Ask ten people why they won’t take creatine and at least three of them will say some version of the same thing: it makes you bloated. It’s probably the single most common objection to the most studied supplement in sports nutrition, and it has kept a lot of people, especially women, from ever trying it.
Here’s the short answer before we get into the studies. Creatine does pull water into your body, but the effect is real mostly in the first week or so, the water goes inside your muscle cells rather than under your skin, and over longer periods the research generally finds that total body water tracks with muscle mass rather than ballooning past it. The puffy, soft, water balloon picture that people carry around in their heads doesn’t match what the fluid measurements actually show. And in women, the newest research found that during the parts of the menstrual cycle when fluid shifts feel worst, loading raised water in every fluid compartment, including outside cells, without changing body weight.
That’s the summary. Now let’s earn it, because the honest version of this story includes some findings that supplement marketing tends to skip.
Why Creatine Attracts Water in the First Place
Creatine is osmotically active. That’s a technical way of saying it attracts water, the same way salt does. When you supplement, creatine travels from your blood into muscle cells through a dedicated transporter that also carries sodium, and water follows to keep the concentration inside the cell balanced. Around 95 percent of the body’s creatine sits in skeletal muscle, so that’s where the water goes too.
The classic dose finding work comes from Hultman and colleagues in 1996, who gave 31 men creatine in different amounts and measured muscle content directly by biopsy. Six days at 20 grams per day raised total muscle creatine by about 20 percent. So did 28 days at 3 grams per day, just more slowly. When supplementation stopped, muscle creatine drifted back to its starting level over about 30 days. That 20 percent bump is the physiological reason water moves. More creatine inside the cell means more osmotic pull, so water comes in and the cell swells slightly. Nobody in the field disputes that mechanism. What’s disputed is how much water, where it goes, and whether it matters after the first month.
This is not a side effect in the sense of something going wrong. It’s the mechanism working as intended. Muscle cells that hold more creatine hold more water, and a better hydrated cell is, if anything, a functional advantage. Researchers have pointed out for years that increased intracellular water acts as a signal that supports protein and glycogen synthesis inside the cell. Cell volume is not just storage, it’s information.
The key distinction, and the one that decides whether “water retention” is a problem or a footnote, is where the water sits. Intracellular water lives inside your muscle cells. Extracellular water sits outside cells, in the spaces between tissues and in plasma, and that’s the compartment responsible for the visible puffiness people mean when they say bloated. Swollen ankles, a soft face, premenstrual water weight, those are extracellular stories. So the real question was never whether creatine adds water. It’s which compartment the water lands in.
What the Classic Studies Actually Measured
The paper that gets cited most on this question is a 2003 randomized trial by Powers and colleagues in the Journal of Athletic Training, and it’s worth describing properly because both sides of the argument quote it selectively. The researchers gave 32 healthy men and women either creatine monohydrate or a sucrose placebo, using a heavy protocol by today’s standards: 25 grams per day for 7 days, then 5 grams per day for 21 more. They measured muscle creatine, body mass, total body water, extracellular water and intracellular water using dilution methods, which are more rigorous than the bioelectrical impedance scales most later studies rely on.
The results cut both ways. Creatine clearly increased muscle creatine content, body mass and total body water. So anyone telling you creatine adds zero water is not reading the data. But the distribution of that water across compartments didn’t change. The proportion of body water sitting inside cells versus outside them stayed where it was. The extra fluid was not pooling in the extracellular space where bloating lives. The paper’s own title says it plainly: total body water went up without altering fluid distribution.
One thing Powers did that later studies mostly haven’t: half the participants were women. The published summary reports the pooled result for the whole group rather than breaking the water findings out by sex, so it can’t tell you whether women responded the same way as men. But it is one of the few body water datasets that included women on a maintenance dose at all, and with 32 people total it’s worth knowing about and worth holding loosely.
Ziegenfuss, Lowery and Lemon in 1998 gave 10 trained men 0.35 grams of creatine per kilogram of fat free mass per day for three days. By day three, total body water had risen about 2 percent, roughly 0.86 liters, and intracellular volume had risen about 3 percent. Extracellular volume, the water outside cells in blood and between tissues, did not change. The authors concluded that the early weight gain from creatine is mostly water and that most of it sits inside cells.
A 2021 review of common creatine questions by Antonio and colleagues in the Journal of the International Society of Sports Nutrition pulled the wider literature together and found a consistent pattern. Short term studies, particularly ones using loading doses of around 20 grams per day, tend to find increases in total body water. Several longer training studies running five to ten weeks found no significant increases in total body water, extracellular water or intracellular water at all. The same review noted the practical upshot people actually care about: loading phases in men commonly produce a body mass jump of roughly 1 to 2 kilograms in the first days, and most of that early gain is water.
