Do GLP-1 medicines cause muscle loss?
The headline trials measured weight, not body composition. The body-composition answers come from smaller DEXA substudies run inside those trials, and from reviews that pooled them. The numbers are consistent enough to plan around.
What did the STEP 1 and SURMOUNT-1 substudies find?
In the STEP 1 trial (Wilding and colleagues, New England Journal of Medicine, 2021), adults with obesity on semaglutide 2.4 mg lost a mean of 14.9 percent of body weight over 68 weeks. The DEXA substudy presented at ENDO 2021 reported total fat mass down 19.3 percent and visceral fat down 27.4 percent from baseline, with lean body mass down 9.7 percent. Because fat fell faster than lean tissue, the lean share of total body mass rose by about 3 percentage points. The trial registry lists the absolute change as about 5.8 kg of lean mass lost on semaglutide versus 1.8 kg on placebo in the small DEXA cohort.
In SURMOUNT-1 (Look and colleagues, Diabetes, Obesity and Metabolism, 2025), 160 participants had DXA scans over 72 weeks. On tirzepatide, body weight fell 21.3 percent, fat mass 33.9 percent and lean mass 10.9 percent. Of the weight lost, 74 percent was fat and 26 percent lean. The placebo group, which lost far less weight, showed almost the same split at 75 to 25. The authors concluded that the ratio was consistent across subgroups and that bigger weight losses simply meant more of both fat and lean mass lost.
That is the key reading. These drugs do not appear to target muscle. They produce large, fast weight loss, and large, fast weight loss always costs some lean tissue unless something is done to protect it.
Why does this matter more for some people?
In their 2024 commentary “Muscle Matters” in The Lancet Diabetes and Endocrinology (volume 12, pages 785 to 787), Carla Prado and Stuart Phillips estimated that fat-free mass loss can account for 25 to 39 percent of total weight lost over 36 to 72 weeks of treatment, and argued that this could worsen sarcopenic obesity in susceptible people (summarised by the University of Alberta, October 2024). A 2024 network meta-analysis in Metabolism found lean mass reductions of roughly 25 percent relative to total weight loss, with the percentage of lean mass relative to body weight essentially unchanged.
Reviews published in 2025 in Pharmacological Research and in a geriatrics-focused analysis raise the same concern for older and frail adults, where muscle is already declining about one percent a year and where strength predicts falls, independence and survival. In India, where many adults carry high visceral fat with relatively low muscle mass, the same caution applies at younger ages than Western cut-offs would suggest.
Two other groups deserve attention: people who stop and restart treatment, because regained weight tends to return as fat, and people using these drugs for aesthetic goals at a normal body weight, where there is little fat to lose and lean tissue takes a larger hit.
How do you protect muscle while losing weight on a GLP-1?
The evidence base is catching up. A six-month prospective study of 200 adults starting semaglutide or tirzepatide with structured resistance training and protein education reported about 13 percent weight loss with roughly 3 percent muscle loss, as reported by Medscape in 2025. It was not randomised and used bioimpedance, so treat it as encouraging rather than definitive. The LEAN-PREP randomised trial (232 adults, targeting 1.6 grams of protein per kilogram per day with or without resistance exercise alongside semaglutide or tirzepatide) is designed to answer the question properly.
Practically: lift something heavy enough that the last few repetitions are hard, cover the major muscle groups, progress the load, and spread protein across meals. Appetite suppression makes protein the first thing to fall, so it usually needs to be planned rather than assumed.
How does MRI measure muscle, and how does it compare with DEXA and smart scales?
A 2018 consensus in the Journal of Cachexia, Sarcopenia and Muscle judged DEXA the best practical reference for muscle mass on feasibility, accuracy, safety and cost, while noting it is not a gold standard. Borga and colleagues (British Journal of Radiology, 2018) showed excellent agreement between DEXA and quantitative MRI for whole-body fat and lean totals in UK Biobank, but that the agreement does not extend to regional or compartment-level detail. A 2023 review describes MRI as comparable to CT for muscle volume and for fat within and between muscles, with Dixon and fat-fraction sequences giving a quantitative map rather than a visual impression.
| Method | What it measures | Strength | Limitation |
|---|---|---|---|
| Bioimpedance (smart scales) | Estimates fat and fat-free mass from electrical resistance. | Cheap, instant, at home. | Hydration swings move the number; no regional or intramuscular detail. Not validated against MRI for muscle volume. |
| DEXA | Whole-body and regional fat, lean and bone mass. | Accepted practical reference for lean mass; fast and low-dose. | Two-dimensional projection; cannot see fat within muscle or separate muscles; agreement with MRI weakens at the regional level. |
| MRI (Dixon or fat-fraction) | Muscle volume per muscle group, intramuscular fat fraction, visceral and liver fat. | Reference standard for muscle volume and muscle quality; no radiation; repeatable. | Higher cost and scan time; needs standardised protocols and trained readers or validated AI segmentation. |
For a person on a GLP-1, the question that matters is not “how much lean mass did I lose” but “did my thigh and trunk muscles shrink, and did fat move into them”. That is a regional, intramuscular question, and it is the one MRI answers.
