evipedia.ai
← Back

Creatine for Health & Longevity

Evidence Review created on 09/08/2026 using AI4L / Opus 5

Also known as: Creatine Monohydrate, Micronized Creatine, Creatine Hydrochloride, Creatine Ethyl Ester, Buffered Creatine, Kre-Alkalyn, Magnesium Creatine Chelate, Creatine Nitrate, Methylguanidoacetic Acid

Motivation

Creatine (methylguanidoacetic acid) is a small compound the body makes from three amino acids and also obtains from meat and fish. It is stored mainly in muscle, where it acts as a rapid-recharge battery for cellular energy, and in smaller amounts in the brain. Because the body’s own production and dietary supply both fall with age and with meat-free eating, interest has grown in whether topping up stores changes how people age.

First isolated in the 1830s and sold as a sports supplement from the early 1990s, creatine began as a performance aid for athletes. Attention has since widened to muscle and bone preservation in later life, to thinking and memory, and to mood, and it is now one of the most heavily studied supplements on the market.

This review examines what the evidence shows about creatine taken for general health and long-term function: where the human trial record is strong, where it is thin or contradictory, what the safety record looks like over years of use, and how the reported effects differ by age, sex, diet, and training status.

Benefits - Risks - Protocol - Conclusion

A short list of high-level overviews of creatine from expert practitioners and longevity-focused publications.

Content from five of the six priority platforms is listed. Chris Kresser’s site discusses creatine only in passing inside broader articles on nootropics (substances taken to sharpen thinking) and on weight gain, so nothing from that platform gave a high-level overview of creatine itself.

Grokipedia

  • Creatine

    Covers chemistry, the body’s own synthesis, dietary sources, supplement forms and the performance and clinical literature, with the safety controversies presented alongside the counter-evidence rather than only asserted.

Examine

  • Creatine

    Graded outcome database spanning 47 conditions with roughly 17,000 participants, plus dosing guidance and a safety section covering kidney cautions, drug interactions and product-quality findings.

ConsumerLab

  • Creatine Supplements Review

    Independent laboratory testing of creatine powders, capsules and gummies for label accuracy and contamination, with price-per-gram comparisons and identification of a product that failed testing.

Systematic Reviews

The pooled evidence on creatine, covering both its claimed benefits and its principal safety concern.

Mechanism of Action

Creatine is made in the liver, kidneys and pancreas from the amino acids glycine, arginine and methionine by two enzymes: AGAT (L-arginine:glycine amidinotransferase, which builds the precursor guanidinoacetate) and GAMT (guanidinoacetate N-methyltransferase, which adds a methyl group to finish the molecule). It enters muscle and brain through SLC6A8 (a sodium-dependent transporter protein that pulls creatine into cells). About 95% of the body’s 120–140 g pool sits in skeletal muscle, roughly two-thirds of it stored as phosphocreatine.

The core mechanism is energy buffering. Creatine kinase (an enzyme that moves phosphate groups between molecules) transfers a phosphate from phosphocreatine onto ADP (adenosine diphosphate, the spent form) to regenerate ATP (adenosine triphosphate, the molecule cells spend as immediate energy) within milliseconds. This sustains force during short, hard efforts and shortens recovery between them. The same shuttle operates in neurons, which is the proposed basis for effects on thinking and mood when brain energy demand is high, such as during sleep loss.

Secondary mechanisms are contested. One account attributes the extra muscle growth mainly to greater training volume plus cell swelling from water drawn into muscle, which is itself a growth signal. A competing account proposes direct effects on satellite cells (muscle stem cells), on muscle-building gene expression, and on reduced protein breakdown. Whether these add anything beyond the training-volume effect remains unresolved.

Historical Context & Evolution

Creatine was isolated from meat extract by the French chemist Michel Eugène Chevreul in 1832. Early twentieth-century work established that ingested creatine is retained in muscle. Its original applied use was not health optimization but the laboratory study of muscle energy metabolism, and later the treatment of rare inherited defects of creatine synthesis and transport, where it remains standard care for two of the three known defects.

The consumer era began around the 1992 Barcelona Olympics, after British researchers demonstrated that oral loading raises muscle creatine content and press reports linked several medallists to its use. Commercial powders followed within roughly a year.

Two developments then moved creatine from the gym toward the clinic. First, animal models of neurodegeneration showed protection of brain cells, which prompted large publicly funded trials. The Parkinson disease trial randomized 1,741 people to 10 g daily for at least five years and was halted for futility, with no difference in clinical decline (Kieburtz et al., 2015); the Huntington disease trial randomized 553 people to up to 40 g daily and was also halted, with decline marginally favouring placebo (Hersch et al., 2017). These null results are specific to progressive neurodegenerative disease at high doses and do not address the muscle and cognitive endpoints studied in healthy adults. Second, trials pairing creatine with resistance training in older adults reported repeatable gains in lean mass and strength, reframing it as a candidate countermeasure for sarcopenia (age-related loss of muscle mass and strength).

