood markers of inflammation and vessel health.

It wasn't that people skipped their capsules. About 96% of doses were taken in both groups, and blood levels of urolithin A's main breakdown product rose by 285 ng/mL in the supplement group against 1 ng/mL on placebo. The compound got in. It didn't visibly change the arteries.

Then the brain. Participants did a letter-memory task wearing a cap that shines near-infrared light through the scalp to track blood oxygen in the outer layer of the brain. After four weeks, the supplement group showed a clearly larger rise in oxygenated blood in the left dorsolateral prefrontal cortex, a region that handles working memory, when the task got harder. The placebo group showed no such rise. This is the result the paper's conclusion points to.

It needs handling with care, for a few reasons. The people with the bigger blood-flow response did no better at the task: accuracy and reaction times were the same in both groups. The authors suggest the task was too easy to show a difference, and they have a point, since 94% of participants scored perfectly on the simpler version before taking anything. But it also means nobody can say the bigger response was useful. In brain imaging, more blood flow for the same performance can be read as a brain being better supplied, or as one working harder for the same result. This trial can't tell which. The brain measure was also one of several secondary outcomes, analysed only in people with complete scans, in a study where 7 of 50 did not come back for the final visit. The paper's abstract says the finding 'warrants further investigation', which is right. A line in its discussion says the supplement 'significantly improved cerebrovascular function', which goes further than the data.

Our reading: this is an honest null result, and those are worth having. It was funded by the American Heart Association and a foundation rather than a supplement company, the authors report no conflicts of interest, and the trial was registered before it began with the artery test as its main outcome. Fifty people for four weeks can't rule out a modest benefit: the range of plausible effects on the artery test ran from slightly worse to almost 5 points better, and the authors want a longer trial of 12 weeks or more. For now, though, there is no evidence that urolithin A improves blood vessel health in people with obesity, and the brain finding is a lead, not a benefit. The things known to help the vessel lining are less exciting: regular exercise, losing weight and not smoking.