Do GLP-1 drugs make you lose muscle?

Transcript

Welcome to the sofidex Health Drive podcast series. Today we are talking about muscle, and about a worry that has attached itself to the weight-loss injections almost from the moment they became popular. If you are taking one of these drugs, you have almost certainly seen the headlines. You might have looked in the mirror and thought your arms had changed. You might have noticed the stairs feel different, and not been sure whether that is the weight coming off or something being lost that you wanted to keep.

And the answers on offer tend to come in two unhelpful sizes. Either it is a scare story and you should ignore it, or you are wasting away and should stop the drug. Neither of those is a real answer, and both leave you doing nothing. So we want to ask a better question. Not whether you lose muscle, because you do. The question is how much, compared to what, and whether any of it is actually avoidable.

We will go through what the body composition studies actually measured, and why about a quarter of the weight lost being lean tissue is both true and less alarming than it sounds. We will look at the comparison nobody puts in the headline, which is what happens to your muscle when you lose the same weight by eating less. And we will get to the two things that genuinely change the number, both of which are things you can start this week.

One thing before we start. Nothing in this episode is a reason to stop a medication you were prescribed, and none of it is a verdict on whether you should be taking it. One thing before we start. This is health information, not medical advice, and nothing in it is a substitute for talking to your own doctor about your own treatment. What we can tell you is where all of it came from — 5 published sources, all listed with the episode. I want to do this one through a single person, if that's alright. Not a real patient — someone we're making up, so we can follow her all the way through. Call her Anne. She's sixty-eight. She's been on one of these drugs for seven months. She's lost a bit over two stone. And last week she saw a photograph of herself from last summer and thought: I look older. And then she did what everybody does, which is go and look it up. And what she found was a sentence saying a quarter of the weight you lose on these is muscle.

And that's the sentence I want to spend this episode on. Because Anne is now frightened by a number that isn't right. And she's frightened in a way that could genuinely make her stop a medication that's doing her good. So it's worth going slowly. Where that number comes from. What it actually says. Whether it applies to her at all. And at the end, what she ought to do on Monday morning.

Start with the photograph though. Because that's what set it off, and I think that's true of most people. It's not a study that frightens anybody. It's a picture.

The photograph is the part I'd deal with first, because it's the most misleading thing in the whole story. Anne is looking at her face. And your face doesn't carry much muscle to begin with. What it carries is fat pads, and those pads are doing structural work you never think about until they're gone. Take them away and the shape changes fast. And it changes in a way we're all trained to read as age — because that's exactly what happens to those pads as people get older anyway. So Anne is seeing something real. She isn't imagining it. But what she's lost from her face is fat. Not muscle. And the thing that frightened her into searching has almost nothing to do with the thing she then searched for.

So the visible change and the thing she's worried about aren't the same thing at all.

They're not even in the same part of the body. The change she can see is in her face. The thing the studies are about is in her thighs, her back, her shoulders. And she can't see any of that from outside. Which is unlucky, because it means the most visible change is driving the fear about the least visible tissue.

Alright. Then let's take the number she found, because that's the thing actually doing the damage. Where does a quarter come from?

It's real, and it's from a good source. There's a network meta-analysis. Metabolism, last year. It pulls together the body composition data across these drugs. And network meta-analysis means it doesn't just pool trials that asked the same question. It compares drugs against each other indirectly. Through what each one was tested against. And the figure that comes out of it: about a quarter of the weight lost is fat-free mass. So the number Anne found is not made up. It's been reported accurately. And it still isn't what she thinks it is.

Why not?

Because fat-free mass doesn't mean muscle. It means everything in your body that isn't fat. So that's muscle, yes. But it's also water. It's also glycogen — your stored carbohydrate. It's connective tissue. It's bone mineral. It's part of your organ mass. It's a category that exists because a scan can separate it from fat, and honestly that's the only thing those tissues have in common. So when Anne reads that a quarter of her weight loss was muscle? Somebody has taken one word out of a technical term. And quietly swapped in a different meaning. A quarter is fat-free mass. Skeletal muscle is about half of that — the Acta Diabetologica review puts it in almost exactly those terms.

So what's Anne's actual number?

Call it an eighth. Of the weight she's lost, something in the region of one part in eight is skeletal muscle. And I want to be careful here. The range across studies is wide. I don't want to hand her a new over-confident number to replace the old one. But an eighth is a very different sentence from a quarter. And there's something else sitting inside it that makes even the eighth look worse than it really is.