Expected Benefits

High 🟩 🟩 🟩

Increased Lean Body Mass

Creatine raises intramuscular water immediately and, over weeks of training, contractile protein, so body-composition scans record more fat-free tissue. A dose-response meta-analysis of 143 randomized controlled trials (studies in which participants are randomly assigned to the treatment or to a dummy treatment) found consistent gains, larger when a maintenance dose was paired with resistance training (Pashayee-Khamene et al., 2024). A separate 35-trial synthesis found the effect concentrated in resistance-training arms and larger in men (Delpino et al., 2022). Without training, the effect is negligible.

Magnitude: +0.82 kg fat-free mass (95% confidence interval, the range in which the true effect most likely lies, 0.57 to 1.06) and +0.86 kg body mass across 143 trials; +1.10 kg lean mass when creatine is combined with resistance training, versus +0.03 kg without exercise.

Greater Gains in Muscular Strength

More work per session — extra repetitions before failure and faster recovery between sets — compounds into larger strength gains than training alone. Meta-analyses in adults under 50 (Wang et al., 2024) and in older adults (Forbes et al., 2021) both report greater one-repetition maximum (the heaviest load that can be lifted once) improvements, clearest in leg press and chest press. Functional tests such as the 30-second chair stand also improve more with creatine (Devries & Phillips, 2014). Gains in women are smaller and less consistent.

Magnitude: +11.35 kg lower-body and +4.43 kg upper-body one-repetition maximum versus placebo in adults under 50; +7.5 kg leg press (95% confidence interval +2.2 to +12.8) in postmenopausal women.

Improved Capacity for Repeated High-Intensity Effort

Phosphocreatine is the limiting fuel for maximal efforts lasting under roughly 30 seconds, and larger stores shorten the recovery needed between bouts. This is the oldest and most replicated finding, established across sprint, jump and repeated-contraction protocols (Branch, 2003). It appears even at low doses without weight gain (Rawson et al., 2011). For this audience the practical value is tolerating higher training volumes, not competition; endurance performance is unaffected.

Magnitude: Fatigue resistance improved 7–11% across repeated maximal knee-extension sets at about 2.3 g/day; pooled effect size 0.24 for tasks under 30 seconds, versus a trivial standardized mean difference (effect expressed in units of variability) of −0.07 for endurance performance in trained athletes.

Medium 🟩 🟩

Cognitive Performance, Especially Memory ⚠️ Conflicted

Raising brain creatine by roughly 5–11% is proposed to support neurons when energy demand outstrips supply. A meta-analysis of 16 trials found benefits for memory, attention time and processing speed, with moderate certainty for memory (Xu et al., 2024); a separate 8-trial analysis found the effect concentrated in adults aged 66–76 (Prokopidis et al., 2023). A systematic review argues the mechanistic case is unmet and results are equivocal (McMorris et al., 2024). Net reading: a modest, replicable memory effect concentrated in older or energy-stressed brains, not a general nootropic.

Magnitude: Standardized mean difference 0.31 (95% confidence interval 0.18 to 0.44) for memory across the 16-trial analysis; in the separate 8-trial analysis the pooled figure is 0.29, splitting into 0.88 in adults aged 66–76 and near zero (0.03) in those aged 11–31.

Reduction of Depressive Symptoms as an Add-On ⚠️ Conflicted

Depression is associated with impaired brain energy metabolism, and creatine has been tested alongside serotonin-targeting antidepressants and talk therapy. A meta-analysis of 11 trials in 1,093 people found a statistically significant but small pooled effect that fell below the threshold considered clinically meaningful, with substantial bias favouring creatine and very low certainty (Eckert et al., 2025). Remission rates, however, favoured creatine strongly. Net reading: a plausible add-on signal, strongest for achieving remission, that is not yet established as clinically important.

Magnitude: Standardized mean difference −0.34 (95% confidence interval −0.68 to −0.00), equivalent to 2.2 points on a 17-item depression rating scale against a 3.0-point minimal important difference; odds ratio (the relative odds of an outcome between groups) 3.60 for remission across three trials.

