Does Creatine Break a Fast?

No. Plain creatine monohydrate doesn't break a fast on any definition people actually use. It carries no usable energy, and when researchers measured the question directly, creatine on its own moved serum insulin not at all.¹

The catch is that plenty of people aren't taking plain creatine any more. Gummies, flavored powders, ready-to-drink shots and pre-workout blends are a different product with a different answer — and that answer is printed on the label, not contained in the creatine.

Here's the arithmetic, the one study everybody skips, and the two formats that genuinely do end a fast.

Does creatine have any calories?

Page one of the search results is confident about this: a 5 g scoop is 15 to 20 calories, because creatine "provides four calories per gram, the same as protein."

We went looking for the source of that figure and could not find one. We found no trial behind it and no regulatory basis for applying protein's four-calories-per-gram conversion to creatine, which the body does not handle like protein at all. One article making the claim contradicts itself within a few hundred words: creatine "still likely provides the same four calories per gram as protein," and then "some creatine products may list calories as zero, but this is because the caloric amount is less than five calories." Four calories per gram across a 5 g scoop is twenty. Those two sentences sit in the same article and are at least fifteen calories apart.

The mechanism settles it. Creatine is not a fuel. Muscle uses it as a phosphate shuttle: phosphocreatine hands its phosphate to ADP to regenerate ATP during hard efforts, and the creatine is left behind to be recharged. It isn't oxidized for energy, and its exit route isn't the mitochondrion — about 1–2% of intramuscular creatine degrades into creatinine and is excreted in the urine, which is why the body has to replenish roughly 1–3 g a day just to hold stores level.⁵ The molecule leaves through the kidney, not the lungs.

Even taken at face value, the internet's own worst-case number sits under every threshold anyone fasts by. A tub of monohydrate declares zero calories, and US labeling rules permit anything under five calories per serving to be declared as zero.⁸ So the honest range is somewhere between zero and about twenty calories, and it does not matter at either end of that range.

One beat nobody makes: a fast lowers your creatine intake. About half the daily requirement normally comes from food, and a pound of uncooked beef or salmon carries roughly 1–2 g of it.⁵ Skipping meals means skipping that. A scoop during a fasting window replaces what the fast removed rather than adding something on top.

Does creatine spike insulin?

No — and unusually for a supplement argument, this one has been measured rather than asserted.

In a 1996 trial, 24 men took 5 g of creatine in solution four times a day for five days. One group followed each dose with 93 g of simple carbohydrate 30 minutes later; the other took creatine alone. Total muscle creatine rose 60% more in the carbohydrate group, and the authors reported it plainly: "Creatine supplementation had no effect on serum insulin concentration, but Cr and CHO ingestion dramatically elevated insulin concentration."¹ The insulin came from 93 grams of sugar. It did not come from the creatine.

The more useful finding is that the arrow runs the other way from what the search results imply. Insulin drives creatine into muscle, not the reverse. In a euglycemic clamp study, seven men received 12.4 g of creatine at four different insulin infusion rates; muscle total creatine rose only at the two highest rates, by 4.5 ± 1.4 and 8.3 ± 1.0 mmol/kg dry mass. The conclusion was that insulin enhances muscle creatine accumulation "only when present at physiologically high or supraphysiological concentrations," and that it works by increasing creatine transport into the muscle rather than delivery to it.²

That's why the old carbs-with-creatine trick works, and it's also why it costs a faster nothing. A follow-up study found that 50 g of protein with 47 g of carbohydrate raised insulin as much as roughly 96 g of carbohydrate alone, and both augmented whole-body creatine retention by about 25% — the food is doing the work in each case.³ In a fasting window you are deliberately not eating that food, so you forgo a modest retention advantage that matters most during a loading phase. You don't break anything. More on that trade-off in our guide to taking creatine with carbs or protein.

Over a longer horizon the picture holds. Creatine loading at 20 g/day for five days followed by 3 g/day for 28 days did not influence glucose tolerance or any measure of insulin sensitivity in healthy untrained men — though the placebo group in that trial ingested matched amounts of glucose, which is an unusual comparator and makes the comparison less clean than it first looks.⁴

"Breaking a fast" depends on which fast you mean

There is no clinical definition of the phrase, so the threshold moves with your reason for fasting. We work through what "breaking a fast" actually means in full elsewhere; here are the rows that apply to creatine.

The autophagy question, honestly

The pages ranking for this state it as settled, in both directions, which should be a clue.