So the honest reading is not “creatine causes no water retention.” It’s that creatine causes a real, measurable, mostly intracellular increase in water early on, especially with loading, and that this effect fades into the background over weeks as it becomes indistinguishable from the water content of new muscle tissue.
The Muscle to Water Ratio, and Why It Matters
The most useful single study for settling the “am I just holding water” worry came from Ribeiro and colleagues in 2020, published in the International Journal of Sport Nutrition and Exercise Metabolism. Twenty seven resistance trained men took either creatine or placebo for 8 weeks while following the same lifting program. The creatine group gained more skeletal muscle, more total body water and more intracellular water than the placebo group, with total body water up about 7 percent and intracellular water up about 9 percent in the creatine group.
That sounds like water retention until you look at the ratio. The relationship between skeletal muscle mass and intracellular water was the same in both groups at the end of the study. In other words, the creatine users weren’t waterlogged versions of the placebo users. They had more muscle, and more muscle contains more water, in the same proportion muscle always does. Muscle tissue is roughly three quarters water by weight whether you supplement or not. The creatine group carried more water for the same reason a bigger sponge holds more water than a smaller one.
There’s a detail in that study worth flagging because it’s the kind of thing this site exists to report: the creatine group did show a small decrease in the ratio of skeletal muscle mass to extracellular water, which is a hint of some extracellular change, even though absolute extracellular water rose similarly in both groups. It’s a minor wrinkle in an otherwise clean result, and with 27 participants it shouldn’t be overinterpreted in either direction, but it’s in the paper and you deserve to know it’s there.
Francaux and Poortmans in 1999 ran a nine week study in young men that combined 42 days of strength training with 21 days of detraining. The creatine group took 21 grams per day for five days and then 3 grams per day for the rest. Body water compartments were measured by bioimpedance spectroscopy. The creatine group gained about 2 kilograms, and the placebo and control groups gained nothing. The authors attributed part of the gain to water but noted that the relative volumes of the compartments stayed constant, and they argued that the medium term weight gain reflected growth of dry tissue with a normal amount of water attached rather than water retention on its own. It’s a small study with a small creatine group of eight, and a later review pointed out that the loading period was short, so I’d treat it as suggestive rather than settled.
A 2024 systematic review and dose response meta analysis on creatine and body composition covering 143 studies added an important caution about the whole evidence base here. Only 8 of those 143 studies used methods capable of separating body water into compartments or reported water data at all. The reviewers concluded, reasonably, that more studies are needed before anyone can speak with full confidence about long term water compartment changes. That’s a genuine limitation of the field. The compartment data we have points consistently in one direction, but there’s less of it than the confident tone of most supplement blogs would suggest.
Where Men and Women Diverge
For most of the history of creatine research, participants were young men, and the water retention folklore was built on their results. Men loading creatine at 20 to 25 grams per day genuinely do gain a kilogram or two quickly, and that experience got generalized to everyone. But female physiology handles creatine differently from the start.
According to a 2021 lifespan review of creatine in women by Smith-Ryan and colleagues, women have around 70 to 80 percent lower endogenous creatine stores than men, consume less dietary creatine on average, and yet show somewhat higher resting creatine concentrations inside muscle relative to their muscle mass. Smaller total muscle mass also means less absolute room for creatine and its accompanying water. The same review notes hormonal effects on creatine metabolism across the menstrual cycle, pregnancy and menopause, which is where the story gets more interesting than a simple men versus women comparison.
One disclosure to keep in view: two authors of that literature, Smith-Ryan and Candow, serve as scientific advisors to a company that manufactures creatine, and they disclose this in their papers. Disclosed advisory roles are normal in this field and don’t invalidate the work, especially randomized controlled trial work where the design constrains wishful thinking, but readers should know it, and most articles citing this research never mention it.
The trial that speaks most directly to the bloating fear in women is a 2023 randomized controlled trial from the University of North Carolina published in Nutrients. Thirty moderately active women, some naturally menstruating and some using hormonal contraceptives, took either 20 grams of creatine monohydrate per day for 5 days or a placebo, in a double blind design measured across both the follicular and luteal phases of the cycle. The follicular phase is the stretch from the start of a period to ovulation, and the luteal phase is the stretch after ovulation, which is when many women experience premenstrual fluid retention and bloating.