Improved Blood Sugar Control Alongside Exercise

Creatine increases movement of GLUT-4 (the transporter that carries glucose from blood into muscle) to the muscle membrane, which should improve glucose disposal. In a 12-week randomized trial in people with type 2 diabetes who were also exercise-training, creatine lowered HbA1c (glycated haemoglobin, a marker reflecting average blood sugar over about three months) substantially more than placebo, with parallel falls in post-meal glucose (Gualano et al., 2011). The trial was small and has not been replicated at this size in people without diabetes.

Magnitude: HbA1c fell from 7.4% to 6.4% with creatine versus 7.5% to 7.6% with placebo, a between-group difference of −1.1% (95% confidence interval −1.9% to −0.4%).

Low 🟩

Preservation of Hip Bone Density ⚠️ Conflicted

A 12-month trial in postmenopausal women found creatine slowed femoral neck bone loss (Chilibeck et al., 2015). A larger 2-year trial in 237 women found no density effect, only improved bone geometry (Chilibeck et al., 2023). Net reading: density probably unchanged; geometry may improve.

Magnitude: Femoral neck loss of 1.2% versus 3.9% over 12 months in the smaller trial; no density difference at any site over 24 months in the larger one.

Lower All-Cause Mortality with Higher Dietary Intake

A national survey cohort followed 4,041 US adults for a median 19.8 years and found lower mortality among those eating at least 1 g of creatine daily (Ostojic, 2025). Observational and borderline; the signal weakened once lifestyle factors were added, so meat intake cannot be excluded.

Magnitude: Hazard ratio (the relative rate at which an event occurs between groups over time) 0.85 (95% confidence interval 0.72 to 1.00) for death from any cause, comparing intake of at least 1 g/day against less than 1 g/day.

Speculative 🟨

Slower Biological Aging on Epigenetic Clocks

Higher dietary creatine correlated inversely with two DNA-methylation mortality predictors in adults over 50 (Ostojic & Kavecan, 2025). The basis is cross-sectional correlation against an unvalidated surrogate, with very weak correlations and no supplementation trial.

Benefit-Modifying Factors

  • Baseline muscle creatine stores: The single largest modifier. People starting with low stores gain the most; those already near the ceiling of roughly 160 mmol/kg dry muscle show little change, which explains the well-described “non-responder” group in trial data.
  • Habitual diet: Vegetarians and vegans consume almost no dietary creatine and start with lower muscle and brain stores, so they show larger increases on supplementation and, in cognitive trials, the clearest gains.
  • Concurrent resistance training: Lean mass gains are roughly 1.10 kg with resistance training versus 0.03 kg without it. Without a training stimulus, most of the effect is water rather than contractile tissue.
  • Sex-based differences: Women have higher resting muscle creatine and lower total stores, and pooled analyses show smaller, often non-significant lean mass and strength gains than men, though bone geometry and mood effects are studied mainly in women.
  • Age: Older adults show the largest cognitive effect (memory effect size 0.88 in those aged 66–76 versus 0.03 in young adults) and the most clinically meaningful functional gains, because their baseline stores and muscle mass are already declining.
  • Genetic polymorphisms: Loss-of-function variants in SLC6A8 (the creatine transporter gene) blunt uptake into muscle and brain and are the mechanistic basis of creatine transporter deficiency; carriers of milder variants may respond poorly. GAMT and AGAT variants reduce endogenous synthesis and raise responsiveness.
  • Pre-existing health conditions: Conditions marked by impaired cellular energy metabolism — type 2 diabetes, depression, post-viral fatigue, sarcopenia — are where the largest effect sizes have been reported, whereas healthy young adults show the smallest incremental change.

Potential Risks & Side Effects

High 🟥 🟥 🟥

Gastrointestinal Distress

The most common complaint, driven by unabsorbed creatine drawing water into the gut. It is dose-dependent per serving rather than per day: a randomized trial in 59 athletes found diarrhoea significantly more frequent with a single 10 g dose than with the same 10 g split into two 5 g doses (Ostojic & Ahmetovic, 2008). Gastrointestinal events were also significantly more common on creatine in a 553-person, 4-year trial using doses up to 40 g daily (Hersch et al., 2017). Symptoms resolve on dose splitting or reduction.

Magnitude: Diarrhoea in 55.6% on a single 10 g dose versus 28.6% on two 5 g doses and 35.0% on placebo; total adverse events not increased at 3–5 g/day.

Increase in Body Mass from Water Retention

Creatine is osmotically active, so intramuscular water rises within days of starting, particularly with a loading protocol. This is not fat gain and largely reverses on discontinuation, but it matters for anyone in a weight-classified activity, monitoring scale weight as a goal, or managing fluid balance. A dose-response meta-analysis of 143 trials quantified the gain, and body fat percentage fell slightly rather than rising (Pashayee-Khamene et al., 2024). Low-dose protocols largely avoid it (Rawson et al., 2011).

