A Barcelona hospital followed 83 adults with obesity through a year of routine care and measured what most weight-loss studies only estimate: fat, lean tissue, resting energy expenditure and handgrip strength. The 44 patients on semaglutide lost substantially more weight than the 39 on a lifestyle programme alone. They also lost lean tissue — about 1.1 kg of fat-free mass across the year. What they did not lose was strength: absolute handgrip was preserved in both groups, and measured against body weight it improved on semaglutide.
The numbers
The study, published in Obesity Facts on 25 August 2026, is a single-centre retrospective review at Hospital Universitari Vall d’Hebron covering January 2024 to December 2025. Adults with a BMI of 30 or above and without diabetes were included if they had both baseline and 12-month measurements; the semaglutide group required at least 70% drug exposure over the follow-up. Body composition came from bioelectrical impedance, resting energy expenditure from indirect calorimetry, and muscle was additionally scored against an age- and BMI-standardised reference index validated against MRI.
The semaglutide group’s 12-month reductions were reported as 11.0 kg of body weight, 4.0 kg/m² of BMI and 7.0 cm of waist circumference, each significantly greater than the lifestyle group. Within the semaglutide group, skeletal muscle mass fell from 29.9 kg to 28.7 kg and fat-free mass from 53.8 kg to 52.7 kg, both at p<0.001.
Do the arithmetic and that is roughly 1.1 kg of fat-free mass against about 11 kg of total weight — near a tenth of what was lost, well below the quarter-to-40% share often quoted for lean tissue during rapid weight reduction. Treat the ratio cautiously, though: the abstract does not make explicit whether the 11.0 kg figure is the semaglutide group’s own change or its difference from the lifestyle arm, and those two readings give you different denominators. We read the abstract; the full text sits behind the publisher’s download and we could not open it.
Strength, and the metabolic slowdown
Handgrip is the part worth dwelling on. Absolute grip strength was preserved in both groups, and relative grip — strength per unit of body weight — improved on semaglutide (adjusted p=0.003). On the reference-based muscle index, the semaglutide group improved by 0.52 against 0.09 for lifestyle alone. The authors’ reading is that the muscle left behind is better matched to the smaller body, even though absolute lean mass fell. That is a more useful frame than counting kilograms of lean tissue in isolation, which tells you nothing about whether a person can carry their shopping.
Resting energy expenditure fell in both groups, and part of that fall exceeded what changes in fat mass, fat-free mass, age and sex would predict — the adaptive thermogenesis that makes maintenance hard. The size of it did not differ significantly between groups: −210.2 kcal/day on semaglutide against −373.4 kcal/day on lifestyle. On this evidence semaglutide neither causes the slowdown nor protects against it.
What this cannot settle: it is 83 people at one hospital, reviewed retrospectively, with no randomisation and no blinding — the groups were not assigned, they were treated. Bioelectrical impedance is not DXA or MRI and is sensitive to hydration, which shifts during rapid weight loss. Handgrip is a reasonable general proxy but says little about leg power or day-to-day function. The energy-expenditure comparison is exactly the question a study this size is underpowered to answer, and −210 against −373 kcal/day is a wide gap to file under “no significant difference”. And twelve months tells you nothing about what happens to muscle in year three.
