FitProven
Evidence Breakdown
Based on 7 studies
THE BIGGER QUESTION
Whether this one statement holds is settled above. What to actually do about it is a wider question, weighed across every claim that bears on it.
Evidence map
For & against, at a glance
Case Study
Pro
Koshy KM et al. · 1999New England Journal of MedicineThe single strongest published signal in favour of the claim, and the origin of much of the alarm. A letter to the editor reporting a previously healthy young man who developed biopsy-confirmed acute focal interstitial nephritis attributed to creatine monohydrate; he recovered after stopping. It is an n=1 temporal association: later reviewers note that the patient's clinical and training history was omitted, leaving causation unestablished. Two further case reports sometimes cited alongside it are confounded — one patient had pre-existing focal segmental glomerulosclerosis and five years of cyclosporin, a known nephrotoxin; the other was taking numerous other supplements.
0.06
The single strongest published signal in favour of the claim, and the origin of much of the alarm. A letter to the editor reporting a previously healthy young man who developed biopsy-confirmed acute focal interstitial nephritis attributed to creatine monohydrate; he recovered after stopping. It is an n=1 temporal association: later reviewers note that the patient's clinical and training history was omitted, leaving causation unestablished. Two further case reports sometimes cited alongside it are confounded — one patient had pre-existing focal segmental glomerulosclerosis and five years of cyclosporin, a known nephrotoxin; the other was taking numerous other supplements.
Design Case Study (0.3) × quality 0.20 = impact 0.06
View sourceMeta-Analysis
Con
Kabiri Naeini E et al. · 2025BMC NephrologySystematic review and meta-analysis of studies published between January 2000 and March 2025: 21 studies reviewed, 12 pooled for serum creatinine (177 creatine vs 263 control) and 5 pooled for glomerular filtration rate (69 vs 74). Creatine produced a small but statistically significant rise in serum creatinine (MD 0.07 µmol/L; 95% CI 0.01-0.12; p = 0.03), while GFR — the measure that actually reflects filtration — showed no significant difference from control. The authors attribute the creatinine rise to metabolic turnover of the supplement itself rather than to renal impairment, and conclude that kidney function is preserved.
0.70
Systematic review and meta-analysis of studies published between January 2000 and March 2025: 21 studies reviewed, 12 pooled for serum creatinine (177 creatine vs 263 control) and 5 pooled for glomerular filtration rate (69 vs 74). Creatine produced a small but statistically significant rise in serum creatinine (MD 0.07 µmol/L; 95% CI 0.01-0.12; p = 0.03), while GFR — the measure that actually reflects filtration — showed no significant difference from control. The authors attribute the creatinine rise to metabolic turnover of the supplement itself rather than to renal impairment, and conclude that kidney function is preserved.
Design Meta-Analysis (1.0) × quality 0.70 = impact 0.70
View sourceDouble-Blind RCT
Con
Lugaresi R et al. · 2013Journal of the International Society of Sports NutritionRandomized, double-blind, placebo-controlled trial in 26 resistance-trained men (creatine n = 12, placebo n = 14) already eating a high-protein diet (at least 1.2 g/kg/day) — precisely the population in which harm is most often alleged. Crucially, GFR was measured over 12 weeks by 51Cr-EDTA clearance, a gold-standard tracer method that sidesteps the serum-creatinine artifact entirely rather than relying on an estimating equation. GFR was unchanged (creatine 101.4 ± 13.1 → 108.8 ± 14.4 vs placebo 103.3 ± 17.6 → 106.7 ± 16.1 mL/min/1.73 m²; group × time p = 0.64), as were urea, electrolytes, proteinuria and albuminuria.
0.43
Randomized, double-blind, placebo-controlled trial in 26 resistance-trained men (creatine n = 12, placebo n = 14) already eating a high-protein diet (at least 1.2 g/kg/day) — precisely the population in which harm is most often alleged. Crucially, GFR was measured over 12 weeks by 51Cr-EDTA clearance, a gold-standard tracer method that sidesteps the serum-creatinine artifact entirely rather than relying on an estimating equation. GFR was unchanged (creatine 101.4 ± 13.1 → 108.8 ± 14.4 vs placebo 103.3 ± 17.6 → 106.7 ± 16.1 mL/min/1.73 m²; group × time p = 0.64), as were urea, electrolytes, proteinuria and albuminuria.
