Creatine: From a Meat Extract in 1832 to the Sports Shelf

Creatine was identified in meat in 1832 and synthesized in a lab in 1868. None of the Eastern texts discussed here names it. Those texts sorted flesh as food and praised bird flesh as a tonic, and the modern trials measured narrower things than the shelf suggests.

Creatine was in meat long before anyone isolated it. The chemist credited with the discovery is Michel Eugène Chevreul, who found it in 1832 as a component of the bouillon tablets of a Dutch company. That date comes from an 1868 paper by J. Volhard that Liebig presented to the Bavarian Academy of Sciences, and a historical review published by Firenze University Press also dates the extraction to 1832. The sports case is much newer, and it rests on a narrower set of trials than most shelf copy admits. This page covers both halves, and it gives the Eastern traditions their own section, because that is where readers most often expect a longer story. ## Where creatine comes from Your body makes some creatine from the amino acids arginine, glycine and methionine, and the rest comes from food. The UK Nutrition and Health Claims Committee notes that creatine occurs naturally in animal-derived food, in particular meat and fish. In muscle and brain, creatine is phosphorylated to phosphocreatine, which supplies high-energy phosphate to ADP to make ATP ([UK Nutrition and Health Claims Committee, 2024](https://www.gov.uk/government/publications/uknhcc-scientific-opinion-creatine-supplementation-and-improved-cognitive-function/uknhcc-scientific-opinion-creatine-supplementation-and-improved-cognitive-function)). According to PubMed, a 1994 review by Balsom and colleagues at the Karolinska Institute estimates that daily turnover of creatine to creatinine in a 70 kilogram man is about 2 grams, and that part of that turnover can be replaced by creatine in food, especially meat and fish ([DOI](https://doi.org/10.2165/00007256-199418040-00005)). The scale is easy to miss. In 1992, Harris, Söderlund and Hultman, also at the Karolinska Institute, noted that a single 5 gram dose of creatine corresponds to the creatine in 1.1 kilograms of fresh, uncooked steak. According to PubMed, that comparison appears in the abstract of their paper in Clinical Science ([DOI](https://doi.org/10.1042/cs0830367)). Keep that number in mind the next time a label promises a dose. ## What the Eastern texts recommended Here is the plain answer first. None of the Eastern texts discussed below names creatine, because the compound was not identified until 1832. What those traditions did with flesh is a different question, and the passages below answer it, with dates where the sources give them. ### Ayurveda The Charaka Samhita is an Ayurvedic treatise. The English translation, published in 1949 by the Shree Gulabkunverba Ayurvedic Society, is posted on wisdomlib ([Charaka Samhita, English translation](https://www.wisdomlib.org/hinduism/book/charaka-samhita-english/d/doc627515.html)). The date given here is the translation's. The fifth chapter of the general-principles section, "Measure in eating," sorts foods by weight and by habit. The flesh of aquatic and wetland animals is heavy, so it should be taken in measure. The flesh of an emaciated animal should not be used, and pork, cow's and buffalo's flesh and fish should not be taken habitually. The flesh of jangala animals, the translation's term, may be taken habitually, along with honey. The chapter also says the daily diet "serves as a prophylactic against coming disease." That is the text's claim. The chapter does not prescribe broth. It treats flesh as food to eat by measure and by habit. ### Traditional Chinese medicine The sixteenth-century materia medica that Penelope Barrett analyses in Medical History is Li Shizhen's Bencao gangmu. Barrett's 2005 article, "Cooking up Fine Remedies," quotes the entry for the flesh of the quail. The entry calls the flesh a tonic for the internal organs. It says the flesh strengthens bones and muscles and helps the body endure cold and heat, and that it relieves inflammation. It also gives a combination with ginger and red mung bean for diarrhoea and dysentery, a warning against eating it before May, and a note that it is given for chronic disorders of children ([DOI](https://doi.org/10.1017/S0025727300009133)). This entry is about flesh, not broth, and it describes a bird, not creatine. Barrett's title places the work in the sixteenth century. ### Korean medicine Dongui Bogam, which a 2018 conference paper renders as Principles and Practice of Eastern Medicine, was compiled by Heo Jun and other literati over about fifteen years. The