Muscle Loss on a GLP-1: Why It Matters for Women, and How to Protect It

The short version

When the scale drops fast on a GLP-1, not all of what leaves is fat. A meaningful share is skeletal muscle — the tissue that lets you carry a toddler, get up off the floor, and stay strong into your later decades. Across randomized trials, lean mass makes up roughly a quarter to a third of the weight lost on these drugs¹. That is not a reason to skip treatment; the same literature shows muscle is largely defensible if you plan for it from day one. But it matters more for women over 40, because that is exactly when muscle is already slipping for reasons that have nothing to do with the drug.

This page is about muscle: function, how much you lose, and the two levers with randomized evidence behind them. The skeleton has its own page — bone density, fracture data, what a DXA actually measured — at bone density and lean mass on GLP-1s.

Muscle is not just a number on a body scan

It is easy to treat "lean mass" as an abstraction. It is not. Skeletal muscle generates strength and power, stabilizes your joints, and does a large share of your day-to-day glucose disposal. It is metabolically active tissue, so losing it can quietly lower what you burn at rest. Lose too much and you feel it as weakness and worse stamina, and over years it becomes a higher risk of frailty and falls. The goal of weight loss for a mother is not just a smaller body; it is a more capable one. That reframes muscle from a side note into something you actively defend.

How much you actually lose — and the trials were mostly women

A 2026 systematic review and meta-analysis of 20 randomized trials (15,782 participants, body composition measured by DXA or MRI) found that lean mass constituted 25–39% of total weight lost with incretin drugs: 35.2% with semaglutide, 25.4% with tirzepatide, 26.8% with liraglutide. The number that reframes the whole conversation sits alongside them — lifestyle-only weight loss showed a statistically indistinguishable 26.2% (p = 0.42)¹. This is mostly what losing weight quickly does, not a toxic effect unique to the molecule.

For the absolute scale of it, the SURMOUNT-1 DXA substudy — 160 people, 73% of them women — measured lean mass falling 10.9% over 72 weeks on tirzepatide, against 2.6% on placebo². Note those are two different kinds of number and are routinely swapped: about 25% is the share of the weight lost that was lean; about 11% is the change in lean mass itself. Because that trial population was overwhelmingly women, both describe you rather than a male average.

Why this lands harder for women over 40

Here is the part that generic "muscle loss on Ozempic" articles miss. Women are already losing muscle at the life stage when many start a GLP-1. A 2026 narrative review of the human data found reductions of about 2.5% in perimenopausal and 5.7% in postmenopausal women compared with premenopausal women, tied to falling estrogen — and noted that most of that evidence comes from DXA lean body mass, which may underestimate the change in muscle specifically⁷. The same review found that muscle protein synthesis responses to resistance exercise and to protein feeding may be blunted in older women⁷.

Layer rapid GLP-1 weight loss on top of an already-declining baseline and the arithmetic is obvious: the tissue you can least afford to lose is under the most pressure. It is the same reason our guide to GLP-1s through perimenopause and menopause treats strength as a first-order goal rather than a bonus.

Lever one: lifting, and the numbers behind it

The best-evidenced defense is resistance training. A 2026 systematic review with pairwise and network meta-analysis pooled 34 randomized trials (1,455 participants) comparing calorie restriction alone against calorie restriction plus exercise³:

Read as a whole, exercise prevented nearly half — 45.7% — of the fat-free-mass loss that dieting alone produced³. Read row by row, be careful: subgroup testing found no significant difference between training modes, so the ranking above is a set of point estimates, not a proven hierarchy. What is solid is the top line — moving while you diet keeps roughly half a kilogram to a kilogram of tissue you would otherwise have lost.

When a GLP-1 is doing the weight loss, the logic carries. In a Danish randomized trial of exercise, liraglutide, or both after an initial low-calorie diet, the combination cut body-fat percentage by 3.9 percentage points — about twice the drop seen with exercise alone or liraglutide alone — and it was the only strategy that also improved insulin sensitivity and cardiorespiratory fitness⁵. (That same trial's bone analysis is on the companion page.)

One nuance that should change how you judge progress: in a calorie deficit you may not build much new muscle, but you still get stronger. A meta-analysis of resistance-training trials found lean-mass gains impaired under an energy deficit (effect size −0.57, p = 0.02) while strength gains were statistically comparable to training without a deficit (−0.31, p = 0.28); its meta-regression put the tipping point at roughly 500 kcal/day, beyond which lean-mass gains stopped⁴. So do not wait until "after" to start lifting, and do not judge it by the scale — judge it by whether the weights go up.

Lever two: protein, even when the drug kills your appetite

The catch with a GLP-1 is that it works by cutting appetite, which is exactly what makes protein hard to hit. A review of the protein literature concluded that intakes between 1.2 and 1.6 g/kg/day, with roughly 25–30 g of protein per meal, improved appetite control and body-weight management, and that higher-protein energy-restricted diets produced greater preservation of lean mass than lower-protein ones in tightly controlled feeding trials — with dietary adherence the main reason longer studies disagree⁶. That is well above the 0.8 g/kg minimum most people are anchored to.

In practice: treat protein as the first thing on the plate rather than whatever fits after a few bites, and lean on easy sources — Greek yogurt, eggs, fish, a shake — on the days appetite is genuinely small.

How strong is the evidence, honestly

Graded piece by piece: that a quarter to a third of GLP-1 weight loss is lean mass is high-certainty — 20 randomized trials with DXA or MRI, and the same proportion appears with lifestyle-only weight loss, which is the finding that de-dramatizes it. That exercise preserves fat-free mass during a deficit is high — 34 randomized trials, a consistent effect, a clear pooled estimate. That strength training specifically beats other modes is low — the point estimates lean that way, but the subgroup test was not significant. That the menopause transition costs muscle is moderate at best — a narrative review of mostly cross-sectional DXA data, with the authors themselves flagging measurement limits. And nobody has yet run the trial that would answer this properly: a resistance-training program randomized on top of semaglutide or tirzepatide at obesity doses, in perimenopausal women, with muscle function — not just mass — as the endpoint.