Ask a room full of marathoners about creatine and most of them will tell you it’s a gym supplement that makes you heavy. That answer is about twenty years out of date, and it’s also not entirely wrong, which is exactly why this topic deserves a careful look rather than a slogan.
Here’s the short version. Creatine has not been shown to make trained endurance athletes faster over steady, continuous efforts, and the meta analyses on this point are genuinely unflattering. But the same literature contains findings on glycogen storage, recovery from long races, and repeated surges that most runners have never heard about, because they sit outside the question everyone keeps asking. If you only read the headline, you’ll dismiss creatine. If you read the studies, the picture is more interesting and more useful.
I want to walk through both halves honestly: the results that justify the skepticism, and the results that suggest endurance runners have been asking creatine the wrong question.
Why Runners Wrote Creatine Off
Creatine’s reputation was built in the weight room, and for good reason. When you supplement with creatine monohydrate, the standard and best studied form, your muscles store more phosphocreatine, a molecule that regenerates ATP during short, intense efforts. ATP is the immediate energy currency of muscle contraction, and phosphocreatine is the fastest way to refill it. That system dominates efforts lasting seconds, not hours, so the mechanism never looked like it belonged in a marathoner’s toolkit.
The 2017 position stand from the International Society of Sports Nutrition calls creatine monohydrate the most effective nutritional supplement available for increasing high intensity exercise capacity and lean body mass during training. Notice what that sentence carefully does not say. It does not say anything about 10K times, half marathon pacing, or your long run.
Then there’s the weight. Creatine is osmotically active, meaning it pulls water into muscle cells along with it. During a classic loading protocol of around 20 grams per day for five to seven days, most people gain somewhere in the region of one to two percent of body mass, largely as intracellular water. For a sport where runners agonize over the mass of their shoes, a kilogram of anything feels like a dealbreaker. Combine an irrelevant looking mechanism with guaranteed scale weight and you get the folk wisdom that has dominated running forums for two decades.
The folk wisdom is not baseless. The next section is the evidence that supports it, and it deserves to be reported plainly rather than buried.
The Inconvenient Findings First
This site’s rule is that the null results come before the promising ones, so here they are.
A 2023 systematic review and meta analysis in Sports Medicine by Fernández-Landa and colleagues pooled 13 placebo controlled trials of creatine monohydrate in trained populations, using outcomes such as terrain runs, treadmill tests, and incremental tests on rowing and cycle ergometers. The pooled result was a non significant change in endurance performance, with a trivial effect size of minus 0.07 and a p value of 0.47. Excluding studies for funnel plot asymmetry didn’t change the conclusion. The authors’ summary was blunt: creatine monohydrate supplementation was ineffective for endurance performance in a trained population.
A separate meta analysis of creatine and VO2max by Gras and colleagues, published in Critical Reviews in Food Science and Nutrition, went further. VO2max is the maximum rate at which your body can use oxygen, a standard laboratory measure of aerobic capacity. Across 19 randomized controlled trials covering 424 people, mostly men with a mean age of 30, VO2max increased in both creatine and placebo groups over the study periods, but it increased less with creatine. The effect size for the difference in changes was minus 0.32, and VO2max after supplementation was lower in the creatine groups with an effect size of minus 0.20. The authors concluded that creatine has a negative effect on VO2max regardless of training, supplementation protocol, or population.
I don’t think either paper should be waved away, and I’m wary of anyone who tries. But both have real limitations worth understanding before you decide what they mean for you.
The most obvious one is body mass. Many of the pooled trials used loading doses, and VO2max is often expressed relative to body weight, in milliliters per kilogram per minute. Add a kilogram of water and your relative VO2max drops on paper even if your heart and lungs haven’t changed at all. Some of the underlying studies measured absolute VO2max in liters per minute and some measured relative values, and the meta analyses could not fully untangle how much of the negative signal is arithmetic rather than physiology. The Gras paper’s own meta regressions found no supplementation characteristic that explained the effect, which cuts both ways: it means the negative signal is consistent, and it also means the analysis couldn’t isolate the water weight question.
The second limitation is what these trials measured. Time to exhaustion tests, incremental ergometer protocols, and short terrain runs are not races. Almost none of the pooled evidence looks like a competitive endurance event, with fueling, surges, hills, a finishing kick, and a hard session again three days later. That’s not a defense of creatine. It’s a description of a gap. The honest reading is that creatine does not improve, and may slightly impair, the laboratory measures of steady aerobic performance, and that the question runners actually care about has been studied far less than you’d expect.
The Weight Question, With Numbers Instead of Fear
Since weight is the reason most runners never try creatine, it’s worth being precise about it.
The initial gain during loading is water, held inside muscle cells rather than under the skin, and the 2025 review of common creatine questions in the Journal of the International Society of Sports Nutrition by Antonio and twenty of the field’s most published researchers addresses this along with the persistent myths about dehydration and cramping, which the controlled evidence does not support. Skipping the loading phase and taking three to five grams daily gets muscles to the same saturation over three to four weeks, with the water arriving gradually instead of all in one week.
Whether one to two percent of body mass matters to you depends on your event and your build. Running is weight bearing, and physics is not negotiable, so a small cost to running economy from added mass is plausible, particularly uphill. One small crossover study in well trained endurance runners examined creatine and glycerol hyperhydration and running economy precisely because researchers worried the added fluid, several hundred milliliters, might make running more expensive. In that context the hyperhydration was deliberate, aimed at heat stress, and the running economy question was the tradeoff being tested. The general point stands: for a runner, creatine’s water weight is a genuine cost to weigh against the benefits, not an imaginary one. It’s just a smaller and more temporary cost than forum folklore suggests, and it’s the only well documented side effect in healthy adults across decades of study.