Design Double-Blind RCT (0.85) × quality 0.50 = impact 0.43
View sourceCase Study
Con
Williamson L & New D · 2014BMJ Case ReportsA case of apparent kidney impairment that turned out to be a measurement artifact — the clearest illustration of why this myth persists. A 42-year-old bodybuilder taking creatine ethyl ester presented with serum creatinine of 198 then 227 µmol/L against a reference range of 60-120, numbers that would normally trigger a chronic-kidney-disease workup. On stopping the supplement his creatinine fell to 104 µmol/L, with no underlying kidney pathology found. The supplement had raised creatinine *production*, not reduced renal *clearance*. The paper's title states it plainly: "How the use of creatine supplements can elevate serum creatinine in the absence of underlying kidney pathology."
0.14
A case of apparent kidney impairment that turned out to be a measurement artifact — the clearest illustration of why this myth persists. A 42-year-old bodybuilder taking creatine ethyl ester presented with serum creatinine of 198 then 227 µmol/L against a reference range of 60-120, numbers that would normally trigger a chronic-kidney-disease workup. On stopping the supplement his creatinine fell to 104 µmol/L, with no underlying kidney pathology found. The supplement had raised creatinine *production*, not reduced renal *clearance*. The paper's title states it plainly: "How the use of creatine supplements can elevate serum creatinine in the absence of underlying kidney pathology."
Design Case Study (0.3) × quality 0.45 = impact 0.14
View sourceCase Study
Pro
Koshy KM et al. · 1999New England Journal of MedicineThe single strongest published signal in favour of the claim, and the origin of much of the alarm. A letter to the editor reporting a previously healthy young man who developed biopsy-confirmed acute focal interstitial nephritis attributed to creatine monohydrate; he recovered after stopping. It is an n=1 temporal association: later reviewers note that the patient's clinical and training history was omitted, leaving causation unestablished. Two further case reports sometimes cited alongside it are confounded — one patient had pre-existing focal segmental glomerulosclerosis and five years of cyclosporin, a known nephrotoxin; the other was taking numerous other supplements.
0.06
The single strongest published signal in favour of the claim, and the origin of much of the alarm. A letter to the editor reporting a previously healthy young man who developed biopsy-confirmed acute focal interstitial nephritis attributed to creatine monohydrate; he recovered after stopping. It is an n=1 temporal association: later reviewers note that the patient's clinical and training history was omitted, leaving causation unestablished. Two further case reports sometimes cited alongside it are confounded — one patient had pre-existing focal segmental glomerulosclerosis and five years of cyclosporin, a known nephrotoxin; the other was taking numerous other supplements.
Design Case Study (0.3) × quality 0.20 = impact 0.06
View sourceMeta-Analysis
Con
Kabiri Naeini E et al. · 2025BMC NephrologySystematic review and meta-analysis of studies published between January 2000 and March 2025: 21 studies reviewed, 12 pooled for serum creatinine (177 creatine vs 263 control) and 5 pooled for glomerular filtration rate (69 vs 74). Creatine produced a small but statistically significant rise in serum creatinine (MD 0.07 µmol/L; 95% CI 0.01-0.12; p = 0.03), while GFR — the measure that actually reflects filtration — showed no significant difference from control. The authors attribute the creatinine rise to metabolic turnover of the supplement itself rather than to renal impairment, and conclude that kidney function is preserved.
0.70
Systematic review and meta-analysis of studies published between January 2000 and March 2025: 21 studies reviewed, 12 pooled for serum creatinine (177 creatine vs 263 control) and 5 pooled for glomerular filtration rate (69 vs 74). Creatine produced a small but statistically significant rise in serum creatinine (MD 0.07 µmol/L; 95% CI 0.01-0.12; p = 0.03), while GFR — the measure that actually reflects filtration — showed no significant difference from control. The authors attribute the creatinine rise to metabolic turnover of the supplement itself rather than to renal impairment, and conclude that kidney function is preserved.
Design Meta-Analysis (1.0) × quality 0.70 = impact 0.70
View sourceDouble-Blind RCT
Con
Lugaresi R et al. · 2013Journal of the International Society of Sports NutritionRandomized, double-blind, placebo-controlled trial in 26 resistance-trained men (creatine n = 12, placebo n = 14) already eating a high-protein diet (at least 1.2 g/kg/day) — precisely the population in which harm is most often alleged. Crucially, GFR was measured over 12 weeks by 51Cr-EDTA clearance, a gold-standard tracer method that sidesteps the serum-creatinine artifact entirely rather than relying on an estimating equation. GFR was unchanged (creatine 101.4 ± 13.1 → 108.8 ± 14.4 vs placebo 103.3 ± 17.6 → 106.7 ± 16.1 mL/min/1.73 m²; group × time p = 0.64), as were urea, electrolytes, proteinuria and albuminuria.
