Creatine Beyond the Gym
Ask most people what creatine does and the answer will involve muscle. It is one of the most researched supplements in existence, and the International Society of Sports Nutrition calls creatine monohydrate the most effective legal supplement for building strength and high intensity exercise capacity. What far fewer people know is that creatine has a second, quieter research story: its relationship with blood sugar.
That story is the subject of a scientific presentation by Bruno Gualano, a professor at the University of São Paulo in Brazil, whose research group ran the key clinical experiment in this area: a randomized, double blind, placebo controlled trial of creatine in people with type 2 diabetes. Gualano is careful to present the work as promising rather than proven, and this article follows his lead. Below is what the science actually shows about creatine and blood sugar: the century of clues that led to the trial, what the trial found, the biological mechanism that seems to explain it, the safety data in people whose kidneys were already under strain, and an honest account of how far the evidence does and does not reach.
A Century of Metabolic Clues
The idea that creatine touches glucose metabolism is remarkably old. Gualano points out that the first observation dates back roughly a hundred years: in 1928, a researcher named Hill reported that giving creatine to dogs lowered their blood sugar, a finding recounted in the 2021 review Gualano coauthored in the journal Nutrients. It was a single early report, not a body of evidence, but it planted a question that took decades to mature.
Laboratory work later added texture. In cell and tissue experiments, creatine and chemically related guanidine compounds modestly stimulated insulin release from isolated rat pancreas, and similar effects appeared in mouse islet cells and insulin producing cell lines. Animal feeding studies pushed further: in one study in rats, three weeks of creatine in the diet increased muscle glycogen, raised the amount of the glucose transporter GLUT4, and activated AMPK, an enzyme that acts as a cellular fuel gauge. Other rodent experiments, however, found no change in glucose handling at all, and the Nutrients review is blunt that animal results diverge so much between species and protocols that they are difficult to generalize.
The human evidence is more consistent on one point: glycogen. Trials found that creatine loading enhanced glycogen supercompensation after exercise, and a six week creatine protocol increased muscle glycogen content by about 18 percent in young adults. Interestingly, that same six week study found no change in the amount of GLUT4 the muscles produced, an early hint that if creatine helps humans handle glucose, it probably does so by changing how existing machinery behaves rather than by building more of it.
The Case for Pairing Creatine With Exercise
To understand why Gualano tested creatine together with exercise, it helps to appreciate just how powerful exercise is for blood sugar. A landmark analysis in JAMA pooled 47 randomized controlled trials covering 8,538 patients with type 2 diabetes and found that structured exercise training lowered HbA1c, the standard three month marker of blood glucose, by an average of 0.67 percentage points, a reduction that compares favorably with some medication combinations. Training more than 150 minutes per week was associated with an even larger drop of 0.89 points. Muscle contraction pulls glucose out of the blood through a route that does not depend on insulin, and research shows that people with type 2 diabetes retain this exercise triggered response even when their insulin signaling is impaired. That is exactly why exercise is a cornerstone of diabetes care.
Creatine’s most intriguing human results happen precisely at this intersection with training. In a Belgian study, healthy volunteers had one leg immobilized for two weeks: GLUT4 protein in the idle muscle fell by about 20 percent in the placebo group but was preserved in those taking creatine, and during the retraining period that followed, the creatine group’s GLUT4 climbed roughly 40 percent above baseline. A related study found that creatine combined with rehabilitation training raised muscle GLUT4 and improved oral glucose tolerance. Gualano’s own earlier trial pointed the same way: in sedentary healthy men, twelve weeks of creatine plus moderate aerobic training produced a greater fall in blood glucose during a glucose tolerance test than training alone.
The contrast case completes the logic. When researchers gave creatine to men who were not training, glucose tolerance, insulin responses, and muscle glycogen did not budge. According to Gualano, this pattern suggested that creatine might act as an amplifier of the training signal rather than a standalone glucose drug, and it raised an obvious clinical question: if exercise is already a proven therapy for type 2 diabetes, could an inexpensive supplement make that therapy work better?
The São Paulo Type 2 Diabetes Trial
To test the idea, Gualano’s team designed a trial lasting 12 weeks, published in Medicine and Science in Sports and Exercise in 2011. It was randomized, double blind, and placebo controlled, which for a supplement study is the standard that matters. Participants were adults older than 45 with type 2 diabetes treated with metformin and a sulfonylurea, with HbA1c below 9 percent and a glomerular filtration rate above 30 mL per minute. That last criterion is important: it deliberately allowed people with meaningfully reduced kidney function into the trial, because one of the study’s goals was to test safety in exactly the population where creatine’s safety had been least studied.