Magnitude: +0.86 kg body mass (95% confidence interval 0.76 to 0.96) pooled across 143 trials, with body fat percentage −0.28%; no measurable weight gain at approximately 2.3 g/day over six weeks.

Elevated Serum Creatinine Misread as Kidney Damage

Creatine breaks down to creatinine, the very molecule used to estimate kidney function, so supplementation raises serum creatinine and lowers estimated filtration rate without any change in the kidney itself. A meta-analysis of 12 trials found a small significant rise in serum creatinine but no change in glomerular filtration rate, the measured rate at which kidneys filter blood (Naeini et al., 2025). The clinical consequence is misdiagnosis, unnecessary imaging and inappropriate discontinuation of other medicines.

Magnitude: Mean serum creatinine difference +0.07 µmol/L overall, largest within the first week of supplementation; no statistically significant change in glomerular filtration rate in any subgroup.

Medium 🟥 🟥

Switch to Mania in Bipolar Disorder

Creatine’s effect on brain energy metabolism appears able to destabilise mood in bipolar illness in the same direction as antidepressants. In a randomized, placebo-controlled trial of add-on creatine for bipolar depression, two of 17 participants receiving creatine switched to hypomania (a milder, shorter episode of elevated mood and activity) or mania early in the trial, with no such events on placebo (Toniolo et al., 2018). The sample was small and the finding has not been replicated in a larger trial, but the signal is consistent with the proposed mechanism.

Magnitude: 2 of 17 participants (12%) on 6 g daily switched to hypomania or mania within the first weeks, versus 0 of 18 on placebo.

Low 🟥

Kidney Injury in Established Kidney Disease or with Kidney-Toxic Drugs ⚠️ Conflicted

Case reports describe reduced filtration in people with pre-existing kidney disease or taking kidney-toxic drugs, with recovery after stopping. Controlled data find no renal damage in healthy people (de Souza e Silva et al., 2019). Net reading: no evidence of harm to healthy kidneys, unresolved risk in compromised ones.

Magnitude: Not quantified in available studies. No controlled trial has enrolled people with established kidney disease, so only isolated case reports exist for that group.

Hair Loss Attributed to Raised Dihydrotestosterone ⚠️ Conflicted

The claim rests on one study reporting raised dihydrotestosterone (the androgen that drives pattern hair loss) in rugby players (van der Merwe et al., 2009). A 12-week randomized trial measuring hormones and follicles found no effect (Lak et al., 2025). Net reading: the direct evidence does not support the claim.

Magnitude: No significant group-by-time difference in dihydrotestosterone, the dihydrotestosterone-to-testosterone ratio, hair density, follicular unit count, or cumulative hair thickness over 12 weeks at 5 g/day.

Muscle Cramping, Dehydration and Heat Intolerance ⚠️ Conflicted

A long-standing warning from sports-governing bodies, based on the reasoning that intramuscular water shifts deplete extracellular fluid. A systematic review with meta-analyses of exercise-heat trials found no impairment of hydration status or heat tolerance (Lopez et al., 2009). Net reading: the concern is not supported.

Magnitude: No significant differences in body temperature, sweat rate, heart rate, or cramping incidence during exercise in the heat; total body water rose in proportion to intracellular water rather than at its expense.

Speculative 🟨

Exposure to Contaminants and Under-Dosed Products

Retail surveys have found creatinine above European limits in over 40% of products, dicyandiamide and dihydro-1,3,5-triazine in buffered forms, near-zero creatine in some gummies, and one steroid-adulterated product. No human outcome study has measured harm.

Risk-Modifying Factors

  • Baseline kidney function: The strongest modifier. Normal filtration carries no demonstrated risk; reduced filtration, a single kidney, or a transplant shifts the risk-benefit balance and makes the creatinine artefact clinically confusing.
  • Baseline biomarker levels: A pre-supplementation serum creatinine and cystatin C pair distinguishes a supplement-driven creatinine rise from genuine decline, because cystatin C is unaffected by creatine intake or muscle mass.
  • Sex-based differences: Pooled adverse-event data in 951 women found no excess total adverse events, gastrointestinal events, weight gain, or kidney and liver abnormalities (de Guingand et al., 2020), so the profile is not worse in women.
  • Pre-existing health conditions: Bipolar disorder raises the risk of a mood switch. Kidney disease, and conditions treated with kidney-toxic drugs, concentrate the renal concern. Pregnancy and lactation lack human safety data entirely.
  • Age: Older adults are more likely to be taking kidney-toxic co-medication, to have reduced filtration reserve, and to have their creatinine used for drug dosing, so the misinterpretation risk rises with age even though direct toxicity does not.
  • Genetic polymorphisms: Variants reducing SLC6A8 transporter function lower cellular uptake, leaving more unabsorbed creatine in the gut and plausibly raising gastrointestinal intolerance while reducing benefit.

