Lifting on Ozempic, Wegovy or Zepbound: How to Keep Your Muscle

Collage of a woman in her forties doing a dumbbell goblet squat, with the headline "Keep Muscle on GLP-1"

Yes, lifting can help you hold on to muscle while you lose weight on semaglutide (Ozempic, Wegovy) or tirzepatide (Zepbound, Mounjaro), although the trials built to measure how much are still under way. In trials that measured body composition, roughly a quarter to two-fifths of the weight lost was lean mass, a share similar to weight loss by diet alone, so the practical plan is simple: lift two to three times a week, add load slowly, and eat enough protein even when you are not hungry.

This is general education, not medical advice. Weight-loss medication, your other health conditions and your other medicines are decisions for your prescriber, and a dietitian or physiotherapist can tailor food and training to you. Nothing here covers drug doses, or starting or stopping a medication.

How much of the weight you lose is muscle?

Two DXA substudies give the clearest numbers. In the STEP 1 subgroup (140 people on semaglutide 2.4 mg or placebo, scanned at week 68), the semaglutide group lost 15.0% of body weight, 19.3% of fat mass and 9.7% of lean body mass (Wilding et al., conference abstract, Journal of the Endocrine Society, 2021). A 2025 joint advisory (covered below), citing STEP 1, reports that 5.3 kg (38%) of an average 13.6 kg loss was lean body mass.

In the SURMOUNT-1 DXA substudy (Look et al., Diabetes, Obesity and Metabolism, 2025; 160 of the trial's 2,539 participants, mostly women, scanned at week 72), tirzepatide users lost 21.3% of body weight, 33.9% of fat mass and 10.9% of lean mass. About 75% of the weight lost was fat and 25% was lean mass. The placebo group, which lost far less weight, showed the same 75/25 split.

A 2026 meta-analysis of 20 randomized trials with 15,782 participants (Eisa and Barood, Diabetes, Obesity and Metabolism) pooled the picture. Lean mass made up 35.2% of weight lost on semaglutide, 25.4% on tirzepatide and 26.8% on liraglutide, against 26.2% for lifestyle-only weight loss. The difference between drugs and lifestyle was not statistically significant, and the trials disagreed with each other moderately (I-squared of 68%). A separate 2026 meta-analysis of seven placebo-controlled trials (Laverde et al., International Journal of Obesity) found that lean mass rose as a share of body weight but fell in absolute terms, with a bigger absolute loss on semaglutide, and its results varied widely between studies.

The takeaway is not that the drugs are harmless to muscle. Large weight loss takes some lean tissue with it by most methods studied, and these drugs make large weight loss common. That is why the plan below matters more on a GLP-1 than on a slow diet.

Lean mass is not the same as muscle

DXA reports lean mass, which includes muscle but also other non-fat tissue such as water and organs. The joint advisory estimates that the STEP 1 lean-mass figure corresponds to roughly 20% of the weight lost being muscle itself, about half the headline number. Home scales and many clinic devices use bioelectrical impedance instead. The advisory describes it as practical and low cost, and says that all of these methods monitor muscle mass but not muscle health, quality or function.

So treat any single percentage with care. A useful second measure is what you can do: how much you can squat, press and row, and whether stairs or carrying shopping feel harder. That is another reason to keep a log, covered below.

What the evidence says about lifting and protein on GLP-1 therapy

The honest answer is that it points one way but is thin. A PubMed search for this post (October 2026) found no completed, peer-reviewed randomized trial that tests lifting during semaglutide or tirzepatide with a direct muscle measurement, though small or unpublished studies could exist. The best-known trial is a protocol: LEAN-PREP (Alawadhi et al., BMJ Open, 2026; NCT06885736) plans to randomize 232 adults to control, home-based resistance exercise three times a week, protein at 1.6 g/kg a day, or both, over six months, with MRI-measured quadriceps size as the main outcome. The registry lists it as recruiting and there are no results yet.

