Creatine side effects: hair loss, kidneys, and headaches

Three worries follow creatine around: hair loss, kidneys, and headaches. They are not the same kind of evidence. Hair has one old hormone study and one newer trial that actually looked at hair. Kidneys have systematic reviews that find a creatinine change and not a filtration change. Headaches, in the search we ran, did not show up as a harm signal in healthy users.

Hair loss

The study that started the rumor is van der Merwe 2009 (PMID 19741313). Twenty college-aged male rugby players took creatine or placebo in a crossover. The creatine arm loaded at 25 g/day for 7 days, then 5 g/day for 14 days. Testosterone did not change. Dihydrotestosterone (DHT) rose 56% after the loading week and was still 40% above baseline after the maintenance fortnight (P < 0.001). The DHT-to-testosterone ratio rose 36%, then stayed 22% up (P < 0.01). The authors did not count hairs, photograph a scalp, or follow anyone long enough to see balding. They said further work was needed and that long-term safety of the hormone shift could not be guaranteed from this trial.

The direct hair trial is Lak 2025 (PMID 40265319, doi:10.1080/15502783.2025.2495229). Forty-five resistance-trained men aged 18 to 40 were randomized to 5 g/day creatine monohydrate or maltodextrin for 12 weeks. Thirty-eight finished. There was no group-by-time interaction for total testosterone, free testosterone, DHT, the DHT-to-testosterone ratio, or the hair measures (density, follicular unit count, cumulative hair thickness). Testosterone moved over the 12 weeks in both groups, and that change was not tied to which powder they took. This is the first trial we found that measured the follicle, and it does not support the claim that 5 g/day for 12 weeks thins hair in young men.

It also does not close the question. It is one trial, in men, for 12 weeks, at a maintenance dose rather than the 25 g load used in 2009. It was not done in women, in people who already have androgenic hair loss, or over years. Several authors advise creatine brands or the International Society of Sports Nutrition. Read it as a failed replication of the hair claim in that group, not as a lifetime guarantee.

Kidneys

Creatine is broken down to creatinine. Creatinine is also the number clinics use to estimate kidney filtration, so a supplement can move the lab result without moving the kidney.

Naeini 2025 (PMID 41199218, doi:10.1186/s12882-025-04558-6) reviewed studies from 2000 through March 2025. Twenty-one studies were in the review. Twelve (177 people on creatine, 263 controls) were in the creatinine meta-analysis. Serum creatinine rose by a mean difference of 0.07 µmol/L (95% CI 0.01 to 0.12). The rise was significant in studies of a week or less and again in studies longer than 12 weeks, and not in the 1-to-12-week band. Glomerular filtration rate was not significantly different. The authors call the creatinine change modest and transient, from turnover rather than kidney damage. We are quoting the unit as the abstract prints it.

de Souza 2019 (PMID 31375416) put 6 trials in the numbers. The abstract says creatine "did not significantly alter" serum creatinine and then prints a standardized mean difference of 0.48 (95% CI 0.24 to 0.73, P = .001) for creatinine and 1.10 (0.34 to 1.85, P = .004) for urea. Those p values are significant. The conclusion sentence is that creatine did not induce renal damage at the amounts and durations studied. Believe the conclusion only as far as "damage was not shown." Do not believe the words "did not significantly alter" when the same paragraph gives P = .001.

In the postmenopausal review (Naddafha 2026, PMID 42141930), adverse events were mild and similar to placebo, and renal indices were unchanged. That is a narrower population than "everyone."

None of these reviews is a trial in people who already have kidney disease. If your creatinine is being watched for a reason, tell the clinician you take creatine before anyone treats a higher number as a new diagnosis.

Headaches

On 21 October 2026 a PubMed search for creatine supplementation and headache did not return a trial of creatine causing headaches in healthy people. The two papers that matched were small studies in long COVID, where headache was one of the illness symptoms that improved, not a side effect that appeared.

• Slankamenac 2024 (PMID 38684388): 15 adults with long COVID, three groups, 8 weeks. Creatine 8 g/day, creatine 8 g plus 3 g glucose, or 3 g glucose. In the creatine-plus-glucose group, several symptoms including headache were lower at 8 weeks. This is a preliminary symptom report, not a headache treatment trial and not a safety signal that creatine causes headache.
• Slankamenac 2023 (PMID 37970399): 12 people with post-COVID fatigue, 4 g/day for 6 months versus placebo. Headache was again one of several illness symptoms that improved at 6 months. Twelve people cannot prove a side-effect rate either way.

The postmenopausal review said adverse events were mild and similar to placebo. It did not publish a headache count. We are not going to invent one.

Questions people ask

Does creatine cause hair loss?

A 2009 study of 20 male rugby players found DHT rose after a creatine load and did not measure hair. A 2025 trial of 5 g/day for 12 weeks in young men found no difference from placebo in DHT or hair-follicle measures. That trial does not cover women, people who already have pattern hair loss, or use over many years.

Does creatine damage kidneys?

Reviews find a small rise in serum creatinine and no significant change in filtration rate. Creatinine is a lab marker that creatine itself can move. These reviews are not trials in people who already have kidney disease.

Does creatine cause headaches?

A PubMed search on 21 October 2026 did not find a trial of creatine causing headaches in healthy users. The matching papers were small long-COVID studies in which headache, as a symptom of that illness, was reported to improve. That is not a rate, and it is not proof creatine treats ordinary headaches.