Key Interactions & Contraindications

  • Nephrotoxic (kidney-damaging) prescription drugs (ciclosporin, tacrolimus, aminoglycoside antibiotics such as gentamicin, cisplatin): Caution. Case reports describe reduced filtration when combined; mitigation is separating creatine from these courses and monitoring cystatin C rather than creatinine.
  • Non-steroidal anti-inflammatory drugs available over the counter (ibuprofen, naproxen, high-dose aspirin): Caution. Both reduce blood flow to the kidney; the clinical consequence is additive filtration decline during dehydration, and mitigation is renal monitoring during chronic combined use.
  • Diuretics (furosemide, hydrochlorothiazide): Caution. Volume depletion plus the intracellular water shift can concentrate the extracellular space; the consequence is dizziness and raised creatinine, and mitigation is maintained fluid intake with electrolyte rechecks.
  • Metformin: Monitor. Metformin is renally cleared and its dosing thresholds are keyed to estimated filtration rate, which creatine artificially lowers; the consequence is unnecessary dose reduction or withdrawal.
  • Caffeine (supplement or high-dose coffee, above roughly 300 mg): Caution. Some crossover studies show caffeine blunts creatine’s performance benefit, possibly via opposing effects on muscle relaxation time; mitigation is separating intake by several hours.
  • Serotonin-targeting antidepressants and lithium: Monitor. Creatine appears to strengthen antidepressant response; in bipolar illness the clinical consequence is a switch to mania, and mitigation is psychiatric supervision of any addition.
  • Statins (atorvastatin, simvastatin): Additive and potentially favourable. Statins may lower muscle creatine, and creatine has been used to reduce statin-associated muscle symptoms; no dose change is needed.
  • Supplements with additive effects: Additive, no restriction. Beta-alanine, HMB (beta-hydroxy-beta-methylbutyrate, a leucine metabolite studied for muscle preservation) and whey protein are co-administered in trials with additive lean mass effects; sodium bicarbonate is additive for repeated high-intensity work.
  • Other interventions: Potentiating, no restriction. Resistance training is the interaction that matters most, converting a largely water-driven mass change into contractile tissue; without it, most benefit does not materialise.

Populations who should avoid Creatine:

  • Chronic kidney disease at stage 3b or worse (estimated filtration rate below 45 mL/min/1.73 m²), single functioning kidney, or kidney transplant
  • Bipolar disorder type I or II, unless supervised by the treating psychiatrist
  • Pregnancy and lactation, where no reliable human safety data exist
  • Anyone taking a kidney-toxic drug whose dose is titrated against serum creatinine
  • People with a personal history of severe gastrointestinal intolerance to creatine at any dose

Risk Mitigation Strategies

  • No loading phase: Starting at 3–5 g daily reaches the same muscle saturation in about four weeks and avoids the diarrhoea and bloating that cluster around 20 g/day loading protocols.
  • Servings capped at 5 g: Diarrhoea rose from 28.6% to 55.6% when 10 g was taken as one serving rather than two, so 5 g portions taken with food prevent most gastrointestinal distress.
  • Baseline creatinine and cystatin C before starting: Recording both prevents the elevated-creatinine artefact from being misread as kidney injury and unrelated medicines being stopped unnecessarily.
  • Creatine declared on medication lists: Flagging it to prescribers prevents renally dosed drugs such as metformin being reduced or withdrawn on the basis of an artificially lowered filtration estimate.
  • Third-party-tested creatine monohydrate: Certification seals guard against the contaminants found in over 40% of surveyed products and against gummies shown to contain almost no creatine.
  • Higher fluid intake for the first two weeks: Roughly 500 mL extra daily offsets the intracellular water shift and reduces the bloating and transient weight gain that drive most discontinuation.
  • Pairing with resistance training: Lean mass gain without a training stimulus is roughly 0.03 kg, so unpaired use carries the water retention and the cost with almost none of the benefit.