same paper, by Liu, Gu, Mao and Wu and published by Atlantis Press, gives 1610 as the completion date. It says two full sets of the first print were registered on UNESCO's Memory of the World Register on 31 July 2009 ([Liu et al., 2018](https://download.atlantis-press.com/article/25896921.pdf)). The work has five parts and 104 items, and it draws on Chinese classics such as the Plain Questions, the Spiritual Pivot and the Treatise on Exogenous Febrile Diseases. ### What the traditions did with flesh In the passages discussed above, Eastern medicine treated flesh as food to be sorted by quality and habit, and it praised at least one bird's flesh as a tonic. None of these passages describes isolating a compound from flesh, and none points toward creatine. The Eastern thread in this article is therefore about meat as food, not about creatine. ## Who found creatine, and when [Volhard's 1868 paper](https://publikationen.badw.de/en/003393259/003393259.pdf) is a primary source for the early history, and it sets out the key steps. Chevreul extracted the substance from meat in 1832 and named it, and a historical review published by Firenze University Press gives the same year ([Firenze University Press review](https://media.fupress.com/files/pdf/16/4380/4140_26026)). Several researchers had tried and failed to isolate the same substance from meat broth. Liebig then described a reliable method in his study of meat liquids, published in volume 62 of the Annalen der Chemie und Pharmacie, which Volhard cites. Liebig showed that creatine is a constant part of the meat liquid of many animals and of human urine. Pettenkofer had been the first to prepare it from urine. Volhard then made a guess that matters for this subject. He compared creatine and creatinine with theobromine and theine, the stimulants in chocolate and tea. He then wrote that it is not improbable that the nerve-strengthening power of a good meat broth rests partly on its content of these two substances. The translation is ours. Volhard offered the idea as a likelihood. None of the trials cited in this article tested broth itself. He also recorded a chemical detail that later mattered. When creatine turns into creatinine, it loses one molecule of water. ## How the name came about Volhard credits Chevreul with the name. The Firenze University Press review also says Chevreul named the substance from the Greek word for meat, kreas (κρέας). ## How creatine was made, then and now Volhard's synthesis is the earliest lab route in this history. He combined sarcosine and cyanamide, both in alcohol, and held the mixture at the temperature of boiling water for several hours. On cooling he got crystals that matched natural creatine in composition and properties. From 20 grams of sarcosine and 10 grams of cyanamide he obtained about 4 grams of creatine. He called the method costly and laborious and did not recommend it for larger amounts. The sarcosine itself came from monochloroacetic acid and methylamine, by a method Volhard had described. The industrial record starts from the same two ingredients. US Patent 5,719,319 was filed by SKW Trostberg AG in 1996 and published in 1998. It describes reacting cyanamide with sodium or potassium sarcosinate, in water or in a mixture of water and an organic solvent, at 20 to 150 degrees Celsius and a pH of 7.0 to 14.0 ([patent](https://patents.google.com/patent/US5719319)). The patent claims yields of 60 to 90 percent by weight and "very high purity." Google Patents lists Alzchem Trostberg GmbH as the current assignee, and that company is the applicant on the EU and UK claim files discussed below. The patent describes process chemistry on record. It does not describe how any plant runs today. ## From lab curiosity to the sports shelf The sports story began with muscle chemistry, not with marketing. In 1992, Harris, Söderlund and Hultman gave people 5 grams of creatine monohydrate four or six times a day for at least two days. According to PubMed, total creatine in the quadriceps rose in 17 subjects, by as much as 50 percent in some. About a fifth or more of the creatine taken up showed up as phosphocreatine, and the authors saw no change in muscle ATP ([DOI](https://doi.org/10.1042/cs0830367)). The loading question came next. A 1996 study from the Queen's Medical Centre in Nottingham tested how quickly muscle could be loaded. According to PubMed, in 31 men on varied creatine regimens, 20 grams a day for six days raised muscle total creatine by about 20 percent. Two grams a day held that level. The authors wrote that 3 grams a day was likely to be as effective at raising tissue levels over the long term ([DOI](https://doi.org/10.1152/jappl.1996.81.1.232)). The first trials were not all positive. In 1996, Barnett, Hinds and Jenkins at the University of Queensland gave 17 recreationally active men four days of creatine, at four doses of 70 milligrams per kilogram of body mass a day, against a glucose placebo. The men then completed repeated ten-second cycle sprints. According to PubMed, sprint performance did not change, and neither did blood lactate, blood pH or post-sprint oxygen use. The authors suggested either that four days did not raise resting muscle creatine enough, or that sprint performance is not improved by higher resting creatine ([PubMed](https://pubmed.ncbi.nlm.nih.gov/8742865/)). The strength result came a few years later. In 1999, Volek and colleagues at Penn State ran a 12-week trial with 19 resistance-trained men. The creatine group took 25 grams a day for a week, then 5 grams a day, while both groups trained hard. According to PubMed, fat-free mass rose 6.3 percent in the creatine group and 3.1 percent in the placebo group. Bench press rose 24 percent against 16 percent, and squat rose 32 percent against 24 percent. The authors reported no negative side effects ([DOI](https://doi.org/10.1097/00005768-199908000-00011)). Marketing followed the trials, and it did not always follow the evidence. By 2004, Volek and Rawson wrote that a large number of creatine studies had been published over the previous decade ([DOI](https://doi.org/10.1016/j.nut.2004.04.014)). The 2017 position stand from the International Society of Sports Nutrition calls creatine one of the most popular nutritional ergogenic aids for athletes ([DOI](https://doi.org/10.1186/s12970-017-0173-z)). In 2022, Kreider, Jäger and Purpura wrote that unsupported misrepresentations about the effectiveness and safety of various creatine "forms" had continued after the authors' own 2011 warning. Two of the 2022 authors list Increnovo LLC as an affiliation ([DOI](https://doi.org/10.3390/nu14051035)). Regulators noticed the shelf too. In December 2017, India's Food Safety and Standards Authority (FSSAI) told states to stop the unlicensed manufacture, import and sale of creatine monohydrate. The agency said that no standard for creatine monohydrate had been prescribed under the Food Safety and Standards Act, 2006. It notified 13 companies and served notices on online marketplaces, including Amazon and Flipkart. A 28 March 2018 newspaper report on enforcement in Kerala, reproduced on the FSSAI website, describes the same campaign ([FSSAI-hosted report](https://mail.fssai.gov.in/upload/media/5b446e7eae81eFSSAI_News_Creatine_IndianExpress_28_03_2018.pdf)). ## How the science changed The science moved in four stages. The first was chemistry: isolation in 1832, Liebig's method for meat liquids, and Volhard's synthesis in 1868. The Firenze University Press review adds dated steps before the 1990s. In 1847, Liebig reported higher creatine in wild animals than in domestic ones, and he concluded that muscle creatine depended on activity. In 1912, researchers at Harvard University showed that ingested creatine could raise muscle creatine content. In 1926, Alfred Chanutin gave 10 grams a day for a week and found more creatine in muscle, and he proposed that creatine had anabolic effects. The second was feeding and loading, from the early 1990s. Balsom and colleagues wrote in 1994 that feeding studies a century earlier had found some ingested creatine retained in the body. Their review also noted that no adverse effects had been identified with short-term feeding ([DOI](https://doi.org/10.2165/00007256-199418040-00005)). The Firenze University Press review also records two inherited disorders. In 1994, a patient with a movement disorder had low brain creatine, traced to a defect in one of the two enzymes that make creatine. In 2001, a patient with developmental delay was found to have a mutation in the gene for the creatine transporter. The third stage was safety. In 2006, Derave and colleagues at an exercise research unit in Leuven, Belgium, tested whether creatine raised urinary N-nitrososarcosine, a carcinogen. According to PubMed, they found no systematic increase after 20 grams a day for a week or 5 grams a day for 20 weeks ([DOI](https://doi.org/10.1016/j.nut.2005.10.004)). Earlier, a 2000 review by Wyss and Kaddurah-Daouk had said that some data suggested creatine and creatinine might be precursors of food mutagens and uremic toxins. That was a suggestion in a review, not a finding ([DOI](https://doi.org/10.1152/physrev.2000.80.3.1107)). The fourth stage is the one we live in. The question stopped being whether creatine helps athletes at all, and became