What this page establishes
- Protein intakes of 1.2 to 1.6 g/kg/day and resistance training are the evidence-backed levers for lean-mass preservation in a deficit, and total daily intake matters more than meal timing in that literature.
- Adding exercise to GLP-1 treatment preserved fat-free mass and improved fitness in randomised data, regardless of fed state, which was not a studied variable.
- ADA guidance directs reducing sulfonylurea and insulin doses when a GLP-1 is added, which is the relevant hypoglycaemia context for fasted exercise.
- Semaglutide injectable: 0.25 / 0.5 / 1.0 / 1.7 / 2.4 mg weekly, 4 weeks per step, 2.4 mg maintenance; 14.9% mean loss at 68 weeks (STEP 1).
Quick answers
I have type 2 diabetes but take neither insulin nor a sulfonylurea. Does the hypo warning apply to me?
The adjustments described on this page are specific to sulfonylureas and insulin, because those lower glucose whether or not you have eaten.
Read the full answer
Other diabetes medicines behave differently and this page does not cover them. Ask your diabetes team what your own regimen does around exercise before making fasted sessions a habit.I keep failing sessions I used to finish. Is that fuelling or the drug?
Both are candidates and reading will not separate them. Low glycogen bites hardest at higher intensities, so easy aerobic work falling apart points somewhere other than the pre-session meal.
Read the full answer
Repeated inability to complete routine sessions is on the list of things to take to a clinician, along with the fatigue and intake questions usually sitting underneath it.I am doing 16:8 on top of this. Do the two stack?
It narrows an eating window that appetite suppression has already narrowed for you, which makes the day's protein target harder to hit rather than easier.
Read the full answer
Total daily intake is the lever with evidence behind it, so a schedule that quietly costs you intake is working against the thing that matters most. Our page on 16:8 time-restricted eating covers the method itself.
Nobody has compared fasted with fed training in people taking GLP-1 receptor agonists. That study does not exist, and any confident answer you find is extrapolating from research done in people who were not on these drugs. What can be said is that the levers with actual evidence behind them are total daily protein intake and doing resistance training at all, and in the general literature those dominate meal timing. On a GLP-1 most people end up training semi-fasted regardless of intention, because appetite is suppressed and eating before a session is unappealing or uncomfortable. The one place fasted training carries a specific danger is if you take insulin or a sulfonylurea, where exercising without carbohydrate on board raises the risk of hypoglycaemia.
Why almost all training becomes semi-fasted
Appetite suppression means breakfast often stops happening. Delayed gastric emptying means eating shortly before a session feels worse than it used to. Total intake falls, and the intake that remains clusters at whatever time of day eating is easiest.
The result is that most people on this drug class end up training several hours after their last real intake without having decided to do intermittent fasting. It is a side effect of the drug rather than a training protocol.
There is a subtlety here that catches people out. Feeling full is not the same as being fed. Delayed emptying can leave you feeling like you ate recently when the last meaningful intake was many hours ago and the amino acids from it are long since dealt with. Judging your fed state by how full you feel is unreliable on this drug.
What fasted training does and does not do
Training fasted increases fat oxidation during the session itself. That is well established and it is the reason the practice has a following. It does not follow that it produces more fat loss over weeks, because total energy balance determines that, not which substrate was burned in a given hour.
At low intensities the glycogen cost is modest and fasted work is largely unaffected. At higher intensities glycogen availability starts limiting, and work capacity falls. Someone doing easy aerobic work fasted will barely notice. Someone attempting intervals or a heavy session in the same state usually will.
In normal energy balance the fasted-versus-fed trade-off is roughly neutral for body composition. In a sustained deficit the picture shifts, which is the situation you are in on a GLP-1.
The muscle-preservation concern
This is where the deficit changes the calculus. Lean soft tissue accounted for roughly 40% of the weight lost in the STEP 1 DXA substudy and about 26% in SURMOUNT-1, with a network meta-analysis putting the class average near 25%. None of those trials included a structured resistance-training programme.
In a deficit, amino acid availability is already low. Training in a fasted state with no protein feeding anywhere nearby leaves muscle protein balance negative for longer than it otherwise would be. That is a mechanistic argument rather than a measured one in this population, and it should be read as such.
What has been measured is the thing that works. Protein intakes of 1.2 to 1.6 g per kg per day combined with resistance training are the evidence-backed levers for preserving lean mass during energy restriction, and in that literature total daily intake matters more than the timing of individual meals. Adding exercise to GLP-1 treatment preserved fat-free mass and improved fitness in randomised data, with fed state never being a variable anyone studied.
So the honest ranking: get the total protein in, do the resistance training, and worry about whether you ate beforehand after those two are handled. Our page on protein targets covers the intake side, and working out covers the training.
| Lever | Evidence status |
|---|---|
| Total daily protein 1.2 to 1.6 g/kg/day | Established in the energy-restriction and resistance-training literature |
| Doing resistance training at all | Established, and randomised evidence in a GLP-1 population shows fat-free mass preservation |
| Training fed rather than fasted | Unstudied in this population, mechanistic argument only |
| Protein timed close to the session | Secondary to total intake in the general literature, unstudied here |
Hypoglycaemia if you take insulin or a sulfonylurea
This is the one part of the page that is a safety matter rather than an optimisation question.