Go on.

Water. That's the rest of it. When you cut carbohydrate, you burn through your stored glycogen in the first week or two. And everyone on these drugs cuts carbohydrate. Because they're eating less of everything. And glycogen is stored wet. Roughly three grams of water tag along with every gram of it. So the first few pounds anybody loses, on any diet ever invented, are substantially water. That water counts as fat-free mass. It shows up on the scan as fat-free mass lost. It was gone within a fortnight of Anne starting, it was never muscle, and it isn't coming back because she doesn't need it any more.

Which means the shorter the study, the worse these drugs look.

Considerably worse. So if you ever see one of these figures quoted without a duration attached to it, you haven't been given enough to judge it by.

Can I put something to you that I only half understand? I read that the proportion of lean mass can actually go up while you're losing weight. Which sounds like it can't possibly be true.

It's true. And it's worth doing slowly, because it changes how Anne should read any number she's given. Say she's lost ten kilos. And say seven and a half of that was fat, two and a half fat-free. So she's lost lean tissue. Definitely. Nobody's disputing that. But she's lost proportionally more fat. Which means the fraction of Anne that is lean tissue is now higher than it was in January. On paper and in fact, she is a higher percentage muscle than when she started.

While the actual amount went down.

Both true. At once. And it's arithmetic rather than a finding — if more of what leaves you is fat than is lean, the share of you that's lean goes up while the amount of it goes down. The meta-analysis measured that directly, and what it reports is that the relative lean mass was unaffected. It also reports that semaglutide was among the least effective of these drugs at preserving lean mass. So the reassuring sentence is true about the arithmetic and not a thing the trials went out and found. And which of those two you lead with decides entirely whether this story is frightening or reassuring. Headlines lead with the kilograms. Every time. Because "loses muscle" is a story and "becomes proportionally more muscular" isn't. Neither one is a lie. Which is exactly what makes it so hard to argue with.

And is some of what she's lost actually meant to go? That's the bit I never see anywhere.

That's the part that matters physically and it hardly ever gets said. Muscle exists to move mass. That's all it's for. So if Anne is moving thirty pounds less, she needs less of it. Think about the leg muscle of somebody heavier. It's doing more work every day. Just to exist. Standing up is a squat with a load on. So when the load comes off, the tissue adjusts. You'd expect that. Of course you would. You'd expect it with weight loss by any route whatsoever. So not all of Anne's lean mass loss is a problem. Some of it is her body correctly downsizing something it no longer has to carry.

Okay. So the number's smaller than Anne thinks. But it isn't zero. So the next thing she'd want to know is whether this is the drug's fault. Whether she'd have avoided it by doing it the hard way.

And the answer is no, and it's about as clean an answer as this field ever gives. There's a paper whose title is more or less the whole argument: GLP-1 receptor agonists induce loss of lean mass, so does caloric restriction. Lose weight by any route at all and some of what leaves you is lean tissue. It's true if you simply eat less. It's true after bariatric surgery. It's true, much more sharply, if you lose weight because you're ill. There is no good evidence that these drugs do something distinctive to muscle that dieting doesn't do. So Anne didn't cause this by choosing an injection. She caused it by losing two stone, which is the thing she was trying to do.

Then why is it a story about the drugs? Nobody wrote frightening articles about muscle loss on the 5:2.

Partly arithmetic. The proportion stays roughly constant while the total goes up. So somebody losing thirty kilos loses far more kilos of lean tissue than somebody losing eight. Same fraction. Bigger number. Nothing different about their physiology at all. And the number of kilos is what gets printed, because it's the more frightening of two ways of saying an identical thing. And partly it's that a drug is something people can be for or against in a way that a diet isn't.

Right. So this is the point where I'd expect a lot of shows to stop. The number's wrong, the coverage is overblown, everybody relax. But you told me before we started that you didn't want to end up there.

No, because that would be a different kind of dishonesty. "The coverage is overblown" and "there's nothing here" are not the same claim, and only the first one is true. There is something here. It just isn't a fact about the drug. It's a fact about the person taking it. And that's the actual turn in this story — the question was never whether people lose lean tissue. It's who it matters to. And Anne is sixty-eight, so it's a fair question whether she's one of them.

So let's ask it properly. Is Anne in the group that should be paying attention?

