Frequent Weight Gain and Loss May Harm Muscle Health, Study Finds - Video

Many people regain weight after dieting and then try to lose it again. But a new study from the University of California, San Francisco (UCSF) suggests that this repeated cycle of losing and gaining weight may speed up muscle loss in middle-aged adults.

The study, published in Radiology, followed 1,433 middle-aged adults for four years. Researchers compared people whose weight stayed fairly stable with those whose weight frequently went up and down.

Using MRI scans and artificial intelligence, they measured the amount of thigh muscle, fat inside the muscles, and fat around the knees.

The results showed that people who repeatedly lost and regained weight lost almost four times more thigh muscle than those whose weight remained stable. Weight cyclers lost about 3.7% of their thigh muscle, while people with stable weight lost only about 1% during the same period.

Interestingly, both groups ended the study at nearly the same overall body weight they had at the beginning. This suggests that while body weight returned, the lost muscle did not.

When people's weight cycled, they lost a lot of muscle along with fat, and they didn't gain the muscle back. The lead researcher also explained that protecting muscle during weight loss is just as important as losing fat.

Although the study did not involve people taking popular weight-loss drugs such as GLP-1 medications, the findings are especially relevant as more people start and stop these treatments.

Researchers say the results highlight the need to focus not only on the number on the weighing scale but also on maintaining healthy muscle mass. Preserving muscle through strength training, adequate protein intake, and healthy weight-loss strategies may be important for long-term health, mobility, and reducing the risk of conditions such as arthritis.

REFERENCE: Marth, A. A., et al. (2026). Association of Weight Cycling with MRI-based Thigh Muscle Quality and Knee-adjacent Obesity: Osteoarthritis Initiative Data. Radiology. DOI: 10.1148/radiol.253988. https://pubs.rsna.org/doi/10.1148/radiol.253988