ADA Standards of Care 2026 direct that when a GLP-1 receptor agonist or dual GIP/GLP-1 receptor agonist is added, sulfonylureas be discontinued or reduced and insulin adjusted, with examples including reducing bolus insulin by 10% to 20% and basal by around 10% where HbA1c is under 7.5%. Those adjustments exist because the combination lowers glucose more than either alone.
Exercising without carbohydrate on board pushes glucose down further, on top of a regimen that may already need adjusting. If you take either of those drug classes, fasted exercise is something to plan with your diabetes team rather than to try and see what happens. Ask specifically about glucose monitoring around sessions and what to have available.
Practical alternatives when solid food is intolerable
Liquid protein before or shortly after a session is the usual workaround when solids are unappealing or sit badly, and it is the most common thing people on this drug class settle on. Liquids also empty from the stomach faster than solids in normal physiology, which helps when gastric emptying is already slowed.
Some people find training later in the day works better, once a couple of small intakes have happened. Others find early morning easiest because nausea is lower. There is no evidence to pick between those, so this is genuine trial and error rather than a hidden right answer.
The one thing worth being deliberate about is the total. If skipping pre-training food means the protein never gets made up later in the day, the timing decision has quietly become an intake problem, and the intake problem is the one that matters.
If food and eating are difficult for you
Training fasted, tracking protein and eating in a deficit are a combination that can be difficult territory for anyone with a history of restrictive eating. Appetite suppression makes under-eating effortless, and adding a training structure on top can make it feel virtuous.
If eating less has started to feel compulsive rather than chosen, that is worth raising with your clinician or an eating disorder support service. It is not a reason to stop training, and it is not something to work through alone.
When to contact a clinician
- Symptoms of hypoglycaemia during or after exercise, particularly if you take insulin or a sulfonylurea
- Dizziness or fainting during fasted sessions
- Repeated inability to complete sessions that used to be routine
- Weeks of training while consistently well under your protein target
- Eating patterns that have started to feel compulsive
Not medical advice
This page reports the general exercise and nutrition literature, randomised evidence in GLP-1 populations, and ADA guidance, and says plainly where no evidence exists for this specific question. It is general educational information, not a training or nutrition plan, and not advice about your diabetes medication.
A first-person account, for context
Everything above is drawn from trial data and prescribing information. What that evidence cannot give you is what a GLP-1 block actually feels like week to week in someone training seriously.
Thomas Prommer, a competitive endurance athlete, has written up training through the side effects, session by session. Read it as one person's experience rather than as evidence. It is n=1, it is uncontrolled, and it describes an athlete population the trials on this page did not study. It is useful for exactly that reason and for no other.
The evidence, one row per claim
| Claim | Tier | Source |
|---|---|---|
| No trial has compared fasted with fed training in people taking GLP-1 receptor agonists. | anecdotal | unsourced |
| Protein intakes of 1.2 to 1.6 g/kg/day and resistance training are the evidence-backed levers for lean-mass preservation in a deficit, and total daily intake matters more than meal timing in that literature. | established | Morton et al., protein supplementation and resistance training meta-analysis (PMC) |
| Adding exercise to GLP-1 treatment preserved fat-free mass and improved fitness in randomised data, regardless of fed state, which was not a studied variable. | established | Physical Fitness with Exercise and GLP-1 RA Treatment (Sports Medicine) |
| ADA guidance directs reducing sulfonylurea and insulin doses when a GLP-1 is added, which is the relevant hypoglycaemia context for fasted exercise. | established | ADA Standards of Care 2026, Section 9 |
| Semaglutide injectable: 0.25 / 0.5 / 1.0 / 1.7 / 2.4 mg weekly, 4 weeks per step, 2.4 mg maintenance; 14.9% mean loss at 68 weeks (STEP 1). | established | STEP 1 |
| Tirzepatide: 2.5 mg initiation then 5 / 10 / 15 mg weekly maintenance; 16.0% / (10 mg intermediate) / 22.5% mean loss at 72 weeks (SURMOUNT-1). | established | Eli Lilly / NEJM |
Sources
- Morton et al., protein supplementation and resistance training meta-analysis (PMC)
- Physical Fitness with Exercise and GLP-1 RA Treatment (Sports Medicine)
- ADA Standards of Care 2026, Section 9
- STEP 1
- Eli Lilly / NEJM
Keep reading
- GLP-1 and Working Out: Strength Training, Muscle Loss and ProteinLean mass loss is real and measured. The trials that measured it mostly had nobody lifting, so treat those num
- GLP-1 and Protein: How Much, and Why It Is Hard to EatThe protein evidence predates GLP-1s and is solid. The new problem is not the target, it is reaching one when
- GLP-1 and Diabetes: What Differs Between the T2D and Obesity IndicationsOzempic and Wegovy are the same molecule with different indications. The part that can hurt you is what happen
- GLP-1 and Energy: Fatigue, Gym Performance and What Causes ItFatigue on a GLP-1 usually has a findable cause. Separate the fixable from the expected instead of accepting i
- GLP-1 and Endurance Event FuellingEndurance fuelling depends on the one thing this drug class slows. The conflict is real, the label documents t