MuscleLedger

Too Much Creatine: What Trials Gave at 20 to 40 g a Day, and What Went Wrong

What happened in the trials that gave the largest amounts of creatine: 20 g a day for loading, 30 g a day in ALS, up to 40 g a day for four years in Huntington disease. What was measured, what went wrong, and what is not established.

Leon H · Edited by Caroline S · Published 2026-11-10

Illustration: An empty, unbranded creatine tub on a damp concrete floor with a sweaty towel in dawn light.
Illustration

"Too much creatine" can mean two things: an amount that causes a problem, or an amount past which nothing more happens. The trial record answers the second question better than the first. It contains studies that gave far more than the 3 to 5 g a day on most tubs, some for years, and what they measured is set out below with the dose, the people and the result.

The first-screen answer

Trial Who took part Creatine dose How long What went wrong
Ostojic and Ahmetovic, 2008 Young soccer players 10 g in one serving, against 2 × 5 g and placebo 28 days Diarrhoea in 55.6% on the single 10 g serving, 28.6% on 2 × 5 g, 35.0% on placebo
Hultman et al., 1996 31 healthy men 20 g a day for 6 days, against 3 g a day for 28 days Up to 36 days Not a safety trial; muscle creatine rose about 20% on both schedules
Atassi et al., 2010 6 people with ALS 5, then 10, then 15 g twice a day 3 weekly steps Authors report no safety concerns at any dose
NET-PD LS-1, 2015 1,741 people with early Parkinson disease 10 g a day At least 5 years (median follow-up 4 years at the interim analysis) No detectable difference in adverse or serious adverse events by body system; stopped for futility
CREST-E, 2017 553 people with early Huntington disease Up to 40 g a day Up to 48 months Gastrointestinal adverse events significantly more common on creatine; serious adverse events, including deaths, more frequent on placebo; stopped for futility

The doses are as the trials printed them. The table is a record of what was given and measured, not a statement of what anyone should take.

The one dose-linked effect the trials show: single large servings and the gut

The clearest finding that a larger amount causes a problem is about how it is taken, not the daily total. In the 2008 soccer trial, players took either one 10 g serving a day or two 5 g servings for 28 days. Diarrhoea was reported by 55.6% of the single-serving group and 28.6% of the split group. The placebo group's rate, 35.0%, shows how common the complaint was anyway in this population, and the trial was small. The authors' reading is that the risk of diarrhoea rises when 10 g is taken in one serving.

The Huntington trial points the same way at much larger daily amounts. Its creatine arm, at up to 40 g a day, had significantly more adverse events than placebo, and the authors describe them as mainly gastrointestinal. The site's creatine and diarrhoea page sets out the gut reports in more detail, and the bloating evidence covers the water and fullness complaints.

Past a point, the muscle store does not take more

The 1996 biopsy study is the reason loading phases exist, and it is also the best evidence that more creatine does not mean more stored creatine. Twenty grams a day for six days raised muscle total creatine by about 20%. Three grams a day for 28 days raised it by a similar 20%, more slowly. Two grams a day held the raised level for a further 30 days, and when supplementation stopped the level drifted back to baseline within about a month. A larger daily amount got to the ceiling sooner. It did not move the ceiling.

What happens to the surplus is the subject of how much creatine the trials gave, which sets the 20 g, 3 to 5 g and trial-average figures side by side with their sources. The loading phase page has the day-by-day schedule from the biopsy trials.

The long, high-dose disease trials

The largest amounts given for the longest time were not given to athletes. Two neurology trials stand out.

Parkinson disease, 10 g a day. The NET-PD Long-term Study 1 randomised 1,741 people with early, treated Parkinson disease to creatine or placebo for at least five years, with up to eight years of follow-up. It was stopped early for futility. The paper reports no detectable differences in adverse or serious adverse events by body system, using a significance threshold set to partly adjust for multiple comparisons.

Huntington disease, up to 40 g a day. CREST-E randomised 553 people with early symptoms to creatine or placebo at 46 sites, for up to 48 months. It was also stopped for futility: decline was slightly faster on creatine (0.82 points a year on the functional scale) than on placebo (0.70), a difference whose confidence interval crossed zero. Gastrointestinal adverse events were significantly more common on creatine. Serious adverse events, including deaths, were more frequent on placebo.

