Muscle Loss on GLP-1s: The Real Numbers and the Two Fixes

“GLP-1s melt your muscles” is a headline. The trial data is calmer but still worth taking seriously: significant weight loss by any method takes lean tissue with it, GLP-1s included — and the size of the loss depends heavily on two things you control.

The actual numbers

In the STEP 1 DXA substudy (body-composition scans on a subset of semaglutide participants), lean mass accounted for roughly 40% of total weight lost; tirzepatide’s SURMOUNT substudies reported broadly similar proportions (about a quarter to 40%, varying by analysis). Sounds alarming — three qualifiers first:

1. It’s not unique to the drugs. Diet-induced weight loss commonly runs 20–35% lean; bariatric surgery similar or higher. Rapid loss without countermeasures costs lean mass, whatever causes the loss.
2. “Lean mass” isn’t all muscle. The lean compartment includes water and organ tissue; skeletal muscle is only part of the DXA number.
3. Ratios still improve. Trial participants ended less fat relative to lean than they started — body composition moved the right way overall.

The honest concern isn’t the average, it’s the tails: people losing very fast, eating very little protein, doing no resistance work — and especially people over ~50, who rebuild muscle more slowly and are nearer the strength threshold where losses cost function. For them the 40% number is a genuine warning; sarcopenia is much easier to prevent than reverse.

Why it matters even if you feel fine

Muscle is metabolic and functional insurance: it drives resting energy burn (losing it makes maintenance harder and rebound easier — regain, unlike loss, comes back mostly as fat), and it’s what stairs, groceries, and independence at 75 are made of. “Lighter but weaker” is a bad trade available to anyone who loses 20% of body weight passively.

Fix one: protein, treated as a floor

During active loss, evidence-based guidance clusters around 1.2–1.6 g of protein per kg per day — substantially above habit for most people, and genuinely hard under appetite suppression. The mechanics of hitting it on a shrunken appetite — protein-first ordering, protein-dense small forms, front-loading the day — are covered in the eating guide. The one-sentence version: spend your scarce appetite on protein first, every meal, and measure the gap once with a week of honest logging instead of assuming.

Fix two: resistance training, minimum effective dose

Muscle is kept by use. During an energy deficit the body decides what tissue is worth feeding, and resistance work is the signal that muscle earns its keep — meta-analyses across weight-loss contexts consistently show resistance training preserving lean mass that diet-only loss surrenders.

The bar is lower than gym culture suggests:

• 2–3 sessions a week, 30–40 minutes.
• Compound basics: squats or sit-to-stands, pushes (wall to floor), rows or pulls, hinges, carries. Bodyweight and resistance bands absolutely count; progressive weights are better over time.
• Progression beats intensity: slightly more reps or load than last month is the entire game.
• Cardio is great for health but is not this signal — walking doesn’t tell your pressing muscles they’re needed.

If you do only one new thing during treatment, make it this. Protein without training lacks the stimulus; training without protein lacks the material. Together they’re most of the answer.

Give muscle a scoreboard

The scale can’t see composition — it reports muscle kept and fat kept identically. If you’re doing the work, track something that shows it:

• Strength benchmarks: push-up count, comfortable carry weight, sit-to-stand reps. Flat-or-rising strength during major weight loss ≈ muscle holding.
• Measurements and photos: waist shrinking at stable weight is recomposition the scale calls “a plateau.”
• Pace check: sustained loss much faster than ~1% of body weight per week raises the lean-loss share — worth a pacing conversation with your prescriber (see titration).

Logging weight, meals, and movement in one place (this is Glu’s home turf) is what makes the pattern visible: protein average, loss rate, and strength trend on the same timeline answer “am I losing the right kind of weight?” better than any single number.

The appointment version

If you’re over 50, losing fast, or already noticing strength slipping: ask your prescriber or a dietitian about a protein target for your case, whether your loss rate warrants slower titration, and — if available — a DXA or bioimpedance baseline so future scans measure change instead of guessing. Five minutes of appointment time; sharper decisions for years.

Sources: STEP 1 (Wilding et al., NEJM 2021, incl. DXA substudy); SURMOUNT-1 (Jastreboff et al., NEJM 2022); Leidy et al. on protein in weight loss (AJCN 2015); Sardeli et al., resistance training preserving lean mass during caloric restriction (Nutrients 2018).