who should take it and at what dose. The 2017 ISSN stand says that up to 30 grams a day for five years is safe and well tolerated in healthy people and in a number of patient groups, from infants to older adults. That is a review's verdict. It is not a dose recommendation for everyone. ## How other countries have studied it Several national bodies have ruled on creatine, and their positions cover different claims. The bodies below appear in no order of rank, and none is set against another. Europe. EFSA's opinion, on an application from AlzChem, found cause and effect for creatine taken with resistance training and improved muscle strength in adults over 55. It required at least 3 grams a day, with regular training three times a week for several weeks. Of 21 human intervention studies and two meta-analyses supplied by the applicant, EFSA drew no conclusion from 11 studies and could not use the meta-analyses ([EFSA opinion](https://efsa.europa.eu/en/efsajournal/pub/4400)). United Kingdom. The UK Nutrition and Health Claims Committee published its opinion on 14 August 2024, on a cognitive-function claim also requested by AlzChem. It found one usable trial out of the ten submitted, a 2008 study at about 2.2 grams a day that showed no difference between creatine and placebo. It concluded that cause and effect for cognitive function at 3 grams a day or less had not been established. The committee also says its opinion is not an authorization for marketing and not a safety assessment ([UKNHCC opinion](https://www.gov.uk/government/publications/uknhcc-scientific-opinion-creatine-supplementation-and-improved-cognitive-function/uknhcc-scientific-opinion-creatine-supplementation-and-improved-cognitive-function)). India. The Food Safety and Standards Authority of India (FSSAI) took the position described in the regulators paragraph above. In December 2017, no standard for creatine monohydrate had been prescribed under the Food Safety and Standards Act, 2006, and states were told to stop unlicensed manufacture and sale ([FSSAI-hosted report](https://mail.fssai.gov.in/upload/media/5b446e7eae81eFSSAI_News_Creatine_IndianExpress_28_03_2018.pdf)). United States. The National Center for Complementary and Integrative Health (NCCIH), part of the National Institutes of Health, summarized the CREST-E trial. That trial, reported in 2017, did not support creatine for slowing functional decline in early Huntington disease ([NCCIH summary](https://www.nccih.nih.gov/research/extramural/major-trial-does-not-support-use-of-creatine-for-early-huntingtons-disease)). Japan. A 2025 systematic review of seven randomized trials, with 263 participants, examined creatine with and without β-alanine. According to PubMed, one of its co-authors is at Waseda University in Tokorozawa, Japan, and the review appeared in Nutrients in 2025 ([DOI](https://doi.org/10.3390/nu17132074)). The authors found that the combination enhanced high-intensity exercise performance compared with either supplement alone. Co-ingestion did not increase maximal strength compared with creatine alone, and the effects on body composition were equivocal. China. A 2026 Bayesian network meta-analysis from Guangzhou Sport University compared dietary supplements for lower-limb power and jump performance. According to PubMed, creatine had favorable effects on relative peak power, and its credible interval for countermovement jump excluded zero. Most comparisons were rated very low certainty. Of 1,636 pairwise comparisons, 1,564 were rated very low, and the authors say rankings alone should not guide individualized supplementation regimens ([DOI](https://doi.org/10.1080/15502783.2026.2740221)). A 2026 narrative review from Jiangnan University adds a caution on lean mass. It says creatine-associated increases may reflect intracellular water expansion and glycogen-associated water storage as well as genuine increases in contractile tissue ([DOI](https://doi.org/10.1080/10408398.2026.2742761)). Where the bodies agree and where they do not. EFSA and the UK committee both use the 3-gram line, and both are narrow. EFSA found cause and effect for muscle strength in older adults who train. The UK committee found none for cognition. The CREST-E trial used doses of up to 40 grams a day in people with Huntington disease, which sits outside both claims. The ISSN's 30-gram, five-year verdict covers a broader population. These bodies are answering different questions, so none of them is the referee for the others. ## What the evidence shows now, and where it is disputed Here is the short verdict. Creatine has its strongest support in muscle and training studies. In the trials cited