The headline finding: creatine loading did not produce significant changes in body mass in these women. Read that again with the male loading studies in mind. The 1 to 2 kilogram scale jump that men reliably see in week one largely failed to appear. Fluid compartments did shift: during the luteal phase total body water rose in every compartment, including extracellular fluid by about 0.46 liters, without the women gaining measurable weight, which does not show creatine reversing a hormone driven fluid shift.
A follow up line of work from the same lab, published in 2025 in the Journal of the International Society of Sports Nutrition, looked at creatine loading, cellular integrity and jump performance across the menstrual cycle using phase angle, a bioimpedance measure that reflects cell membrane health and the balance of water inside versus outside cells. Creatine supported improved cellular fluid measures across the cycle, particularly in the mid to late luteal phase, exactly the window when extracellular fluid shifts tend to make women feel puffy. The researchers’ interpretation is that creatine may help draw water into muscle cells during a phase when hormones are pushing fluid the other way. The 2023 trial’s own numbers do not show that, since extracellular fluid rose too, and the idea comes from small studies in one lab and should be treated as promising rather than settled.
What Happens at Normal Daily Doses
Almost everything scary in the water retention literature comes from loading protocols: 20 to 25 grams per day, four to five times the standard maintenance dose. That matters, because most people considering creatine today, especially for cognitive rather than athletic reasons, take 3 to 5 grams per day and never load. A finding produced by a week of 25 grams daily is not a finding about your 5 gram routine, and the two keep getting blurred together.
The studies that used moderate doses or measured after several weeks paint a calmer picture. In the Antonio review’s summary of the training studies, five to ten weeks of supplementation alongside exercise produced no significant increases in any water compartment in several trials. A 2023 six week study in female collegiate dancers using 0.1 grams per kilogram per day, roughly 6 grams for a typical participant, did find a statistically significant increase in total body water, but the size of it was about half a kilogram, from 32.2 to 32.7 kilograms, alongside a lean mass increase. Half a liter of water distributed through your musculature is not something anyone can see in a mirror.
More recently, a 2025 study of creatine supplementation strategies in female recreational athletes compared 14 days of loading style and moderate dosing against placebo during a normal training period. Body mass, fat mass, lean tissue, intracellular water, extracellular water and total body water all failed to change significantly in any group. The only signal was a shift in the ratio of extracellular to total body water in the loading group, in the direction of relatively more water inside cells. Fourteen days of creatine in these women did not make the scale move, and to the extent water moved at all, it moved inward.
There is also a small 2026 conference abstract from Lambright and colleagues at Western Kentucky University. Eighteen recreationally active young adults, eight of them women, took 5 grams per day for seven days with bioimpedance spectroscopy at baseline, three hours, six hours and day eight. They found no significant change in intracellular or extracellular fluid. That’s a preliminary analysis presented at a conference, not a peer reviewed paper, and with 18 people it could easily miss a real effect. But it’s also the dose most people actually take, and it found nothing in a week. The loading studies are not necessarily telling you what 5 grams a day does.
It’s worth sitting with how consistent that direction is. Across designs, doses and populations, when researchers find a water effect, it’s intracellular. Nobody has produced a controlled trial showing creatine expanding the extracellular compartment the way the bloating story requires.
The Findings That Don’t Fit Neatly
An article like this one is only trustworthy if it reports the awkward parts, so here they are.
First, the early total body water findings are real and shouldn’t be waved away. Powers in 2003 found increased total body water. Earlier work from the same era found similar things at loading doses. If you load creatine, especially as a man with a lot of muscle mass, you will very likely gain water weight in week one, and pretending otherwise is marketing, not science. The weight is intracellular and mostly harmless, but it exists, it shows up on the scale, and for someone in a weight class sport or someone emotionally sensitive to scale numbers, it’s a legitimate consideration rather than a myth.
Second, the measurement tools have limits. Most of the compartment data comes from bioelectrical impedance, which estimates water compartments from how electrical current moves through the body. It’s validated but noisy, sensitive to hydration, food, exercise and, in women, cycle phase. The dilution methods Powers used are stronger but rare. The 2024 meta analysis’s point stands: only a small fraction of creatine studies measured body water carefully at all, so the field is drawing confident conclusions from a thin slice of well measured data.
Third, the women’s studies are small. Fifteen women per arm in the 2023 fluid distribution trial, similar numbers elsewhere. Small trials from a single research group produce findings that deserve attention and replication, not certainty. The direction of the female data is encouraging and biologically plausible, but a couple of larger independent trials could still move this picture.