Two things are worth saying. First, creatine is not an amino acid. It is a guanidino compound: your body builds it from arginine and glycine via the enzyme AGAT to form guanidinoacetate, which is then methylated by GAMT to make creatine.⁵ The popular argument on page one — amino acids inhibit autophagy, creatine is made of amino acids, therefore creatine blunts autophagy — breaks at the first link. Swallowing creatine does not deliver free amino acids.

Second, even if the premise held, there would be nothing to check it against. A 2025 review of intermittent fasting and autophagy concluded that identifying reliable biomarkers for monitoring autophagic activity in humans remains an open challenge.⁶ Fasting is a well-established autophagy trigger in laboratory models; whether creatine touches that process in a living person has never been measured.

So: no plausible mechanism, and nobody has checked. "Creatine blunts autophagy" and "creatine doesn't blunt autophagy" are both guesses, and anyone stating either one confidently is guessing too.

Where the answer actually changes: gummies, flavored powders and blends

This is the part page one rushes past, and it's where the answer actually flips.

Gummies. A gummy needs sugar or a sugar alcohol to be a gummy — that's the format, not a formulation choice. The trap is the serving size. Most gummy creatine products need three or four gummies to reach a 3–5 g dose, so a product declaring "1 g sugar" per gummy is 3–4 g of sugar at your actual dose. Read the total sugars line, then multiply by the number of gummies you take. And know the rounding rule: total sugars under 0.5 g per serving may be declared as zero, and calories under 5 may be declared as zero.⁸ Per serving — and the panel's serving may not be your dose. To be clear about where the effect comes from: the sugar in a creatine gummy is what breaks your fast, not the creatine in it.

Flavored creatine powders. The split that matters is sugar-sweetened versus non-nutritively sweetened. On a blood-sugar definition, sucralose, stevia and aspartame look fine — a meta-analysis of 29 randomized controlled trials in 741 participants found non-nutritive sweeteners did not raise blood glucose, with no difference by sweetener type.⁷ Two caveats we carry every time we cite it: glucose was tracked against participants' own baseline rather than a separate control arm, and four chemically unrelated sweeteners were pooled. On a strict gut-rest definition, a flavored drink is still a flavored drink.

Pre-workout blends and ready-to-drink shots. Usually the clearest yes. Maltodextrin, dextrose, cyclic dextrin and juice bases are there on purpose, and they are food.

The rule to remember: read the carbohydrate and calorie lines, not the word "creatine."

Fitting creatine into a 16:8 window

Creatine works by saturation, not by an acute hit — the variable that changes your results is whether you take it every day, not the hour you take it.⁹ So the honest answer to "in the window or outside it?" is: whichever one you'll never skip.

If you want the marginal retention edge, take it inside the eating window with a meal, since that's where the food-driven insulin effect lives.³ Be realistic about its size: that effect matters most while you're loading, and on a steady 3–5 g maintenance dose your muscles reach the same plateau either way. If your training sits in the fasted part of the day, taking creatine fasted costs you nothing on the endpoint — see when to take creatine for the fuller version.

What we won't do is invent a protocol. There is no evidence for "take it 30 minutes before you break your fast," and precision nobody has tested is just decoration.

Does taking creatine fasted upset your stomach?

Gastrointestinal upset from creatine is dose-dependent, not fasting-dependent. In a controlled trial in 59 athletes, taking 10 g as a single serving produced significantly more diarrhea than the same 10 g split into two 5 g servings, while the split dose was no worse than placebo.¹⁰

No trial has compared fasted versus fed creatine for GI tolerance. We looked; that comparison hasn't been run. What's established is that complaints cluster around large single doses and aggressive loading, and it's reasonable to expect an empty gut makes a big dose feel worse — but that's a mechanism-level expectation, not a finding, and we're labeling it as one.

Practically: 3–5 g dissolved fully in a decent volume of water, and if it bothers you inside a fasting window, move it to a meal. For who tends to react and why, see creatine side effects.

One last thing, since it comes up in the same breath: if the scale moved up in your first weeks of creatine, that's intracellular water, not food and not fat. A fast isn't broken by water.

So the verdict stands: plain creatine monohydrate doesn't break a fast, on the calorie definition, the insulin definition, or any definition that has actually been measured. The gummy might. The pre-workout probably does. Read the label, not the word "creatine."

This page is educational and is not medical advice.