here, it has weak or no support for sprint performance after short loading, for cognition at the doses the UK committee reviewed, and for slowing Huntington disease. The safety record is broad, but it is not closed. The training evidence is the most consistent part of the record. Volek's 12-week trial, the EFSA opinion and the ISSN position stand all point the same way. EFSA's wording is narrow. It found cause and effect for muscle strength with regular resistance training in adults over 55, at 3 grams a day or more. It found no such effect when similar doses were given only on training days, which matters for anyone who trains on some days and not others. The null results are just as useful. Barnett's four-day sprint trial found no change. Rawson and colleagues found no cognitive benefit in 22 young adults given 0.03 grams per kilogram a day for six weeks. According to PubMed, the authors suggested that any benefit might be limited to people whose cognitive processing was already impaired ([DOI](https://doi.org/10.1016/j.physbeh.2008.05.009)). The UK committee found that this was the only usable trial among the ten AlzChem submitted. The largest disappointment is CREST-E, a trial of 553 people with early Huntington disease, randomized to creatine (275) or placebo (278). According to PubMed, the estimated decline on the primary measure was 0.82 points a year on creatine and 0.70 on placebo, with confidence limits that span zero ([DOI](https://doi.org/10.1212/WNL.0000000000004209)). Gastrointestinal side effects were significantly more common on creatine. The journal abstract says serious adverse events, including deaths, were more frequent in the placebo group. NCCIH's summary says those numbers did not differ significantly, so the two sources give different figures for these outcomes ([NCCIH](https://www.nccih.nih.gov/research/extramural/major-trial-does-not-support-use-of-creatine-for-early-huntingtons-disease)). The trial used doses of up to 40 grams a day, far above the 3 grams in the EU and UK claims. On safety, the 2017 ISSN stand holds that up to 30 grams a day for five years is safe and well tolerated in healthy people. A 2023 narrative review by Longobardi and colleagues says controlled trials do not support the idea that creatine impairs kidney function. Creatine can raise serum creatinine, but that does not necessarily mean the kidneys are failing, because creatine converts spontaneously to creatinine. The same authors say people with existing kidney disease still need study ([DOI](https://doi.org/10.3390/nu15061466)). If you have kidney disease, that question belongs with your clinician, not with a history article. The form question is simpler than the marketing suggests. Kreider and colleagues' 2022 critical review says creatine monohydrate remains the only form with substantial evidence for bioavailability, efficacy and safety, and that professional societies recommend it explicitly. Read that verdict with the affiliation list in mind ([DOI](https://doi.org/10.3390/nu14051035)). The dose question is still open. Hultman's loading data point to 20 grams a day for six days as the fast route, with 2 grams a day to hold the level ([DOI](https://doi.org/10.1152/jappl.1996.81.1.232)). EFSA's claim is built on 3 grams a day with training, and the UK committee tested cognition at 3 grams a day or less. A label that promises a faster result at the same grams should be read against those trials. ## Why this history is interesting Three facts from this history are worth repeating. The first is the bouillon tablets, which put creatine on a shelf before anyone had a name for it. The second is the steak: one 5 gram dose matches 1.1 kilograms of fresh, uncooked steak. The third is Volhard's guess from 1868, that the nerve-strengthening power of broth might partly come from creatine and creatinine. It is a plausible idea, and none of the trials cited here tested it in broth. The broth thread runs through modern research too. In 2000, Rennard and colleagues tested a traditional chicken soup in a laboratory assay and found that it inhibited neutrophil migration. Commercial soups varied widely in that activity. The study used cells in a dish, not people, and its abstract does not mention creatine ([DOI](https://doi.org/10.1378/chest.118.4.1150)). If you want the cell side of this story, our [mitochondrial health pillar](/health/mitochondrial-health-complete-guide) covers the machinery that makes ATP, the molecule that creatine helps regenerate. Magnesium has its own [monograph](/ingredients/magnesium). If the Ayurvedic side interests you, our [ashwagandha monograph](/ingredients/ashwagandha-withania-somnifera) covers