Which blood markers matter alongside imaging?
Imaging shows structure. Blood shows whether the body has what it needs to maintain it, and whether the medication is being tolerated. A sensible monitoring panel groups into four themes:
- Protein and nutrition status: serum albumin and pre-albumin, vitamin B12, vitamin D and ferritin. Appetite suppression plus reduced intake makes deficiencies common and easy to miss.
- Metabolic response: HbA1c, fasting glucose and fasting insulin (HOMA-IR), plus ApoB, triglycerides and the LDL to HDL ratio, to confirm the weight loss is improving the things it is meant to improve.
- Organ safety: creatinine and eGFR with sodium and potassium, because dehydration and vomiting are common early; serum amylase and lipase as a pancreatic baseline; liver markers alongside liver fat on MRI.
- Anabolic reserve: TSH and testosterone, which influence how readily muscle is maintained or rebuilt.
How does Evo GLP put this together?
Evo GLP is a longitudinal oversight layer for people on GLP-1 medication. It does not prescribe or supply the drug. It tracks whether weight loss is fat-dominant and safe, and flags early when it is not.
- Body mapping (MRI): visceral adipose tissue volume, a myosteatosis index of fat infiltration within skeletal muscle, and hepatobiliary safety through liver fat fraction and gallbladder morphology.
- Metabolic and cardiovascular panel: HbA1c, fasting glucose and insulin (HOMA-IR); ApoB, triglycerides and LDL to HDL ratios.
- Organ and systemic safety: creatinine, eGFR and electrolytes; serum amylase and lipase; TSH and testosterone.
- Nutritional bioavailability: vitamin B12, vitamin D and ferritin; serum albumin and pre-albumin.
The tests are calibrated to your progress through three to five checkpoints per annual cycle, read through 5C Network's imaging and AI platform, and summarised by a physician into actions: adjust protein, change the training emphasis, review the dose with your prescriber, or simply continue. Evo GLP is currently a waitlist programme; see plans and pricing or the FAQ for how it fits alongside Evo Core.
Frequently asked questions
Do GLP-1 drugs cause muscle loss?
They cause weight loss, and a share of any substantial weight loss is lean tissue. In the STEP 1 semaglutide body-composition substudy, lean body mass fell about 9.7 percent from baseline over 68 weeks; in the SURMOUNT-1 tirzepatide substudy, about 26 percent of weight lost was lean mass. The proportion is similar to diet-induced weight loss, so the concern is the amount and speed of loss, not a unique drug effect.
How much of the weight lost on semaglutide is muscle?
Estimates from DEXA substudies and reviews range from roughly 25 to 40 percent of total weight lost being lean mass. Lean mass includes water, organ tissue and bone as well as skeletal muscle, so the muscle-specific share is smaller. The 2024 Lancet Diabetes and Endocrinology commentary by Prado and Phillips put fat-free mass loss at 25 to 39 percent of weight lost over 36 to 72 weeks.
Who is most at risk of losing muscle on GLP-1 medication?
People over 60, those starting with low muscle mass or sarcopenic obesity, people losing weight very quickly, those eating well under 1.2 grams of protein per kilogram per day, and anyone not doing resistance training. Reviews published in 2025 single out older and frail adults as the group where muscle loss could precipitate sarcopenic obesity.
Can you prevent muscle loss on Ozempic, Wegovy or Mounjaro?
You can reduce it. Progressive resistance training two to three times a week and adequate protein, commonly 1.2 to 1.6 grams per kilogram per day during active weight loss, are the two interventions with the best support. A 2025 prospective study of 200 adults given training and protein guidance reported around 13 percent weight loss with only about 3 percent muscle loss, though it was not randomised.
Is MRI better than DEXA for tracking muscle on GLP-1 drugs?
DEXA is the accepted practical reference for whole-body lean mass and is cheaper and faster. MRI is the reference for regional muscle volume and for fat inside muscle, which DEXA cannot see. If you want to know whether your quadriceps shrank or filled with fat, MRI answers that; DEXA tells you total lean mass moved.
How often should muscle and body composition be checked on GLP-1 therapy?
There is no formal guideline yet. A practical approach is a baseline before or soon after starting, a check at the point of fastest weight loss (often three to six months in), and then every six to twelve months while on treatment or during tapering. Evo GLP is built around three to five checkpoints per annual cycle.
Medical disclaimer