Everyone in the trial completed supervised exercise training at the hospital’s exercise physiology laboratory, and each participant took either 5 g of creatine or a matching placebo daily. From an initial pool of more than a hundred volunteers who responded to recruitment, 25 participants were included in the final analysis: 13 in the creatine group and 12 on placebo. Gualano notes the group skewed female, had low aerobic fitness at baseline as measured by VO2 max, and was balanced between arms for age, sex, and stage of kidney disease.
The measurements were unusually thorough for a supplement study. The primary endpoint was HbA1c. Secondary measures included glucose, insulin, and C-peptide responses to a meal tolerance test, lipid profile, muscle strength and function, aerobic conditioning, body composition by DXA scanning, and liver enzymes. Muscle phosphorylcreatine was tracked with phosphorus magnetic resonance spectroscopy to confirm the supplement was actually being absorbed, muscle biopsies allowed the team to examine GLUT4 directly, and a full kidney panel anchored the safety analysis.
Blood Sugar Results
The first result was a compliance check that passed: muscle phosphorylcreatine rose significantly in the creatine group and not with placebo, with an estimated difference of about 23.6 mmol per kg of muscle. The supplement reached its target.
The headline finding was the primary endpoint. HbA1c fell from an average of 7.4 to 6.4 percent in the creatine group, while drifting from 7.5 to 7.6 percent with placebo. The between group difference was 1.1 percentage points, with a 95 percent confidence interval spanning 0.4 to 1.9 points and a P value of 0.004. For context, that effect size is in the neighborhood of what the JAMA pooled analysis attributes to structured exercise itself, so the trial suggests creatine roughly doubled the glycemic payoff of the training program, though a small study cannot pin that number down precisely.
The meal tolerance test explained where the improvement came from. Blood glucose was significantly lower in the creatine group at the start of the test and at 30 and 60 minutes after the meal, and the change in total glucose area under the curve favored creatine over placebo. In plain terms, the creatine group’s blood sugar climbed less after eating.
Just as telling is what did not change. Lipid profile, fat mass, strength, muscle function, and aerobic conditioning were all unaffected, so the blood sugar benefit was not a side effect of getting fitter or leaner than the placebo group. And insulin and C-peptide, a marker of the body’s own insulin output, did not rise either, which points away from the pancreas and toward the muscle itself.
The GLUT4 and AMPK Mechanism
GLUT4 is the transporter that moves glucose from the bloodstream into muscle cells. Most of the time it sits in storage vesicles inside the cell; insulin signaling or muscle contraction moves it to the sarcolemma, the cell’s outer membrane, where it can actually collect glucose. The Nutrients review notes that in type 2 diabetes this translocation step is often suboptimal even when the total amount of GLUT4 is normal, which makes it a natural therapeutic target.
The biopsy data from the trial map onto that picture almost exactly. Gualano reports that total GLUT4 content did not change with creatine, but the fraction of GLUT4 translocated to the sarcolemma increased significantly versus placebo, reaching levels similar to those his team measured in healthy reference subjects. An ancillary analysis of the same trial, published in Amino Acids in 2012, found that AMPK protein increased in the creatine group, that AMPK correlated directly with GLUT4 translocation, and that it correlated inversely with the change in HbA1c: the more AMPK, the bigger the blood sugar improvement.
Gualano walks through the candidate explanations with unusual candor. The first, creatine boosting insulin secretion from beta cells, has support in cell and animal experiments but not in people; his own data showed no insulin change, so he sets it aside. The second is osmotic: creatine draws water into muscle cells, cell swelling is a documented stimulus for glycogen synthesis, and a gene expression study found creatine altered osmosensing genes and proteins in healthy men. This remains a hypothesis rather than a demonstrated pathway. The third, and the one his data support, is the GLUT4 route, acting through translocation rather than production.
He also names the explanation that his study design cannot rule out: creatine may simply amplify what exercise was already doing. Because every participant trained, the trial cannot separate a direct effect of creatine from creatine enhancing the training adaptations. Exercise itself drives GLUT4 to the membrane through AMPK in an insulin independent way, so a supplement that strengthens that signal would produce exactly the pattern observed. Scientifically the distinction matters; practically, for a patient who is exercising anyway, it may matter less.
Creatine Safety and Kidney Function
The kidney question deserves its own section, because it is the most common worry about creatine and the most misunderstood. Creatine is naturally converted into creatinine, the very molecule doctors measure to estimate kidney function. Someone taking creatine can therefore show higher creatinine without any actual kidney harm. Gualano’s team sidestepped this trap by using chromium EDTA clearance, a tracer technique regarded as a reference method for measuring true filtration rate, reported in a companion paper in the European Journal of Applied Physiology.