Possibly. And age is the reason. At sixty-eight she is already losing muscle every year without doing anything at all. That's sarcopenia — not a disease you catch, but a slope everybody's on from somewhere in their thirties. So anything that takes a bit more off the top is a considerably bigger deal for her than for somebody who's forty. Same percentage of weight lost, same fraction of it lean, entirely different consequence, because there's less held in reserve. The second thing I'd want to know about Anne is whether she was well nourished before she started. Because you can carry a great deal of fat and still be badly nourished. Those two things aren't opposites. And the third is speed. The Acta Diabetologica review is clear on this. In rapid or marked weight reduction, the decline in lean mass may account for up to twenty-five to thirty per cent of the total. So the fraction itself gets worse at the fast end.

Two stone in seven months. Is that fast?

That's an ordinary pace, and I'd be much more worried if she'd told me she'd done it in three. So on that one, Anne is fine. Which leaves age, which she can't change, and nutrition, which she can.

Here's the thing I'd actually want if I were her though. She's sixty-eight, she's had a fright, and she wants to know whether it's happened to her specifically. Not to a study population. To her. How does she find out?

Not with a scan. Definitely not. And that's worth saying loudly, because a scan is exactly what she'll be tempted to go and buy.

Why not? That's the obvious move.

Because the number wouldn't change what anybody does next. Think it through. If Anne's lean mass is down, the advice is protein and resistance training. If Anne's lean mass is up, the advice is protein and resistance training. The scan doesn't change the plan. So what does it actually buy her? A figure to worry about between appointments, and a reason to go and get another one. And UK guidance agrees, incidentally. That's NICE. The overweight and obesity guideline, updated at the start of this year. It doesn't recommend routine body composition monitoring for anyone on weight management medication. That isn't an oversight. It's the same reasoning.

Then what does she do instead?

She checks the things that actually matter, and all of them are free and none of them need an appointment. Can she get out of a chair without pushing off with her hands. Can she carry the shopping in from the car in the same number of trips she could in the spring. Are the stairs the same, better, or worse. And that last one is the one I'd have her pay real attention to — because the expected direction is better. She's two stone lighter. The stairs ought to be easier. That's rather the point of the whole exercise. So if Anne's stairs are getting harder while she's losing weight, something is going the wrong way, and that's a genuine signal worth taking to her GP. It's a better signal than any scan, because it's continuous, it's free, and it measures the thing she actually cares about instead of a stand-in for it.

And that chair test — that's a real thing? It sounds too simple.

It's a real clinical test. Used precisely because it's simple. Standing up from a low chair without your hands. And here's the part I like: the movement that tells you whether you're in trouble is the same movement that gets you out of it. Sit to stand from a chair is a leg exercise.

Which takes us to what she can actually do. And you've been holding this back the whole episode.

Two things, and one of them is better established than the other. The first is protein — and everybody says protein, and the reason it matters here is not the reason people assume. A review in Advances in Therapy this year looked at what actually happens to protein intake on these drugs. And the proportion holds up: protein stays at roughly the same share of what you eat. The problem is that the share is of a smaller plate. Eat considerably less in total and the absolute grams fall, even though nothing about your diet looks worse. And it's the grams that build muscle, not the percentage.

Why? If everybody's saying it, why is it still the thing that slips?

Because it isn't an information problem. It's a logistics problem, and Anne is a good illustration of why. These drugs work by making you not want food. That's the mechanism — it isn't a side effect, it's the drug doing its job. And protein is the most filling thing on the plate. So the food Anne needs more of is exactly the food that's hardest to get down. When she's already full. A third of the way through dinner. Telling her to eat more protein, once, at the start, does almost nothing about that.

So what would actually work for her?

Front-load it. First in the day, and first on the plate. Whatever appetite she's got, spend it on protein. Rather than saving it for the part of the meal she's looking forward to. Which isn't a fun instruction. I know that. And if solid protein won't go down, liquid will. The shakes aren't magic. They're not a substitute for food either. But they solve exactly one problem. And given what we just said about appetite, that happens to be the problem worth solving.

And the second thing.

Resistance training. Weights, bands, her own bodyweight — the label matters far less than people think. And it's the stronger of the two, because muscle responds to demand. That's the whole of it. Tissue you use, you keep. Tissue you don't use, your body is quite ruthless about, because it's expensive to maintain, and if nothing is asking for it, it goes. Which reframes Anne's entire situation. The drug isn't taking her muscle. The drug is reducing her total. What decides how much of that comes out of muscle is, substantially, whether anything is asking for the muscle. So this isn't being done to her. There's a lever, and it's been there the whole time.