Fourth, for completeness, an old strand of concern went the other way entirely: not that creatine bloats you, but that it dehydrates you and causes cramping by pulling water into muscle at the expense of everywhere else. The American College of Sports Medicine cautioned along these lines around 2000. Subsequent controlled research did not support it, and reviews since have concluded there’s no good evidence that creatine causes dehydration or cramping, with some athlete data pointing to fewer cramps among users. A 2025 analysis of NHANES survey data covering tens of thousands of Americans found dietary creatine intake associated with normal hydration biomarkers at the population level, though as observational data about food sources rather than supplements, it can only weakly support the trial evidence, not replace it.
Lopez and colleagues in 2009 systematically reviewed ten controlled trials of athletes exercising in the heat and found no adverse effect of creatine at recommended doses on heat tolerance or hydration status, and a pooled analysis found no relationship between creatine and core body temperature. Some trials found slightly lower body temperature and heart rate in the creatine groups, which is the opposite of the concern. The 2021 Antonio review reached the same conclusion on cramping.
I raise this because it tells you something about how creatine water retention actually behaves. If the extra water were being pulled out of the extracellular space and locked inside muscle, you’d expect trouble in the heat. You don’t see it. The body seems to add the water rather than move it, which is consistent with Powers finding total water up and distribution unchanged.
Who Is Writing These Papers
A reader trying to judge this evidence should know where it comes from. The 2021 misconceptions review, its 2024 sequel, and the 2024 body composition meta analysis share overlapping author lists of researchers who have received creatine donations, industry funding or conference travel support, which they disclose. The 2025 safety analyses led by Kreider and colleagues, which reviewed 685 clinical trials and found creatine well tolerated, had their publication charges paid by a creatine manufacturer as part of a conference special issue. A companion analysis found creatine mentioned in only 203 of 28.4 million adverse event reports worldwide, which is reassuring, and it also found that nearly half the US reports naming creatine involved products that didn’t contain it.
None of this makes the work wrong. The underlying trials were run by many groups over thirty years, and the water findings are consistent across them. But the summaries and the framing, the part that turns “total body water increased” into “creatine hydrates your cells,” tend to come from people with a stake in the ingredient. I read the primary studies for that reason, and I’d suggest you weigh the raw findings more heavily than the interpretive gloss. That applies to this site too.
How to Read a Body Water Study Without Being Misled
Because this topic gets argued about with dueling citations, it helps to know the three questions that decide whether a given study actually applies to you. Most bad takes on creatine and water retention come from skipping one of them.
The first question is the dose and whether the study used loading. A trial that gave participants 25 grams per day for a week is studying an intervention most people will never do. The physiological logic is simple: the faster you push creatine into muscle, the faster water follows, and the more dramatic the early scale change. When someone cites a loading study to warn you off a 5 gram daily habit, they’re describing a different intervention. The reverse error happens too, when maintenance dose studies showing no water changes get cited to claim loading is consequence free. Match the protocol to your plan.
The second question is duration and timing of the measurement. Nearly every positive water finding in this literature comes from measurements taken within the first one to four weeks. Nearly every null finding comes from measurements taken after five or more weeks, usually alongside training. Those aren’t contradictory results, they’re two points on the same curve. Water arrives early with the creatine, then muscle adaptation catches up and the ratio normalizes. A study is only “conflicting” if you ignore when it looked.
The third question is who was measured. Young resistance trained men loading creatine are the population where water gain is largest and most reliable. Moderately active women at the same doses barely registered a body mass change in the North Carolina trial. Older adults, adolescent athletes and untrained beginners are all measured less often, and extrapolating across these groups is exactly the kind of blur this literature suffers from. Population, dose and duration together tell you whether a finding is about you. Any citation offered without all three attached is closer to rhetoric than evidence.
There’s one more layer worth understanding, which is what the measurement device can and cannot see. Bioelectrical impedance, the method behind most compartment claims, sends a small current through the body and infers water volumes from resistance. It’s quick and cheap, which is why it dominates, but its compartment estimates carry real error bars, and things as ordinary as a salty dinner, a hard workout or a luteal phase can move the numbers. Deuterium and bromide dilution, the methods in the Powers trial, trace actual water spaces chemically and are considered the reference standard, but they’re expensive and rare. When an impedance study and a dilution study disagree, the dilution study usually deserves more weight. When two impedance studies disagree by small margins, the honest interpretation is often that neither difference is real.
None of this is a reason to distrust the field. It’s a reason to notice that the highest quality measurements and the largest studies point the same direction as the quick ones: water in, cells first, proportional over time.
What This Means If You’re Worried About Bloating
Pulling the threads together, here is what the evidence supports for someone deciding whether creatine water retention is a reason to stay away.