0.43
Randomized, double-blind, placebo-controlled trial in 26 resistance-trained men (creatine n = 12, placebo n = 14) already eating a high-protein diet (at least 1.2 g/kg/day) — precisely the population in which harm is most often alleged. Crucially, GFR was measured over 12 weeks by 51Cr-EDTA clearance, a gold-standard tracer method that sidesteps the serum-creatinine artifact entirely rather than relying on an estimating equation. GFR was unchanged (creatine 101.4 ± 13.1 → 108.8 ± 14.4 vs placebo 103.3 ± 17.6 → 106.7 ± 16.1 mL/min/1.73 m²; group × time p = 0.64), as were urea, electrolytes, proteinuria and albuminuria.
Design Double-Blind RCT (0.85) × quality 0.50 = impact 0.43
View sourceCase Study
Con
Williamson L & New D · 2014BMJ Case ReportsA case of apparent kidney impairment that turned out to be a measurement artifact — the clearest illustration of why this myth persists. A 42-year-old bodybuilder taking creatine ethyl ester presented with serum creatinine of 198 then 227 µmol/L against a reference range of 60-120, numbers that would normally trigger a chronic-kidney-disease workup. On stopping the supplement his creatinine fell to 104 µmol/L, with no underlying kidney pathology found. The supplement had raised creatinine *production*, not reduced renal *clearance*. The paper's title states it plainly: "How the use of creatine supplements can elevate serum creatinine in the absence of underlying kidney pathology."
0.14
A case of apparent kidney impairment that turned out to be a measurement artifact — the clearest illustration of why this myth persists. A 42-year-old bodybuilder taking creatine ethyl ester presented with serum creatinine of 198 then 227 µmol/L against a reference range of 60-120, numbers that would normally trigger a chronic-kidney-disease workup. On stopping the supplement his creatinine fell to 104 µmol/L, with no underlying kidney pathology found. The supplement had raised creatinine *production*, not reduced renal *clearance*. The paper's title states it plainly: "How the use of creatine supplements can elevate serum creatinine in the absence of underlying kidney pathology."
Design Case Study (0.3) × quality 0.45 = impact 0.14
View sourceShowing the 4 strongest of 7 studies. Tap any node to expand its detail.
Evidence
PRO (1)
PRO Case Studyn=10.20 Koshy KM, Griswold E et al. (1999)
A single previously healthy young man developed biopsy-confirmed acute focal interstitial nephritis attributed to creatine monohydrate, and recovered after stopping it
The single strongest published signal in favour of the claim, and the origin of much of the alarm. A letter to the editor reporting a previously healthy young man who developed biopsy-confirmed acute focal interstitial nephritis attributed to creatine monohydrate; he recovered after stopping. It is an n=1 temporal association: later reviewers note that the patient's clinical and training history was omitted, leaving causation unestablished. Two further case reports sometimes cited alongside it are confounded — one patient had pre-existing focal segmental glomerulosclerosis and five years of cyclosporin, a known nephrotoxin; the other was taking numerous other supplements.
Weighted 0.20 — n=1 correspondence letter -- a temporal association only. The patient's clinical and training history was omitted from the report, so causation is unestablished and confounding cannot be excluded. No funding or conflict-of-interest statement exists (NEJM letters carried none in 1999), so funding is unrecorded rather than declared none. Weak even by case-report standards, yet it remains the single strongest published signal in favour of the claim, which is itself informative about the state of that evidence.
New England Journal of Medicine
AGAINST (6)
AGAINST Meta-Analysisn=4400.70 Kabiri Naeini E, Eskandari M et al. (2025)
Serum creatinine rose slightly (MD 0.07, 95% CI 0.01-0.12, p = 0.03) while glomerular filtration rate — the measure of actual filtration — did not differ from control
Systematic review and meta-analysis of studies published between January 2000 and March 2025: 21 studies reviewed, 12 pooled for serum creatinine (177 creatine vs 263 control) and 5 pooled for glomerular filtration rate (69 vs 74). Creatine produced a small but statistically significant rise in serum creatinine (MD 0.07 µmol/L; 95% CI 0.01-0.12; p = 0.03), while GFR — the measure that actually reflects filtration — showed no significant difference from control. The authors attribute the creatinine rise to metabolic turnover of the supplement itself rather than to renal impairment, and conclude that kidney function is preserved.
