Key takeaways
- Four inputs, in order: resistance training is the signal, protein is the substrate, creatine buys training quality, and sleep is when the work is done.
- Appetite suppression makes dense protein the least appealing food on the plate, so protein intake falls by default unless it is eaten first at every meal.
- Strength on key lifts is the cheapest signal that lean mass is holding; if strength falls, the correction is protein and training, not a steeper deficit.
Weight comes off on GLP-1 therapy whether or not you do anything else. What comes off with it is the part you control. Lean mass is not protected by default in a large energy deficit, and appetite suppression quietly removes the single input that protects it most — protein. This is the protocol, in the order that matters.
Why this deficit behaves differently
Every energy deficit costs some lean tissue. The body draws on whatever is metabolically cheap, and skeletal muscle that is not being loaded is cheap — expensive tissue to maintain with no current demand attached to it.
What makes GLP-1-assisted loss distinctive is not that it strips muscle faster per pound lost. Hunger is what usually stops people under-eating for months at a time, and removing that brake makes the deficit larger and more sustained than anything the person has held before — while suppressing the appetite that would have dragged protein intake up. The typical picture is someone eating half their former volume of food and taking in less protein than when they were heavier and not trying. All of it is invisible unless you measure it.
The four pillars, in priority order
Four inputs, and the order is not arbitrary:
- Resistance training — the signal that the tissue is still needed
- Protein adequacy — the substrate to act on the signal
- Creatine — the amplifier for training quality
- Sleep and recovery — when the building actually happens
Skip any pillar and muscle loss accelerates. Hit all four and muscle is largely preserved through a meaningful share of body weight loss (Deller et al., Diabetes Obes Metab 2026). If you can only do one thing, do the training.
Pillar one: the signal
Loaded muscle is retained muscle. Mechanical tension tells the body this tissue is earning its keep, and the retention decision follows. Cardio does not send that signal, which is why walking a great deal while eating very little is a reliable way to arrive at goal weight smaller and softer than expected.
- Frequency: 2-3 sessions per week
- Volume: 8-12 working sets per major muscle group per week — the lower end during the heaviest phase of loss
- Intensity: 70-85% of one-rep max, 6-12 reps, near but not to failure on most sets
- Movements: compound emphasis — squat or leg press, deadlift or RDL, press, row, pulldown or pull-up
- Progression: hold or modestly increase load. The goal is not a record; it is proof the signal is still being sent
- Form: quality over load — in a deficit, sloppy heavy sets buy injury rather than adaptation
Training in a deficit is harder than training in a surplus, and the session that felt routine three months ago will not. That is a reason to stop training to absolute failure on everything, not a reason to stop. The muscle-preservation program is built for this phase, and pairing training with GLP-1 therapy covers why the combination outperforms either alone.
Pillar two: the substrate
Protein is the most-cited and least-followed muscle preservation factor (Longland et al., Am J Clin Nutr 2016). It matters more in a deficit because when energy is scarce, amino acids get diverted toward fuel unless there is enough of them to spare for structure.
- Daily target: 1.6-2.0 g/kg of goal body weight
- Per meal: 25-40 g across 3-4 meals
- Leucine threshold: roughly 2.5 g per meal is what maximally triggers muscle protein synthesis, which is about 25-30 g of animal protein
- Quality: animal sources carry a complete amino acid profile; plant sources need deliberate combining and a higher total
- Anchors: 8 oz chicken (~50 g), 6 oz fish (~40 g), 4 eggs (~25 g), a scoop of whey (~25 g), a cup of Greek yogurt (~20 g)
The execution problem is specific to this therapy. Appetite suppression does not reduce all foods equally — dense protein becomes the least appealing thing on the plate while small amounts of carbohydrate stay tolerable, so intake drifts toward whatever goes down easily. The fix is order of operations: protein first at every meal, while there is still room. Shakes are not a compromise when food volume is genuinely small; often they are the only way the target gets hit (more on the numbers).
Pillar three: the amplifier
Creatine is the most evidence-supported supplement for muscle preservation and strength (Kreider et al., J Int Soc Sports Nutr 2017). It keeps phosphocreatine stores topped up, which is what lets you finish the last two reps of a set — small differences in training quality, repeated across months.
- Dose: 5 g of monohydrate daily
- Timing: irrelevant. Consistency is not
- Effect: around 5-10% strength preservation and 2-4 lb of apparent muscle (intracellular water plus real tissue)
- Safety: among the best-characterised supplements in existence
One expectation to set: the scale stalls for a week or two after starting, because creatine pulls water into muscle cells. People read that as failure and stop. It is intracellular water, and it is the point (full creatine guide).
Pillar four: when it is actually built
Short sleep in a deficit shifts the composition of what is lost toward lean tissue (Nedeltcheva et al., Ann Intern Med 2010). This is the pillar treated as optional, and the one that determines whether the other three convert.
- Seven or more hours, with consistent timing mattering more than the exact number
- Alcohol limited — it degrades sleep architecture and blunts protein synthesis the same night
- Caffeine finished early enough to clear
Recovery management is the same principle awake: treat soreness and joint complaints as data rather than badges, and pull volume down during high-stress weeks rather than pushing through them.
Sequencing, timelines and how to know it is working
The common mistake is starting all four pillars on the day of the first dose and abandoning three within a fortnight. Establish the training habit and protein floor first, add creatine immediately because it costs no willpower, and treat sleep as a standing condition. Strength should hold roughly steady through the first several weeks — that is the signal it is working.
The scale cannot answer the composition question. It reports total mass and says nothing about what the mass was made of. What does answer it:
- Strength on key lifts — the fastest and cheapest signal. Holding load in a deficit is a good outcome
- DEXA every 3-6 months — the closest thing to a direct measurement of lean mass available outside a lab
- Tape at waist, arm and thigh — limb circumference falling faster than waist is a warning
- Photographs in the same light and pose every 2-4 weeks
- The scale, as one input rather than the verdict
Weight down with strength stable means fat is coming off. Weight down with strength falling means you are losing the tissue you are trying to keep, and the correction is protein and training volume, not a steeper deficit.
The clinical pearl: muscle preservation is a structural problem, not a motivational one. Patients who follow the four pillars lose a far greater share of the total as fat. Those who skip training and protein reach the same scale weight with worse body composition — and pay for it when weight returns.
Bottom line
Training is the signal, protein is the substrate, creatine buys training quality, sleep is when the work is done. The therapy handles appetite; it does not handle body composition, and it makes protein intake harder rather than easier — which is exactly why the protocol has to be deliberate. Measure strength and lean mass rather than weight, and correct with protein and training rather than a bigger deficit. The 60-second assessment routes the medical side to a physician who can look at both.
Educational content, not medical advice. Laboratory interpretation and any treatment decision are made by a licensed physician after individual evaluation. Individual results vary.