ALS, 30 g a day. The 2010 phase I study escalated six people through 5, 10 and 15 g twice a day in weekly steps. Plasma creatine rose in step with the dose (20.3, 39.3 and 61.5 µg/ml before the morning dose) and brain-scan creatine signals rose by 8% at the higher doses. The authors report no safety concerns. Six people over three weeks can show tolerability in the short term; it cannot show long-term safety.

Two limits apply to all three. The participants had a neurological disease and were monitored in a trial, which is not the same population as healthy people taking supplements on their own. And a trial stopped for futility has fewer person-years of exposure than planned.

What the pooled trial record says about dose

Two analyses from the same Texas A&M group pooled the controlled trials. The 2025 analysis counted side effects across 685 trials with 12,839 people on creatine and 13,452 on placebo. Nearly all the trials gave creatine monohydrate, at an average of 0.166 g per kg of body weight a day. Side effects were reported in 13.7% of creatine arms and 13.2% of placebo arms (Kreider et al., 2025). A 2026 follow-up grouped the 684 trials by total dose and duration and reported no consistent increase in side effects at higher doses or longer use, with effect sizes it calls uniformly small (Gonzalez et al., 2026).

Both are study-level analyses: they count whether a trial reported a side effect, not each person's dose and outcome. Several authors on both papers disclose industry-sponsored research or advisory roles with supplement companies. The 2017 ISSN position stand, also from overlapping authors, states that supplementation "up to 30 g/day for 5 years" is safe and well tolerated (Kreider et al., 2017).

What the FDA's report database can and cannot show

The FDA's adverse-event database for foods and supplements is the place where a pattern of overdoses would surface if one existed. Read through openFDA on 10 November 2026 (data last updated 7 July 2026), it holds 210 reports naming a creatine-branded product. This page's own reading of those 210:

What was looked for Reports
Reaction coded as overdose, accidental or intentional overdose, or excessive dose 0
Reaction coded as "poisoning" 2 (both 2016, one product each, no amount recorded)
More than one product named in the report 117
A field recording how much was taken none exists

Zero overdose codes is a real finding, but it is not reassurance. The database records no amount, so it cannot say whether any report followed a large intake. More than half the reports name other products as well. The 2025 pooled analysis found that 46.3% of the reports it examined had no creatine in the listed products at all. How the full set of 210 reads, outcome by outcome, is on creatine side effects.

Kidneys and the blood-test problem

The most common worry about large amounts is kidney damage. Two kinds of evidence exist. Case reports describe kidney problems in people taking creatine, often with other supplements or existing kidney disease: a 1998 Lancet letter titled "Renal dysfunction accompanying oral creatine supplements" (Pritchard and Kalra) and a 2006 report of acute renal failure in a weightlifter taking several supplements including creatine (J Ren Nutr 2006). Case reports cannot show cause. The controlled trials, summarised on creatine and kidneys, measured kidney function in healthy people and found no change beyond the expected one.

That expected change matters. Creatine breaks down to creatinine, the marker most kidney blood tests use. More creatine raises blood creatinine without any change in kidney function, so the test can read worse than the kidneys are working. The site's creatine in bloodwork page explains the artefact and the cross-check clinicians use.

What is not established

  • An upper limit. No regulator or position stand on this page publishes a tolerable upper intake for creatine. The figures above are amounts trials gave, not ceilings.
  • Years at 20 g or more in healthy people. The long high-dose trials enrolled people with neurological disease. Healthy lifters taking these amounts for years have not been studied as a group.
  • Large amounts in people with kidney disease. Most trials excluded them.
  • Whether any FDA report involved a large intake. The database cannot say.

Who should ask a clinician first

People with kidney disease or reduced kidney function, people taking medicines that act on the kidneys (set out on how creatine fits with prescriptions), and people who are pregnant or breastfeeding are the groups these trials either excluded or did not study. Anyone with a kidney blood test due should tell the clinician about creatine use, because of the creatinine artefact above. The trials on this page report amounts and outcomes; they do not say what anyone should take.

