Creatine: dangers and side effects, one by one

Creatine is the most studied supplement on the shelf, and the one surrounded by the most warnings: kidneys, liver, cramps, water, hair. Let's go through them one by one, with what trials have actually measured — including reserves, which do exist.

"It makes you retain water"

That is true, and it's the first thing you see on the scale. Five days at 20 g per day in 15 men and 15 women increase total body mass and fat-free mass — without affecting blood pressure, plasma creatinine or creatine kinase. After six weeks of continuous intake, the weight gained is still there, and it corresponds to an increase in fat-free mass.

What these trials measure is "fat-free mass" — a category that contains water. The authors of the second study say it themselves: water retention associated with creatine counts towards lean mass. It is therefore neither instant muscle nor fat: it is weight, and it was not accompanied by any rise in blood pressure.

"It gives you cramps and dehydrates you"

This is the most widespread fear, and the one that holds up the least. An entire review is dedicated to it: research has refuted the idea, and data point rather in the other direction — maintenance of hematocrit, aided therm thermoregulation, lowered heart rate and sweat rate during exertion in hot or humid conditions.

One trial put athletes to the heat test after a loading phase: 0.75 kg more body mass, unchanged urinary sodium, potassium and creatinine excretions, identical thermoregulation, no adverse effects reported — including cramps, and better performance in repeated sprints. A safety review classifies reports of cramps and digestive disorders as isolated signals, "not necessarily related to creatine".

"It's bad for the kidneys"

The benchmark meta-analysis reviewed 15 studies, of which 6 were quantitatively analyzable: neither serum creatinine nor plasma urea are significantly changed. In long-term consumers — from ten months to five years — plasma and urinary creatinine, urea and albumin are the same as in non-consumers. Three months at about 10 g per day in sedentary men produced no sign of renal dysfunction.

The most telling case is elsewhere: in type 2 diabetics, twelve weeks of creatine with training affected no renal parameter — neither the clearance measured with ⁵¹Cr-EDTA, nor proteinuria, nor albuminuria.

"And the liver?"

Much less discussed, and yet measured. Ten collegiate American football players took an average of 13.9 g per day for 2.9 years: albumin, alkaline phosphatases, ALT, AST, bilirubin, urea, creatinine and creatinine clearance do not differ from those of the control group. Another series, from five days at 20 g up to nine weeks at 3 g, leaves hematological indices and hepatic and renal markers within normal limits.

This same study incidentally explains the misunderstanding that fuels the fear of the kidneys: serum creatinine rises a little, but because more is produced — not because the kidney filters less well.

"It makes you lose your hair"

This fear is not based on a safety study but on a single trial that was heavily extrapolated, and it deserves its own article — this one, which takes the subject head-on, including the 2025 trial where someone finally looked at the hair.

How much, for how long

The International Society of Sports Nutrition position stand is the broadest reference: creatine is well tolerated in healthy subjects as in various patient populations, from infants to the elderly, and data cover up to 30 g per day for five years. This is not a dose recommendation: it is the extent of what has been observed without harm.

The 2019 review adds a useful point for those who eat little or no meat: vegetarians and vegans have lower muscle creatine concentrations, and it is in them that supplementation makes the biggest difference, on muscle performance as well as on neuropsychological tests.

What we don't know

It must be said too. A 2007 review notes isolated reports of effects on muscle, kidney and the digestive tract — the majority of clinical trials do not find them, but they exist in the literature. It points above all to what is missing: data on very long-term consequences are rare. The longest follow-up cited on this page is the 2.9 years of the university football players — and there were ten of them.

Every figure in this article comes from a study you can open: underlined mentions lead to the summary of the work from which it is taken, written from the abstract alone. Sample sizes are often small and follow-ups short — this is the main limitation of this literature, and it is stated above. Informative content, which does not replace the advice of a health professional: if you have kidney disease, the question arises with your doctor, not on the shelf.