What exists is weaker evidence:

  • A six-month study of 200 adults prescribed semaglutide or tirzepatide, presented at the European Congress on Obesity in 2025, reported about 12% to 13% weight loss with roughly 0.6 kg (women) and 1 kg (men) of lean mass lost by impedance. The participants were counselled on lifting and protein, but the study had no comparison group, used impedance, and was preliminary when presented.
  • A 2026 retrospective chart review (Edens et al., Obesity Science and Practice) of 171 patients who stayed in a dietitian-plus-progressive-exercise program for 12 months found skeletal muscle made up about 12% of weight lost. Only adherent patients were analyzed, and there was no control group.
  • A 2026 non-randomized trial (Stanton et al., Diabetes, Obesity and Metabolism) compared 151 people in a virtual program with strength-focused coaching against 94 on usual care for 12 weeks. The coached group lost 6.0% versus 3.3% of weight and showed a slightly larger rise in muscle percentage by home impedance scales (+0.6% versus +0.2%).

All three point in the same direction, but none was randomized, people who join programs may be more motivated, and two of the three relied on impedance rather than DXA or MRI. For context, in the Eisa and Barood meta-analysis lifestyle programs that included resistance training had a lean share of 17.5%, versus 26.2% without. Those were not drug trials, so they show what lifting does during weight loss in general, not what it adds to a drug.

Guidelines are firmer than the data. The 2025 joint advisory from the American College of Lifestyle Medicine, the American Society for Nutrition, the Obesity Medicine Association and The Obesity Society (Mozaffarian et al., Obesity) recommends strength training at least three times a week plus at least 150 minutes of moderate aerobic activity weekly to protect muscle and bone. It adds that structured strength training, or strength plus aerobic training, is well established for preserving lean mass during weight reduction, and that increased protein alone is likely inadequate to preserve muscle without it. A 2024 review in Diabetes Care (Locatelli et al.) argues the same and notes that supervised resistance training lasting more than 10 weeks has produced gains of around 3 kg of lean mass and about 25% strength in men and women, in studies that were not specific to these drugs.

A weekly plan to start from

The anchor for general muscle building in healthy adults is the 2026 ACSM position stand (Currier et al., Medicine and Science in Sports and Exercise): train all major muscle groups at least twice a week, with roughly 10 sets per muscle per week for growth, and failure is not strictly necessary. Our breakdown of the ACSM guidelines goes through it. That stand covers healthy adults, not people on these drugs, so treat it as something to build toward, not a week-one target. When you are eating much less, you may recover more slowly, so starting lower is sensible.

  • Frequency: two or three full-body sessions on non-consecutive days. Three matches the advisory and the LEAN-PREP design; two is a fine start.
  • Exercises: five or six big movements per session covering legs, push, pull and hips.
  • Sets and reps: start with two sets of 8 to 12 per exercise and move to three over several weeks as recovery allows. This is a practical convention, not a tested drug-specific dose.
  • Effort: finish each set with about two or three reps left in the tank. No need to grind to failure.
  • Cardio: add brisk walking or similar toward the 150 minutes the advisory suggests, after the lifting is in place.

Two sample sessions to alternate

  • Session A: goblet squat, dumbbell bench press, seated cable row, Romanian deadlift, plank.
  • Session B: leg press, dumbbell shoulder press, one-arm dumbbell row, dumbbell lunges, farmer's walk.

If you are new to the weight room, the beginner strength routine and the week-by-week starter plan cover technique and setup.

How to progress

Use double progression. Pick a rep range such as 8 to 12. When you reach 12 reps on every set with two or three reps still in reserve, add the smallest weight jump available and drop back to 8. As your body weight falls, push-ups, pull-ups and lunges get easier, so add reps or load there too. Strength and muscle size are related but not the same thing, so a steady log is encouraging but not proof.

A rough rule: if your logged weights and reps fall across several exercises for a few weeks in a row, raise it with your care team. It is a practical flag, not a validated threshold.