Sources and dates

  • Hersch SM, Schifitto G, Oakes D, et al. The CREST-E study of creatine for Huntington disease: a randomized controlled trial. Neurology. 2017;89(6):594-601. PubMed 28701493; NCT00712426. Opened 2026-11-10.
  • Writing Group for the NINDS Exploratory Trials in Parkinson Disease (NET-PD) Investigators; Kieburtz K, et al. Effect of creatine monohydrate on clinical progression in patients with Parkinson disease. JAMA. 2015;313(6):584-593. PubMed 25668262; NCT00449865. Opened 2026-11-10.
  • Atassi N, Ratai EM, Greenblatt DJ, et al. A phase I, pharmacokinetic, dosage escalation study of creatine monohydrate in subjects with amyotrophic lateral sclerosis. Amyotroph Lateral Scler. 2010;11(6):508-513. PubMed 20698808. Opened 2026-11-10.
  • Hultman E, Söderlund K, Timmons JA, Cederblad G, Greenhaff PL. Muscle creatine loading in men. J Appl Physiol. 1996;81(1):232-237. PubMed 8828669. Opened 2026-11-10.
  • Ostojic SM, Ahmetovic Z. Gastrointestinal distress after creatine supplementation in athletes: are side effects dose dependent? Res Sports Med. 2008;16(1):15-22. PubMed 18373286. Opened 2026-11-10.
  • Kreider RB, Gonzalez DE, Hines K, Gil A, Bonilla DA. Safety of creatine supplementation: analysis of the prevalence of reported side effects in clinical trials and adverse event reports. J Int Soc Sports Nutr. 2025;22(sup1):2488937. PubMed 40198156. Opened 2026-11-10.
  • Gonzalez DE, Dickerson BL, Hines K, et al. Creatine supplementation dose and duration are not associated with increased side effects. Sports (Basel). 2026;14(4):137. PubMed 42043069. Opened 2026-11-10.
  • Kreider RB, Kalman DS, Antonio J, et al. International Society of Sports Nutrition position stand: safety and efficacy of creatine supplementation in exercise, sport, and medicine. J Int Soc Sports Nutr. 2017;14:18. PubMed 28615996. Opened 2026-11-10.
  • Pritchard NR, Kalra PA. Renal dysfunction accompanying oral creatine supplements. Lancet. 1998;351(9111):1252-1253. PubMed 9643752 (letter; no abstract — cited by title only). Opened 2026-11-10.
  • Acute renal failure in a young weight lifter taking multiple food supplements, including creatine monohydrate. J Ren Nutr. 2006. PubMed 17046619 (record read; cited by title only). Opened 2026-11-10.
  • FDA CFSAN Adverse Event Reporting System via openFDA, products.name_brand:creatine, 210 reports, data last updated 2026-07-07, read 2026-11-10. Reaction terms and product counts tallied for this page.
  • Trial abstracts were read on PubMed. Doses, sample sizes and results are as printed there.

Frequently asked questions

How much creatine is too much?

No regulator or position stand publishes a single upper limit for creatine. The trials give a range: 3 to 5 g a day for maintenance, 20 g a day for five to seven days of loading, and 10 to 40 g a day in long disease trials. The clearest dose-linked problem is diarrhoea when 10 g or more is taken in a single serving. What a given person can tolerate, especially with kidney disease, is a question for a clinician.

Can you overdose on creatine?

No published human report describes a creatine overdose as a distinct poisoning with its own pattern, and none of the 210 FDA adverse-event reports naming a creatine product uses an overdose term. That is not proof that very large amounts are harmless: the database does not record the amount taken, and most of its reports name other products as well.

What happens at 20 g of creatine a day for longer than a loading phase?

Long trials of 10 to 40 g a day exist, but in disease populations: 10 g a day for at least five years in 1,741 people with Parkinson disease, and up to 40 g a day for up to four years in Huntington disease. Neither found a pattern of serious harm attributable to creatine; the Huntington trial found more gastrointestinal adverse events on creatine. Healthy gym users at these amounts for years have not been studied as a group.

Does extra creatine build more muscle?

The biopsy work says the muscle store has a ceiling. In the 1996 study, 20 g a day for six days and 3 g a day for 28 days ended at a similar rise of about 20% in muscle total creatine. A larger amount got there sooner. When supplementation stopped, levels returned to baseline within about 30 days.

Who should ask a clinician before taking large amounts of creatine?

Anyone with kidney disease or reduced kidney function, anyone taking a medicine that affects the kidneys, and anyone who is pregnant or breastfeeding. Creatine also raises blood creatinine, which can make a kidney blood test read worse than kidney function is; a clinician who knows about the supplement can read the result correctly.