The results were reassuring across the board. EDTA clearance did not change: roughly 90 mL per minute before and 96 after in the creatine group, adjusted for body size, with an estimated between group difference of essentially zero. Proteinuria and albuminuria were stable, and Gualano highlights that even the participants who entered the trial with existing albuminuria, a sign of established kidney damage, showed no deterioration. Electrolytes, creatinine clearance, urea, and liver enzymes were unaffected, and no adverse events were attributed to the supplement.
This fits a much larger safety literature, one Gualano credits in part to the late Jacques Poortmans, a pioneer of exercise biochemistry. Poortmans’s team compared people who had taken creatine continuously for anywhere from 10 months to 5 years against nonusers and found normal filtration rates, tubular function, and membrane permeability in both groups. The International Society of Sports Nutrition’s position stand, which reviews hundreds of studies, concludes that supplementation at doses up to 30 g per day for as long as five years is safe and well tolerated in healthy people and a range of patient populations. A pooled analysis published in 2026 in International Urology and Nephrology reached the same conclusion using measured filtration rates and kidney injury markers.
Two honest caveats remain. The number of people with genuine kidney disease who have been studied on creatine is still small, this trial included, so anyone with kidney disease should involve their doctor before supplementing. And because blood sugar medicines sometimes need adjusting when glucose control improves, people with diabetes should treat any supplement experiment as something to manage with their clinician, not around them.
The Wider Evidence on Creatine and Glucose
A single positive trial, however well run, is a beginning rather than a verdict, and the surrounding literature counsels patience. A 2016 systematic review of creatine and glycemic control found the animal results so divergent across species and protocols that they resist interpretation, and the human trials few and small. A tiny open label crossover study reported that five days of creatine lowered blood glucose comparably to metformin in patients with type 2 diabetes, but its design was too weak to carry much weight.
There is also a genuinely null result to reckon with. A pilot randomized, double blind, placebo controlled trial in older adults combined 5 g of creatine daily with resistance training for 12 weeks and found no improvement in fasting glucose, insulin, or HOMA scores. The Nutrients review speculates that a ceiling effect may explain the difference: those participants started with better glucose control than Gualano’s patients, leaving less room to improve. That interpretation is plausible but unproven, and the null finding stands as a reminder that the São Paulo result has not yet been reproduced.
Gualano is refreshingly frank about the field’s pace. He recalls lecturing on creatine and glucose metabolism alongside Paul Greenhaff at a creatine research conference in Cambridge back in 2010, presenting the supplement as a possible adjunct in diabetes management, and he laments that more than a decade later, little progress has been made. His own 2021 review states the position plainly: preliminary findings in type 2 diabetes are promising, but the efficacy of creatine for improving glycemic control is yet to be confirmed, and the evidence overall remains speculative.
Limits and Open Questions
The trial’s limitations are the ones Gualano volunteers himself. Twenty five analyzed participants is a small sample; twelve weeks is a short window for a chronic disease; and the study ran at a single research center in patients on one specific medication combination, metformin plus a sulfonylurea, which limits how far the findings generalize. The combined design means no one can yet say whether creatine acts directly on glucose handling or works by enhancing exercise adaptations. And an exploratory trial with a positive primary outcome is a hypothesis strengthened, not a treatment established: as Gualano puts it, only large randomized controlled trials can answer whether creatine belongs in type 2 diabetes care.
Transparency is worth noting too. The 2011 trial was supported by Brazilian public research funding, and its authors declared no conflicts of interest at the time. In more recent work and in the presentation itself, Gualano discloses research support, product donation, and advisory ties to AlzChem, a company that manufactures creatine. None of that invalidates the data, which have been public and scrutinized for over a decade, but readers weighing enthusiasm in this field should know the landscape.
Practical Takeaways
So where does this leave someone thinking about creatine and blood sugar? First, creatine is not a diabetes treatment, and nothing in the current evidence justifies using it as one or adjusting any medication around it. The intervention with the strongest proof in this whole story is exercise: structured training reliably lowers HbA1c, and every benefit seen with creatine in humans appeared alongside training, never instead of it.
Second, the specifics of the research are modest and practical. The São Paulo trial used 5 g of creatine monohydrate per day, in line with the common maintenance range of 3 to 5 g daily used across creatine research, taken while following a supervised exercise program three times a week. Nothing exotic, nothing high dose.
Third, the safety picture is genuinely good for healthy people, and early evidence in patients with diabetes and reduced kidney function is reassuring, but the studied numbers are small. Anyone with diabetes, kidney disease, or any chronic condition should talk with their doctor before adding creatine, both because individual circumstances vary and because improving glucose control can itself change what medications and monitoring a person needs.
The honest bottom line: one small, careful trial found that creatine amplified the blood sugar benefits of exercise in type 2 diabetes, through a plausible and partially demonstrated mechanism, with no safety signals. That is a genuinely interesting result sitting at the early stage of the evidence ladder, waiting for the large trials that would move it up.
This article is for information only and is not medical advice.