How much though? Because "do resistance training" is exactly the advice people nod at and ignore.

It gets ignored because it sounds like a lifestyle, and it isn't one. Twice a week. That's the number that keeps coming up. And it isn't very much. Twice a week, and not for an hour. The evidence for the big benefits doesn't require Anne to become somebody who goes to a gym. What it requires is loading the big movements — legs, back, chest — and something that gets hard by the end. Effort, rather than duration. And she can do all of it in her front room. Sit to stand from a chair. Push-ups against the worktop if the floor's too far away. Carrying the shopping in one trip instead of two, which is a legitimate exercise that everybody already does by accident.

What if she's already lost more than she should have? She's seven months in. She didn't do any of this.

Then the advice doesn't change. The timeline does. Rebuilding is slower than keeping, and that gets more true past sixty. But it isn't remotely impossible — muscle responds at every age that's been studied, it just responds slower and needs more protein to do it. And I'd want Anne to hear that specifically, because the feeling that she's missed the window is the thing most likely to stop her starting. She hasn't missed it. Nowhere near.

There's one more thing that I think does more work in this than any study does. The comments. Anne will be getting them.

The you've-lost-weight-but-are-you-okay comments. Everybody gets them. Nobody warns you about them. And they land hard, because you've done a difficult thing and the feedback arrives as concern. And concern reads as disapproval. Even when it genuinely isn't meant that way.

But that's not really evidence about anything, is it.

No. And that's the part I'd flag for Anne specifically. Somebody says you look gaunt. And it gets filed away in her head as data about her muscle. When what it actually was is a remark about her face. Made by somebody who last saw her in March. That's all it was. And we said at the start what her face has actually lost, and it wasn't muscle. So the loop closes badly. The most visible change drives the fear about the least visible tissue. And it sends her to a search engine instead of to her GP.

Is there anything useful to say back to them?

I don't have a clever answer. Sorry. But knowing what they're really about helps a bit. Somebody is describing her face. She's hearing a verdict on her legs.

So bring it back to her. It's Monday morning. What does she actually do?

She stops looking for a scan. First thing. She checks whether the stairs feel worse than they did in spring, and if they do, that goes to her GP. She eats protein first at every meal for a fortnight. And sees whether that's even achievable. Because for some people on these drugs it genuinely isn't. Which is worth knowing. And she gets out of a chair without her hands. Ten times. Twice a week. And she stops calling that exercise. It's the thing that keeps her independent at eighty.

And the thing she was frightened of? The photograph?

Was mostly her face. And mostly fat. And mostly not the thing she went looking for. What she found instead was a real question that happens to have a good answer. Which isn't the usual way these searches end.

Before we finish — what don't we know? Because you've been careful about that line and I want you to draw it.

The biggest gap is time. Most of the composition work runs a year, maybe two, and Anne could be on this for a decade. So what ten years of this looks like in somebody's seventies isn't something anybody can tell you from data, because it hasn't happened yet. The second is what happens when people stop. Weight comes back — that much is well established. But what the regained weight is made of? That's genuinely open. And if fat came back and muscle didn't, somebody cycling on and off could end up worse than when they started. That's a mechanism, not a finding, and I'd want it said in that order. And the third is function. We have a great deal of body composition data. And much less on whether any of it changes who can get out of a chair in five years. Which is the outcome that actually matters, and it's the slowest one to measure.

If Anne takes one thing from this, what is it?

That the thing frightening her is real. It's smaller than she was told. And it's mostly just what happens to anybody who loses weight, by any route at all. But there's a lever. It takes about two sessions a week. And nobody's going to pull it for her.

And three things to take to her doctor.

First. Should my weight loss be checked with anything other than the scale? And the honest expectation is that the answer will often be no. Which is a reasonable answer. Not a brush-off. Second. How much protein should I be eating, in grams, for my body? Ask for a number. Not a principle. Because a principle is what she already has, and it demonstrably isn't working. And third. Is resistance exercise something I should start now rather than later? And that framing is deliberate, because it's considerably easier to keep muscle than to rebuild it, and that gets more true every year. Which is the most useful sentence in this episode. And also the least interesting one. Which is usually how it goes.