Repeated weight loss and regain may leave the body with less muscle even when the number on the scale returns to roughly where it started, according to an observational study of middle-aged and older adults.
Researchers found that participants whose weight repeatedly fluctuated lost nearly four times as much thigh muscle over four years as those whose weight remained relatively stable. However, the study did not determine whether the weight changes were intentional, prove that weight cycling caused the muscle loss or examine people taking today’s GLP-1 weight-loss medications.
The study, published in the journal Radiology, included 1,433 participants in the Osteoarthritis Initiative, a long-running federally funded study of adults with knee osteoarthritis or risk factors for developing it. Researchers used MRI scans and health data to track changes in weight, thigh muscle volume and fat around the muscles and knee.
Participants classified as weight cyclers lost an average of about 3.7% of their thigh muscle volume over four years. Those whose weight remained relatively steady lost about 1%.
The difference remained after researchers accounted for age, sex, starting body mass index, physical activity, diet and several health factors. Still, other differences between the groups may have contributed to the results.
“When people’s weight cycled, they lost tremendous amounts of muscle along with the fat, and they didn’t gain the muscle back,” said Thomas Link, a professor of radiology at the University of California, San Francisco, who led the study. “This tells us that understanding how to protect muscle while people are losing weight is important for overall health.”
Weight cycling is commonly called “yo-yo dieting,” but that description is not necessarily accurate for this study. Researchers tracked fluctuations in participants’ weight but did not establish why those changes occurred.
Some participants may have intentionally lost weight through dietary changes. Others could have experienced weight changes related to illness, medication, stress, appetite, pain or changing activity levels.
That distinction is especially important because everyone in the study had knee osteoarthritis or an elevated risk of developing it. Knee pain and limited mobility could affect both a person’s weight and the amount of muscle in the thighs.
Because the study was observational, researchers also could not determine which came first. Weight cycling might contribute to muscle loss, but declining muscle, worsening joint health or reduced activity could also make a person’s weight more likely to fluctuate.
The study nevertheless highlights a limitation of using body weight or BMI alone to judge what is happening inside the body.
Body weight includes muscle, fat, bone and water. A person can lose fat and muscle while losing weight, then regain mostly fat afterward. The scale may return to its starting point without the person regaining all the muscle that was lost.
One MRI example included in the study showed a 62-year-old participant whose thigh muscle volume declined by 16% over four years while his BMI fell by only 1.6%. That individual image does not represent the average participant, but it illustrates how a relatively small change in BMI can obscure a much larger change in muscle.
“We need to better understand how weight loss affects the body in ways that aren’t about the number on the scale,” Link said.
The researchers measured muscle volume in the thighs using MRI scans. They did not measure total-body muscle loss, so the findings should not be interpreted as showing that participants lost the same proportion of muscle throughout their bodies.
Muscle volume is also not identical to muscle strength. The study does not establish from imaging alone that weight cyclers became weaker, lost mobility or experienced a higher risk of falls.
Even so, thigh muscles are particularly important for standing, walking and supporting the knees. Losing muscle with age can make everyday movement more difficult, especially for people already experiencing joint problems.
“Fat loss is just one piece of a much larger puzzle that includes how weight loss affects arthritis, muscle, bone, and other factors,” Link said. “As more people use effective weight-loss therapies, we need to be thinking about how those pieces fit together.”
The growing use of GLP-1 medications makes the question of muscle preservation increasingly timely, but the new study did not evaluate those drugs.
Participants were followed through the Osteoarthritis Initiative, which began years before medications such as semaglutide and tirzepatide became widely used for weight management. The results therefore cannot show what happens when people begin, stop or restart a GLP-1 medication.
They do raise broader questions about how repeated weight loss and regain might affect body composition, regardless of how the weight was lost.
Weight loss often includes some loss of lean tissue along with fat. How much muscle a person retains can depend on factors such as age, starting body composition, the speed and amount of weight loss, protein intake, physical activity and resistance exercise.
The study did not identify a particular diet, exercise plan or protein target that prevented muscle loss. It also should not be interpreted as evidence that people should avoid medically appropriate weight loss.
For people with overweight or obesity, weight loss may improve blood sugar, blood pressure, mobility and other aspects of health. The findings instead suggest that preserving muscle deserves attention alongside changes in body weight.
Future research will be needed to determine whether weight cycling directly contributes to muscle loss, which people are most vulnerable and whether nutrition or resistance training can help people regain muscle after their weight rises again.
Longer studies could also examine whether the pattern affects strength, mobility, independence or the risk of developing age-related muscle loss.
The research was supported by the National Institutes of Health and the UCSF Core Center for Musculoskeletal Biology and Medicine.