a different herb from the same tradition. ## What this history supports Creatine's strongest record is in muscle and training studies. EFSA's claim starts at 3 grams a day. The UK committee said cause and effect for cognitive function at 3 grams a day or less had not been established. If you read a creatine label, check two things against this history: the form, where the 2022 review found substantial evidence only for creatine monohydrate, and the grams per serving, where the EU claim starts at 3 grams a day with training. The last detail is the one to keep. A 5 gram dose of creatine matches the creatine in 1.1 kilograms of fresh, uncooked steak. ## Frequently asked questions ### Who first identified creatine, and from what? Michel Eugène Chevreul extracted creatine from meat in 1832, according to Volhard's 1868 paper on its synthesis and a historical review published by Firenze University Press. Volhard's paper also places it in the bouillon tablets of a Dutch company. Later work found creatine in the meat liquid of many animals and in human urine. ### Where does the name creatine come from? The Firenze University Press review says Chevreul named the substance from the Greek word for meat, kreas. Volhard's 1868 paper also credits Chevreul with the name. The chemistry is the firmer ground: when creatine turns into creatinine, it loses one molecule of water. ### How did creatine move from lab curiosity to sports supplement? Three steps mattered. Volhard made creatine in a lab in 1868 and called the method costly and laborious. A 1998 patent described an industrial route. Then the 1990s trials gave researchers something to measure, with Harris and colleagues in 1992 measuring muscle loading and Volek and colleagues in 1999 measuring training gains. By 2004, reviews described a large body of research, and the 2017 ISSN stand called creatine one of the most popular ergogenic aids for athletes. Unsupported claims about "new forms" followed, as the authors of a 2022 review noted. ### What did the 1990s creatine studies measure? Three trials show the range. Harris and colleagues in 1992 measured how creatine loaded into muscle, with increases of as much as 50 percent in some subjects. Barnett and colleagues in 1996 measured repeated ten-second cycle sprints after four days of loading and found no change. Volek and colleagues in 1999 measured 12 weeks of training in 19 men and found greater gains in fat-free mass, bench press and squat in the creatine group, with no negative side effects reported. None of the three measured general health. ### How is creatine monohydrate made today? The most recent process record in this history is US Patent 5,719,319, filed by SKW Trostberg AG in 1996 and published in 1998. It describes reacting cyanamide with sodium or potassium sarcosinate, in water or a water and organic solvent mixture, at 20 to 150 degrees Celsius and a pH of 7.0 to 14.0, and it claims yields of 60 to 90 percent by weight. Volhard's 1868 lab route used sarcosine and cyanamide as well, and he called it costly and laborious for larger amounts. ### What do the early safety studies on creatine actually show? Several early studies addressed specific worries. Balsom and colleagues in 1994 reported that no adverse effects had been identified with short-term feeding. Derave and colleagues in 2006 found no systematic rise in urinary N-nitrososarcosine after 20 grams a day for a week or 5 grams a day for 20 weeks. Later reviews carry the picture forward. The 2017 ISSN stand calls up to 30 grams a day for five years safe and well tolerated in healthy people, and a 2023 review by Longobardi and colleagues found that controlled trials do not support kidney harm, while noting that people with existing kidney disease still need study. ### Did Eastern medicine use creatine? None of the Eastern texts discussed in this article names creatine, which was not identified until 1832. The Charaka Samhita sorts flesh by quality and habit, and a Bencao gangmu entry calls quail flesh a tonic. ### Related reading - [History of collagen, from boiled bones to peptides](/blog/history/history-of-collagen-from-boiled-bones-to-peptides) - [History of turmeric and curcumin, from Ayurveda to isolation](/blog/history/history-of-turmeric-and-curcumin-from-ayurveda-to-isolation) - [History of vitamin B12, from liver cure to cobalamin](/blog/history/history-of-vitamin-b12-from-liver-cure-to-cobalamin) - [the carnitine origins post](/blog/history/history-of-carnitine-from-meat-extract-to-cellular-fuel) - [The history of cordyceps and its exercise trials](/blog/history/history-of-cordyceps-from-tibet-to-athletic-supplement)