Therapeutic Protocol

  • Standard maintenance dose: 3–5 g of creatine monohydrate daily, taken continuously. This is the dose used in most trials and endorsed in the International Society of Sports Nutrition position stand (Kreider et al., 2017).
  • Conflicts behind that guidance: The society issuing that position stand draws revenue from supplement-industry members and sponsors, and much of the creatine trial literature comes from the same small network of laboratories.
  • Competing approach — rapid loading: 0.3 g/kg daily in four doses for 5–7 days, then 3–5 g. Popularised by the Nottingham laboratories of Harris and Greenhaff; saturates muscle in a week rather than a month, with more gastrointestinal upset.
  • Competing approach — brain-directed dosing: 10–20 g daily, or 0.35 g/kg acutely. Advocated by Darren Candow’s group at Regina on the argument that 5 g raises muscle but barely moves brain creatine.
  • Time of day: No consistent advantage. Post-exercise dosing has a slight edge in some older-adult trials; on non-training days timing is irrelevant because saturation, not acute availability, drives the effect.
  • Half-life: Plasma half-life is roughly 1–3 hours, but muscle stores wash out over 4–6 weeks, so the practically relevant time constant is weeks, not hours.
  • Single versus split dosing: A single daily dose is adequate at 3–5 g. Above 5 g per serving, splitting into 5 g portions is the standard method of avoiding diarrhoea.
  • Genetic polymorphisms: SLC6A8 transporter variants reduce uptake and are the usual explanation offered for non-response; GAMT and AGAT synthesis variants raise dependence on dietary and supplemental intake. No routine pharmacogenetic testing is used.
  • Sex-based differences: Women show smaller pooled lean mass and strength responses and are studied more at higher relative doses; trials in postmenopausal women commonly use 0.1–0.14 g/kg daily rather than a flat 5 g.
  • Age: Older adults are the group in which combined creatine plus resistance training shows the clearest functional gains; loading followed by a low maintenance dose produced the largest strength effects in that population.
  • Baseline biomarker levels: Low baseline muscle creatine, typical of vegetarians and of low meat consumers, predicts the largest response; those already saturated respond minimally regardless of dose.
  • Pre-existing health conditions: Type 2 diabetes, depression and post-viral fatigue are the conditions in which supra-standard responses have been reported; kidney disease and bipolar disorder shift the calculus the other way.

Discontinuation & Cycling

  • Intended duration: Continuous, indefinite use. Trials have run up to five years at 10 g daily and up to four years at up to 40 g daily without cumulative toxicity emerging, and benefit depends on maintained saturation.
  • Withdrawal effects: None described. Muscle stores fall to baseline over roughly four to six weeks, with loss of the associated water weight and gradual reversal of the strength and mass advantage.
  • Tapering: Not required. There is no rebound below baseline, and the body’s own synthesis, which is partly suppressed during supplementation, recovers within weeks.
  • Cycling: Not supported for efficacy. Cycling simply desaturates muscle and forfeits the benefit; the practice originates from anabolic steroid protocols, where receptor downregulation makes cycling meaningful.
  • Training-day-only dosing: A meta-analysis found supplementing only on resistance-training days still significantly increased lean tissue mass and strength versus placebo (Forbes et al., 2021), making this a legitimate reduced-exposure option.

Sourcing and Quality

  • Preferred form: Creatine monohydrate. Hydrochloride, ethyl ester, buffered and nitrate forms are more expensive and none has outperformed monohydrate in head-to-head trials; ethyl ester and liquid serums degrade to creatinine.
  • Third-party testing: Products carrying NSF Certified for Sport or Informed Sport seals are the tested tier. Over 40% of 33 retail products surveyed exceeded the European limit of 100 mg creatinine per kg of creatine, a marker of poor manufacturing or storage.
  • Contaminants identified in testing: Dicyandiamide and dihydro-1,3,5-triazine have been detected above limits in buffered products, and one supplement was found adulterated with anabolic steroids.
  • Gummy formats: Independent testing found four of six popular creatine gummies contained virtually no creatine, alongside elevated creatinine, making the delivered dose unverifiable.
  • Micronization: Micronized monohydrate has smaller particles and dissolves more readily. It offers no bioavailability advantage but reduces the gritty residue that drives non-adherence.
  • Storage conditions: Creatine degrades to creatinine in solution and under heat and humidity, so dry storage below room temperature matters; pre-mixed bottles left standing and summer shipping both reduce potency.
  • Reputable manufacturers: Products carrying the Creapure designation, and brands regularly passing independent testing such as Thorne, Optimum Nutrition, Nutricost, Momentous and Sports Research, are the ones repeatedly identified in laboratory reviews.

