Twenty-three men with metabolic syndrome ate only between 8 a.m. and 4 p.m. for one month, and an MRI-based model estimated their brains as about 0.76 years (roughly nine months) younger at the end than at the start. The study was published in Frontiers in Aging on February 18, 2026, and the journal posted a minor correction on November 5, 2026, which is when the coverage picked up.

The Study Design

Researchers at two hospitals in Xi'an, China, recruited men aged 25 to 55 who met at least three of five criteria for metabolic syndrome, including a waist of 90 centimeters (about 35 inches) or more, raised triglycerides, low HDL cholesterol, raised blood pressure and raised fasting glucose. Of 29 men screened, 23 were enrolled. Their average age was 35, and their average weight was 95.6 kilograms (about 211 pounds), with an average body mass index of 31.4.

Everyone followed the same plan, so this is a single-group study that compares each man with himself, with no control group. After a one-month run-in on their usual diet, participants spent one month eating only within an 8-hour window from 8 a.m. to 4 p.m., fasting the other 16 hours. There was no calorie target and no restriction on food types inside the window. Adherence was tracked with continuous glucose monitors, which alerted the study endocrinologist when readings suggested eating outside the window, and with daily meal photos sent to a group chat with the clinical team. Daily calories were not measured.

How Brain Age Was Estimated

Brain age is not a clinical measurement. It is a prediction. Each participant had a structural MRI on a 3 Tesla scanner at the start and end of the month. The gray matter maps from those scans were fed to brainageR, a machine-learning model trained on scans from 3,477 healthy adults, which outputs a predicted age for the brain. Subtracting a person's real age from that prediction gives the brain age gap. A positive gap means the model reads the brain as older than the calendar says.

At baseline the group's average gap was 2.79 years. After the month it was 2.03 years, a drop of 0.76 years, with a reported p value of 0.038. The standard deviation around both averages was close to 5.9 years, so the change is small against the spread between individuals. The paper also reports larger gray matter volume after the month in four small regions on the left side of the brain, including part of the hippocampus and the thalamus.

Memory and Metabolic Results

On a standard word-list recall test, average immediate recall rose from 45 to 54 words across the trials, and delayed recall from 9.4 to 11.3 words. A short-delay interference measure did not change. Repeat testing on the same list format can lift scores through practice, and with no control group the study cannot separate that from any effect of the eating window.

Body weight fell by an average of 4.4 kilograms (about 9.6 pounds), BMI dropped from 31.4 to 30.0, triglycerides and fasting glucose fell, and fasting insulin roughly halved. Total, LDL and HDL cholesterol did not change. Waist and blood pressure are not reported as outcomes. The authors report no outside funding and no conflicts of interest.

What the Authors Say It Cannot Show

The paper lists four limits: the sample is small, it is all men, the intervention and follow-up lasted one month, and because calories were not counted, the study cannot tell whether the changes came from meal timing itself or from simply eating less inside a shorter window. Two more limits sit outside the paper. With no comparison group, any change, including the brain age estimate, could reflect regression to the mean, scanner variability, weight loss or the intensive daily monitoring. And a brain age model's output is a statistical estimate, not a measure of cognition, dementia risk or any clinical outcome.

Where It Sits Against Larger Trials

Randomized trials of time-restricted eating have mostly measured weight and metabolic markers, not the brain, and their results are mixed. The TREAT trial, published in JAMA Internal Medicine in September 2020, randomized 141 adults with overweight or obesity to a noon to 8 p.m. eating window or three structured meals a day for 12 weeks, with no calorie targets in either group. Weight fell 0.94 kilograms in the fasting group and 0.68 kilograms in the control group, a difference that was not statistically significant, and the fasting group lost more lean mass. It was funded by UCSF, the NIH and a foundation, and the authors note that calorie intake was not measured.

A 14-week trial published in the same journal in August 2022 randomized 90 adults with obesity to an early 7 a.m. to 3 p.m. window or a window of 12 or more hours, with both groups also counseled to cut calories. The early window group lost 2.3 kilograms more (6.3 versus 4.0 kilograms), with a 4 mm Hg larger drop in diastolic blood pressure, but no significant difference in fat loss. It was NIH funded, mostly female and smaller than planned because of the pandemic.

Against that record, a one-month, uncontrolled study of 23 men is an early signal that a question is worth asking in a randomized trial with a brain outcome, and not an answer. The number to carry away is 0.76 years on a model estimate, in people who also lost about ten pounds, with nobody in a comparison group. A reader considering an 8-hour eating window, especially anyone with diabetes or on glucose-lowering or blood pressure medication, has a concrete question to bring to a clinician: how a 16-hour fast would interact with their medications and their training schedule.

This article reports research and expert guidance. It is not medical advice. Talk to your clinician before changing medication, treatment or a medical diet.

Sources