Weighted 0.70 — 21 studies reviewed but only 12 pooled for creatinine (440 participants; 177 creatine vs 263 control) and 5 for GFR (143 participants) — modest pooled samples. No funding and no competing interests declared, which is a strength. The outcomes are surrogate biochemical markers over mostly short trials in healthy people, so this cannot exclude harm in long-term use or in people with existing kidney disease.
Funding: none declared
BMC Nephrology
AGAINST Double-Blind RCTn=260.50 Lugaresi R, Leme M et al. (2013)
GFR by 51Cr-EDTA clearance unchanged over 12 weeks (creatine 101.4 to 108.8 vs placebo 103.3 to 106.7 mL/min/1.73 m2; group x time p=0.64); urea, electrolytes, proteinuria and albuminuria also unchanged
Randomized, double-blind, placebo-controlled trial in 26 resistance-trained men (creatine n = 12, placebo n = 14) already eating a high-protein diet (at least 1.2 g/kg/day) — precisely the population in which harm is most often alleged. Crucially, GFR was measured over 12 weeks by 51Cr-EDTA clearance, a gold-standard tracer method that sidesteps the serum-creatinine artifact entirely rather than relying on an estimating equation. GFR was unchanged (creatine 101.4 ± 13.1 → 108.8 ± 14.4 vs placebo 103.3 ± 17.6 → 106.7 ± 16.1 mL/min/1.73 m²; group × time p = 0.64), as were urea, electrolytes, proteinuria and albuminuria.
Weighted 0.50 — The gold-standard 51Cr-EDTA tracer measurement of GFR is the strength; the sample is the weakness. 46 were randomised but only 26 were analysed (12 creatine, 14 placebo) — 43% attrition — and the planned 12-month follow-up was cut to 12 weeks, with the authors stating the follow-up is too short to permit definitive conclusions. A small, underpowered null in healthy recreational lifters. Publicly funded by Brazilian agencies with no declared conflict.
Funding: FAPESP (Sao Paulo Research Foundation) and CNPq
Journal of the International Society of Sports Nutrition
AGAINST Case Studyn=10.45 Williamson L, New D (2014)
Serum creatinine 198 then 227 umol/L (reference 60-120) on creatine ethyl ester, falling to 104 umol/L after stopping, with no underlying kidney pathology found
A case of apparent kidney impairment that turned out to be a measurement artifact — the clearest illustration of why this myth persists. A 42-year-old bodybuilder taking creatine ethyl ester presented with serum creatinine of 198 then 227 µmol/L against a reference range of 60-120, numbers that would normally trigger a chronic-kidney-disease workup. On stopping the supplement his creatinine fell to 104 µmol/L, with no underlying kidney pathology found. The supplement had raised creatinine production, not reduced renal clearance. The paper's title states it plainly: "How the use of creatine supplements can elevate serum creatinine in the absence of underlying kidney pathology."
Weighted 0.45 — n=1. Within its class this is a strong case report — a clean positive dechallenge with no pathology found, illustrating a mechanism (raised creatinine production, not reduced renal clearance) that is independently established elsewhere. But it is still a single patient with no gold-standard clearance measure (no cystatin C or inulin), so it illustrates the artifact rather than demonstrating it.
BMJ Case Reports
AGAINST Narrative ReviewCONFLICTED0.40 Longobardi I, Gualano B et al. (2023)
controlled trials show no reduction in measured renal filtration; creatine degrades to creatinine at ~2% of the body pool per day, raising serum creatinine without impairing clearance
A review devoted to this exact question, titled "Is It Time for a Requiem for Creatine Supplementation-Induced Kidney Failure?". It concedes that isolated case reports and animal studies have suggested harm, but finds that controlled clinical trials do not support the claim, and that studies assessing filtration by reliable non-creatinine-based methods consistently find creatine safe. It examines the classic case reports individually and finds them confounded. It also documents the mechanism behind the false alarm: creatine spontaneously degrades to creatinine at roughly 2% of the body pool per day, so supplementing raises blood creatinine without reducing renal clearance — "a known confounder" that can produce a false-positive diagnosis. The authors flag one genuine gap: people with pre-existing kidney disease remain understudied.