Practical Considerations

  • Time to effect: Water-driven weight gain appears within 3–7 days. Muscle saturation takes about 28 days at 3–5 g daily, or 5–7 days with loading. Strength and lean mass differences require 8–12 weeks of training.
  • Time to effect for cognition: Memory effects have been reported after five days and after 24 weeks, with no consistent duration relationship; the acute sleep-deprivation effect appeared within three hours of a single large dose.
  • Common pitfall — expecting effects without training: Lean mass gain without exercise is essentially nil. Creatine amplifies a training stimulus rather than substituting for one.
  • Common pitfall — stopping over a creatinine result: A raised creatinine on routine bloodwork is expected and is not evidence of kidney damage; discontinuing on that basis forfeits the benefit for no reason.
  • Common pitfall — chasing exotic forms: Hydrochloride, nitrate and buffered products cost several times more per gram with no demonstrated advantage over monohydrate.
  • Regulatory status: In the United States creatine is a dietary supplement under the 1994 Dietary Supplement Health and Education Act, so it is not pre-approved for safety or efficacy by the Food and Drug Administration. It is not on the World Anti-Doping Agency prohibited list.
  • Cost and accessibility: Among the cheapest interventions in this field, typically well under USD 0.20 per day for bulk monohydrate. It is sold without prescription in most jurisdictions. Being unpatentable, it draws little industry- or payer-funded research.

Interaction with Foundational Habits

  • Sleep: Indirect and potentially compensatory. Creatine does not improve normal sleep, but a single 0.35 g/kg dose partially reversed the cognitive and brain-energy deterioration of 21 hours of sleep deprivation, with effects visible from three hours out. It does not substitute for sleep, and large evening doses may cause gastrointestinal discomfort at night.
  • Nutrition: Direct and dose-relevant. Roughly 1–2 g daily comes from meat and fish, so vegetarians start depleted and respond more. Co-ingestion with carbohydrate or a mixed meal raises muscle uptake through insulin-driven transporter activity, making dosing with food the practical default.
  • Exercise: Strongly potentiating and close to obligatory. Resistance training converts a 0.03 kg non-exercise lean mass change into roughly 1.10 kg. Timing around a session matters little; adherence across weeks matters greatly. Endurance training gains nothing, with a pooled effect of essentially zero in trained athletes.
  • Stress management: Indirect. Creatine buffers brain energy when demand is high, which is the proposed basis for benefits under sleep loss and in depression, but no trial has measured cortisol or a stress response as a primary endpoint. It does not replace stress-reduction practices.

Monitoring Protocol & Defining Success

Before starting, a baseline panel establishes the reference points that make later results interpretable: serum creatinine and cystatin C together, blood urea nitrogen, liver enzymes, fasting glucose and glycated haemoglobin, blood pressure, body weight, and a body-composition scan or bioimpedance reading. The creatinine and cystatin C pair matters most, because creatine intake raises the first and leaves the second untouched. A recorded strength benchmark and grip-strength measurement complete the baseline. Ongoing monitoring is light: repeat body weight weekly for the first month to separate water gain from fat gain, recheck the kidney and liver panel at 3 months, then every 6–12 months thereafter, with body composition and strength benchmarks reassessed at 3 months and then annually. Where kidney function is reduced or medication is kidney-cleared, that interval is typically shortened to every 3–6 months.

Biomarker Optimal Functional Range Why Measure It? Context/Notes
Serum creatinine 0.6–1.0 mg/dL (women), 0.7–1.1 mg/dL (men) Detects the expected supplement artefact and any real change Expected to rise slightly on creatine; conventional labs flag above 1.2 mg/dL as abnormal, which is misleading here. Fasting not required
Cystatin C 0.60–0.95 mg/L Kidney filtration marker unaffected by creatine or muscle mass The arbiter when creatinine rises. Pair with creatinine on the same draw
eGFR Above 90 mL/min/1.73 m² Tracks true filtration capacity over years eGFR is estimated glomerular filtration rate, a calculated measure of how fast the kidneys clear blood. The cystatin C-based equation is preferred over the creatinine-based one in creatine users
BUN 10–16 mg/dL Second-line kidney and protein-turnover check BUN is blood urea nitrogen, a protein waste product cleared by the kidneys. Unchanged by creatine in pooled trial data. Requires an 8–12 hour fast for the cleanest reading
HbA1c Below 5.4% Captures the glucose-disposal benefit reported with training Best paired with fasting insulin. No fasting is needed for this test itself
Fasting glucose 75–86 mg/dL Early signal of the metabolic effect before HbA1c moves 8–12 hour fast, morning draw. Best paired with fasting insulin
ALT 10–26 U/L Confirms no liver strain, the second most common unfounded concern ALT is alanine aminotransferase, an enzyme released when liver cells are stressed. Unchanged in pooled trial data. Morning draw, not after intense exercise, which raises it transiently
Body weight Rise of 0.5–1.5 kg in the first 2 weeks, then stable Distinguishes expected water retention from fat gain Same time of day, same conditions. A continuing rise past 4 weeks is not creatine water
Lean body mass No established target; track change from the individual’s own baseline The primary outcome creatine is taken for Dual-energy X-ray absorptiometry is the reference method; bioimpedance is acceptable if conditions are standardised. Measure fasted and before training
Grip strength No established target; track change from the individual’s own baseline, with below 27 kg (men) or 16 kg (women) marking clinical weakness Cheap, validated proxy for whole-body strength and function Best paired with a chair-stand count. Measure at the same time of day, rested
Blood pressure Below 120/80 mmHg Confirms fluid shift is not raising pressure Seated, rested 5 minutes. No creatine-driven rise has been reported

