Why more muscle alone will not fix metabolic health

In 1989, in Denmark, six healthy young men did an exercise that worked one leg and left the other at rest. Four hours later researchers put catheters in both femoral veins, infused insulin, and measured how much glucose each leg wanted. The legs disagreed. The one that had just exercised took up more glucose than the one that had not, in the same body, with the same blood, the same hormones, and the same muscle mass. The only difference was whether the tissue had recently contracted. That result is a problem for a popular claim: that obesity is a disease of too little muscle and that the fix is to build more.

The one leg experiment

Muscle centric medicine, as argued by Dr. Gabrielle Lyon, says skeletal muscle is the first organ to fail and that you then become obese. Part of that is right. Muscle is the largest site of insulin driven glucose disposal, and muscle insulin resistance shows up years before blood sugar looks wrong. But the sequence is wrong, and so is the prescription. Ten men with type 2 diabetes trained one leg for six weeks and left the other alone. The trained leg cleared more glucose, and both legs had exactly the same amount of muscle. What the trained leg had more of was glucose transporters, insulin receptors, and the signaling proteins in between.

Muscle takes up glucose through doors, not size

Glucose cannot enter a cell on its own. It needs a transporter, a door in the cell wall. In muscle, more than nine out of ten of those doors sit stored inside the cell at rest rather than installed in the wall. Insulin installs doors. Muscle contraction during exercise does the same job by a different route, with or without insulin. So the limit on how much sugar a muscle takes up is how many doors are at the surface, and how much tissue stands behind them matters much less. In a Yale study of 24 young lean people matched on age, weight, and device measured activity, the insulin resistant group made 61 percent less muscle glycogen after two high carb meals. The carbs went to the liver instead, where de novo lipogenesis and liver triglyceride synthesis both more than doubled, blood triglycerides rose 60 percent, and HDL fell 20 percent. Nobody in that study was short of muscle.

The number that predicts nothing

The largest analysis on this, the Sarcopenia Definitions and Outcomes Consortium, pooled 18,800 older adults and looked at four outcomes: falls, fractures, mobility loss, and death. Grip strength predicted all four. Gait speed predicted all four. Lean mass on a DEXA scan, the way we measure muscle mass, showed no relationship to any of them. Its recommendation was to remove DEXA lean mass from the definition of sarcopenia. And people with obesity on average carry more muscle mass than lean people and are stronger in absolute terms. What they have is muscle that works worse per kilogram, because fat has infiltrated the tissue.

Fat in the wrong rooms: location, type, and flux

When the fat store under the skin fills up, fat goes where it should not: the liver, the pancreas, the muscle. But endurance athletes carry as much fat inside their muscle fibers as people with type 2 diabetes, and they are among the most insulin sensitive people ever measured. The difference is speed: athletes turn that intramuscular fat over about two and a half times faster. Fat that sits still is what makes a muscle weaker and more insulin resistant. The liver behaves the same way: having liver fat worsens insulin resistance, but going from 6 to 25 percent changes nothing. What matters is the type of fat, reactive diacylglycerol versus inert triglyceride, and whether it moves. Three things decide whether someone is in trouble: where the fat is, what type it is, and whether it is flowing. None of them is a number on a scale.

What this means for training and GLP-1 drugs

A single 45 minute elliptical session more than tripled the amount of carbohydrate that 12 insulin resistant people put into muscle after a meal, and cut the fat their liver built by 30 percent, with no muscle gained. The effect lasts: still present at 48 hours, gone by day five. A large muscle that has not contracted in five days is, by this measure, an untrained one. On GLP-1 drugs like Wegovy and Ozempic, the panic over lean mass loss rests on an inflated number: DEXA counts water, glycogen, organs, and much of the fat lost from the wrong places as lean mass. In the semaglutide study with no exercise, grip strength improved by about 4.5 kilograms even as the scan said otherwise. The share of people with sarcopenic obesity fell from 49 to 33 percent just by losing fat.

Bottom line

Yes, lift, eat enough protein, and if you carry excess fat, lose it. But muscle mass is not the target. What decides metabolic health is the rate at which tissue handles what arrives, and the way to improve it is to use the muscle and get fat out of where it does not belong, not to chase a number on a scan.