Weighted 0.40 — Narrative (non-systematic) review, so no pooled sample and no formal risk-of-bias appraisal. Directly on topic and argued in detail, but the senior author declares research grants, creatine donations, honoraria and a Scientific Advisory Board seat with AlzChem, a creatine manufacturer — an industry-linked author reaching a favourable safety conclusion. The authors themselves flag scant data beyond 16 weeks and almost none in people with pre-existing kidney disease.
Funding: "This research received no external funding." But the conflict-of-interest statement discloses: "B.G. has received research grants, creatine donation for scientific studies, travel support for participation in scientific conferences, and honorarium for speaking at lectures from AlzChem (a company which manufactures creatine). Additionally, he serves as a member of the Scientific Advisory Board for Alzchem." — the funder profits from this result.
Nutrients
AGAINST Cross-Sectional0.30 Poortmans JR, Francaux M (1999)
No difference between long-term creatine users (self-reported 1-80 g/day for 10-60 months) and control athletes in plasma content, urinary excretion or clearance of creatinine, urea or albumin
Compared national- and international-level athletes who had used creatine long-term (reported as 1-80 g/day for 10-60 months) against control athletes, finding no differences in plasma content or urinary excretion of creatinine, urea or albumin. Read alongside the same group's short-term (20 g/day for 5 days) and medium-term (21 g/day for 5 days, then 3 g/day for 58 days) placebo-controlled work, the authors concluded that creatine causes no detrimental renal effects in healthy people. The weakest design among the studies here — cross-sectional, small, and reliant on self-selected users — so it corroborates the controlled trials rather than carrying the case on its own.
Weighted 0.30 — Group sizes are not reported in the accessible record. The weakest design in this literature — cross-sectional, small, self-selected users, no baseline and a self-reported exposure spanning an implausibly wide 1-80 g/day range, so healthy-user bias is unaddressed. It corroborates the controlled trials rather than carrying the case on its own. Funding could not be verified (1999, paywalled).
Medicine & Science in Sports & Exercise
AGAINST Expert OpinionCONFLICTED0.20 Kreider RB, Kalman DS et al. (2017)
Two positions. On performance: creatine monohydrate is 'the most effective ergogenic nutritional supplement currently available to athletes with the intent of increasing high-intensity exercise capacity and lean body mass during training', with high-intensity performance 'generally increased by 10-20%' after loading. On safety: no compelling evidence that up to 30 g/day for five years harms otherwise healthy people, and specifically no support for renal dysfunction -- creatinine clearance, glomerular filtration rate, tubular resorption and glomerular membrane permeability all unaffected
The ISSN position stand on creatine, which takes positions on two separate questions — which is why it is cited by more than one claim, with opposite polarity.
On performance. Position statement 1: creatine monohydrate is "the most effective ergogenic nutritional supplement currently available to athletes with the intent of increasing high-intensity exercise capacity and lean body mass during training." Position statement 5 says the same of the monohydrate form specifically. Quantitatively: "After creatine loading, performance of high intensity and/or repetitive exercise is generally increased by 10–20% depending on the magnitude of increase in muscle PCr." It also cites a meta-analysis of 357 older adults in whom creatine plus resistance training produced "greater gains in muscle mass, strength, and functional capacity" than training alone.
On safety. Position statement 2 holds that there is no compelling scientific evidence that short- or long-term creatine monohydrate use (up to 30 g/day for 5 years) has any detrimental effect on otherwise healthy individuals; on kidneys specifically it finds no support for renal dysfunction, citing an absence of effect on creatinine clearance, glomerular filtration rate, tubular resorption and glomerular membrane permeability.
Carried as expert opinion, not primary evidence: it is a consensus statement, it was funded by the supplement industry's trade association, and the author list is commercially entangled. Weigh it accordingly on both claims — including the one where its conclusion happens to be right.
Weighted 0.20 — Not a systematic review -- a consensus position stand with no inclusion criteria, no pooled sample and no risk-of-bias appraisal, so it carries no independent evidential weight (sample_size omitted by design). It was written at the REQUEST of, and funded by, the supplement industry's trade association, and the author list is deeply commercially entangled: the lead author is a university-approved advisor to a creatine seller (Nutrabolt), two are ISSN founders, and others include a supplement-company chief science officer, a supplement-company CEO, and multiple supplement patent-holders. An industry-funded safety reassurance is precisely the configuration warranting most caution -- though in this instance its conclusion is independently corroborated by trial evidence.
Funding: Council for Responsible Nutrition (the dietary-supplement industry trade association) — the funder profits from this result.
Journal of the International Society of Sports Nutrition