If you skip loading and take 3 to 5 grams of creatine monohydrate daily, the measurable water effect ranges from small to undetectable in the controlled studies that looked. Muscle creatine still saturates on this approach, it just takes three to four weeks instead of one. There is no performance or cognitive rationale that requires loading for the general user, and skipping it removes most of the week one scale jump that built this fear in the first place.
The other thing people do that the studies rarely model is take creatine while cutting calories or sodium, then blame creatine for the scale refusing to drop. A kilogram of intracellular water is real weight and it will sit on the scale for as long as you keep taking creatine. It is not fat, and it is not extracellular bloat, but if your goal is a number rather than a look, you should expect the number to be about a kilogram higher than it otherwise would be. That’s a reasonable thing to weigh against whatever you’re taking creatine for.
If you do load, expect the scale to rise, more if you’re male and muscular, less or possibly not at all if you’re female, based on the North Carolina loading data. The gain is water inside muscle cells. It is not fat, and the compartment studies give no reason to expect it under your skin or in your face.
One more distinction, because the word gets used loosely. Stomach bloating after a dose is a gut symptom, not a body water symptom, and it has a different cause. The 2024 sequel to the misconceptions review attributed the gastrointestinal complaints that do occur to large single doses of more than 10 grams taken at once, or to taking creatine alongside a lot of caffeine, and the 2025 safety analysis of 685 trials found no higher rate of gastrointestinal distress or weight gain in creatine groups than in placebo groups. A tight stomach an hour after 20 grams in one glass is real and it is not fluid retention. It’s also avoidable by splitting the dose or not loading at all.
If you’re a woman whose real concern is cyclical bloating, the current data, small as it is, suggests creatine is unlikely to make the luteal phase worse, though in the 2023 trial total body water rose in every compartment during it, extracellular fluid included, without a change in body weight. That’s a hypothesis with early supporting trials, not an established benefit, and it would be dishonest to promise it. But the fear that creatine stacks extra bloat on top of premenstrual water retention has no experimental support at all.
And if you’re taking creatine for your brain rather than your muscles, note that everything in this article concerns muscle and body water. Brain creatine uptake works through the same transporter family but on a slower timeline and at lower magnitudes, and none of the fluid findings here say anything about cognition either way. Water retention is a body composition question, and on that question the evidence is unusually reassuring for anyone using sensible doses.
Water follows creatine into muscle because muscle takes up a lot of creatine quickly. Brain creatine rises much more slowly and by much less in the studies that have measured it, and I haven’t found a trial measuring brain water in response to supplementation in healthy adults. The cell swelling that some researchers propose as a signal for muscle growth is a muscle hypothesis with rodent and cell data behind it, and nobody has shown it operates in neurons in any useful way. If you’re taking creatine for cognitive reasons, the water retention question is really a question about whether you mind a kilogram of intracellular water in your muscles, not a question about your head.
One practical note that applies to everyone: creatine is filtered by the kidneys, and while the safety literature in healthy people is strong and long running, anyone with kidney disease, anyone pregnant, or anyone on medication affecting fluid balance should talk with their doctor before supplementing. That’s not legal boilerplate. Individual medical situations genuinely change the calculus, and no article can see yours.
What Would Change This Picture
The state of the evidence on creatine and water retention is stronger than for almost any other supplement question, but it isn’t finished. A few things would meaningfully update it.
Larger trials in women, run by groups outside the current handful of labs, measuring fluid compartments across full cycles with dilution methods rather than impedance alone, would either cement or complicate the encouraging female data. Longer studies at the 3 to 5 gram maintenance dose, rather than loading protocols, would fill the gap between what researchers study and what people actually take. And any future trial finding a genuine extracellular water expansion from creatine would be a real surprise worth taking seriously, because after roughly three decades of measurement, nobody has produced one.
A trial of 3 to 5 grams daily over six months or more, with body water measured by a reference method like deuterium dilution rather than impedance alone, would tell us whether maintenance dosing produces any lasting water increase beyond what the added muscle accounts for. And a study that measured subcutaneous fluid directly, by imaging rather than whole body impedance, would finally address the question people are actually asking, which is whether creatine changes how you look.
Until then, the fair summary of creatine and water retention is this: yes, early and intracellular, especially with loading in men; small to absent at normal doses; proportional to muscle over time; and in women, a luteal phase rise in every fluid compartment, extracellular included, without a change in body weight. The bloating objection, examined study by study, turns out to be mostly a story about where people assumed the water went. The measurements say it went somewhere better.