Protein: the numbers, and what to do when you are not hungry

The joint advisory lists the standard recommended allowance as 0.8 g/kg a day, notes that 1.2 to 1.6 g/kg a day has been proposed during active weight loss, and says protein intake below about 0.4 to 0.5 g/kg risks muscle atrophy while prolonged intake at or above 2 g/kg should be avoided. It also floats an absolute target of 80 to 120 g a day for simplicity, and 1.5 g per kg of fat-free mass if you know yours. It admits that the best way to calculate this at higher body weights is unclear.

A worked example: at 90 kg, 1.2 to 1.6 g/kg is 108 to 144 g a day, which overlaps the top of the simple 80 to 120 g range. Ask a dietitian whether to base the figure on current or goal weight. If you have kidney disease, protein targets are a clinician decision, not a rule of thumb. Our protein guide covers the general muscle-gain evidence.

When appetite is low, the advisory suggests:

  • Eat small, frequent meals and set reminders to eat.
  • Eat the protein food first, before the rest of the plate.
  • Choose compact protein foods such as fish, eggs, Greek yogurt, cottage cheese and nut or seed butters.
  • Use smoothies or protein drinks when solid food is unappealing.

Training around side effects

The advisory notes nausea often shows up in the morning or after long gaps without food, so schedule lifting when you feel best and eat something small beforehand. It also warns that dehydration from severe nausea, vomiting or diarrhea can cause acute kidney injury; the Wegovy prescribing information (revised 06/2026) lists kidney injury from volume depletion as a warning too. Drink fluids through the day, and skip a session on days you cannot keep fluids down.

Low blood sugar can occur in people with type 2 diabetes who also take insulin or sulfonylureas, according to both documents. If that is you, ask your prescriber how to handle training days.

Stop and get medical help for chest pain, fainting, a racing or irregular heartbeat, severe stomach pain that does not ease (especially with vomiting), vomiting you cannot control, very dark urine or little urine, or sudden weakness. The advisory also links large, fast weight loss (14% or more over three to four months) to bone loss and cites one randomized trial where a year of GLP-1 therapy plus exercise training preserved bone density, and it recommends its strength-plus-aerobic plan to protect bone as well as muscle.

Who this does not apply to, and common mistakes

These ideas are for generally healthy adults cleared to exercise. If you are frail, much older, recovering from surgery, pregnant, have heart, kidney or joint conditions, or have a history of disordered eating, a clinician-led or supervised program is the safer route. A supervised physiotherapy or exercise-physiology program is also a good choice if you are new to lifting and unsure of your form.

  • Cardio only. Walking helps, but the advisory notes that aerobic activity alone has a smaller effect on preserving lean mass.
  • Waiting. The advisory links rapid weight loss to muscle and bone loss, so once you are cleared to exercise, start lifting in the first weeks, not at month six.
  • Skipping protein because food feels unappealing. Plan it as the first item in each meal.
  • Judging progress by the scale. Watch strength and how clothes fit too.
  • Doing too much at once. Three sets of everything on a very low intake can leave you wiped out.
  • Testing a one-rep max. Use rep-based progress instead.

Keeping a log that shows what is happening

On any weight-loss program, the training log becomes your strength check-in between scans. Easy Reps is a free lifting log that records every set with its weight and reps, keeps a filterable history with charts and a calendar, and updates your estimated one-rep max as you go, so a flat or falling trend on your main lifts is easy to spot. It does not track food, protein or body measurements, so keep those in whatever else you use. The post on stopping a GLP-1 and the post on the new GLP-1 pill cover those separate questions.

Bottom line

Expect some lean mass to go when weight drops quickly, and do not assume the drug is uniquely to blame. Lift two or three times a week, progress gradually, aim for the protein range your dietitian or prescriber agrees with, and watch your logged strength as an early warning. The evidence for lifting on these drugs is promising but unfinished, and your prescriber should make the medical calls.

Sources