Qualitative markers matter as much as laboratory values, because most of what creatine changes is felt before it is measured.

  • Repetitions completed at a fixed load, and how quickly recovery occurs between sets
  • Perceived effort during the last set of a session compared with the first
  • Recall of names, lists and short sequences, particularly on poorly slept days
  • Mental clarity and processing speed during cognitively demanding work under fatigue
  • Ease of rising from a chair, climbing stairs, and carrying loads over distance
  • Absence of bloating, abdominal discomfort or loose stools after dosing
  • Mood stability, with any elevation, reduced sleep need or racing thoughts treated as a stop signal

Emerging Research

  • Muscle preservation during cancer treatment: NCT06112990 randomizes 200 men with metastatic prostate cancer to creatine plus resistance training or training alone, with total lean mass by body-composition scan at 52 weeks as the primary endpoint.
  • Cognition in mild cognitive impairment: NCT06948149 enrols 140 older adults with mild cognitive impairment for creatine plus resistance training, using visuospatial span and reaction time as primary outcomes — the population in which the memory signal should be largest.
  • Depression as a combination therapy: NCT05895747 is a phase 2 trial of 106 people with major depressive disorder combining creatine with 5-HTP (5-hydroxytryptophan, a serotonin precursor), measured on a 17-item depression rating scale.
  • Postoperative brain health: NCT07310043 will test creatine in 120 surgical patients for postoperative delirium and cognitive change, extending the sleep-deprivation and energy-stress rationale to a new stressor.
  • Recovery after joint replacement: NCT06925880 enrols 262 older adults after total knee arthroplasty, with muscle mass, strength and sarcopenia prevalence as primary endpoints.
  • Evidence that could weaken the case — cognition: A systematic review argues the brain-energy rationale is not met by the data and that most cognitive trials never measured brain creatine (McMorris et al., 2024); a review of older-adult evidence rated three of six studies methodologically poor (Marshall et al., 2026).
  • Evidence that could weaken the case — depression: The most rigorous synthesis to date found the pooled antidepressant effect below the clinically important threshold with substantial small-study bias (Eckert et al., 2025).
  • Evidence that could weaken the case — bone: The largest and longest bone trial found no effect on density at any site after two years, against a positive 12-month result (Chilibeck et al., 2023).
  • Open question — dose for the brain: Whether the 5 g dose that saturates muscle does anything meaningful for brain creatine is unresolved; the Alzheimer pilot needed 20 g daily to raise brain creatine by 11% (Smith et al., 2025).
  • Open question — mortality: Whether the observed association between dietary creatine and survival reflects creatine or the meat and diet quality it tracks can only be settled by a long-duration supplementation trial (Ostojic, 2025).

Conclusion

Creatine is a compound the body makes and also gets from meat and fish, stored mainly in muscle as a fast-recharge energy reserve. Taken daily at a few grams, it reliably adds a modest amount of muscle tissue and strength when paired with regular resistance training, and improves the ability to repeat short, hard efforts. Without training, most of what it adds is water. Beyond muscle, the record thins: memory gains show up mainly in older adults and in people whose brains are under energy stress, mood effects look real but small and uncertain, blood sugar findings rest on one small study, and the bone and survival evidence points in more than one direction at once.

The safety record is unusually long and unusually reassuring at ordinary doses, with digestive upset and a few pounds of water being the realistic costs. The one genuinely important hazard is not to the body but to interpretation: creatine raises the blood marker doctors use to judge kidney function, which invites false alarms. The groups where the evidence gives most pause are people with existing kidney disease, people with bipolar illness, and pregnancy, where human safety data are absent.

Much of this evidence comes from a small network of researchers and a sports-nutrition body whose members and sponsors sell the product. Because creatine is cheap and cannot be patented, neither manufacturers nor health systems carry a commercial incentive to study it, which is part of why the record looks the way it does.

Top - Benefits - Risks - Protocol