Glycogen: The Finding That Should Get More Attention
Now for the half of the literature runners rarely hear about, starting with the finding I think deserves the most attention.
Muscle glycogen, the stored carbohydrate that fuels the second half of a marathon, is one of the few variables in endurance performance nearly everyone agrees on. Run out and you hit the wall. So it should raise eyebrows that creatine appears to help muscles store more of it.
In a controlled trial from Paul Greenhaff’s group at Nottingham, published in Amino Acids, fourteen healthy men cycled to exhaustion at 70 percent of their peak oxygen uptake to empty their glycogen stores, then ate a prescribed high carbohydrate diet for six days while taking either 20 grams of creatine daily or a placebo. The creatine group stored noticeably more glycogen than placebo after just one day of recovery, and the advantage held through day six. This confirmed the group’s earlier work, and the augmentation happened before muscle creatine had even risen much, which tells you something subtle is going on beyond simple cell swelling.
Earlier work in the Journal of Applied Physiology found the same interaction ran through exercise itself: combining creatine and carbohydrate after glycogen depleting exercise boosted glycogen resynthesis specifically in the exercised muscle, and the authors judged the size of the increase sufficient to matter for endurance performance.
Two honest caveats. These were small studies in healthy men, using loading doses alongside aggressive carbohydrate intake, not three grams a day alongside normal eating. And storing more glycogen is a mechanism, not a race result; no trial has yet shown that creatine assisted carb loading translates into a faster marathon. What we can say is that the best characterized limiter of long race performance responds to creatine in controlled conditions, in the right direction, and that this finding has been sitting in the literature for years while the conversation stayed stuck on VO2max.
Recovery After Long, Hard Races
The second overlooked cluster of evidence concerns what happens to your body after an event long enough to damage it.
The clearest study for runners is from Santos and colleagues, who supplemented experienced marathoners before a 30 km race. Thirty four male athletes, all with multiple marathons and personal bests between 2.5 and 3 hours, took either 20 grams of creatine plus carbohydrate daily for five days before the race, or carbohydrate alone, in a double blind design. After the race, the placebo group showed what a 30 km effort normally does to a body: plasma creatine kinase, a marker of muscle cell damage, rose 4.4 fold, lactate dehydrogenase rose about 43 percent, and the inflammatory signals prostaglandin E2 and TNF alpha rose 6.6 fold and 2.3 fold. The creatine group’s rises were substantially attenuated, and the authors reported no cramping, gastrointestinal trouble, or other adverse effects despite the high dose and the race conditions.
A companion study by Bassit and colleagues in triathletes finishing a half Ironman found the same pattern: five days of 20 grams daily before the race markedly reduced the post race rise in TNF alpha, interferon alpha, interleukin 1 beta, and prostaglandin E2 at 24 and 48 hours, compared with placebo.
Before anyone gets carried away, the limitations here are real and I want to name them. The Bassit study had eleven participants. Eleven. The Santos study is over twenty years old and has not been replicated at scale in runners. Both measured blood markers, not how fast anyone ran the following week. And when researchers have pooled the broader muscle damage literature, the picture gets messier: a 2022 systematic review and meta analysis in Sports Medicine on creatine and muscle damage markers describes the pooled effects as paradoxical, with results varying by exercise type and timing, and creatine kinase itself is an unreliable proxy for how recovered an athlete actually feels or performs.
So the fair summary is this: after single, very long, very damaging efforts, the two most relevant studies both found meaningfully lower inflammation and damage markers with creatine, no study has found the opposite in that context, and the evidence base is thin enough that I’d call it promising rather than established. For a runner stacking a long race onto a training block, faster recovery would be worth more than a small VO2max debate. It’s genuinely strange that this question has attracted so little modern research money.
Surges, Hills, and the Finishing Kick
Races are not conducted at a single steady output, which brings us to the third overlooked area: what creatine does to the high intensity moments embedded inside endurance efforts.
Here the results split in an instructive way. A 2023 review in the Journal of the International Society of Sports Nutrition on creatine and endurance performance collects the trials. In one study of twelve triathletes, creatine at 6 grams daily for five days improved power output by about 18 percent during repeated 15 second intervals inserted into prolonged cycling. In a study of elite cyclists by Vandebuerie and colleagues, creatine loading before two and a half hours of endurance cycling improved the five maximal 10 second sprints that followed. Rico-Sanz’s protocol, alternating between 30 and 90 percent of maximal aerobic power, extended time to exhaustion from about 30 to 36 minutes in highly trained athletes after a standard loading week.
And yet a 28 day trial at 3 grams per day in trained cyclists, using a simulated two hour road race followed by a sprint to exhaustion, found nothing on performance despite a 24 percent rise in resting muscle creatine. The design differences matter: repeated short sprints with brief recoveries lean hard on phosphocreatine, while a single continuous sprint to exhaustion leans on other systems. Creatine helps the first kind of effort and appears not to help the second.
Whether the same holds late in a running race is a fair question, since almost all of this work was done on bikes, where the extra body mass is cheaper to carry. Cycling protocols dominate this corner of the literature because power is easy to measure on an ergometer, not because researchers think runners are different, but the transfer is an assumption rather than a demonstrated fact, and I’d rather flag that than gloss it.
Translate that to running and you get a testable, modest claim rather than a hype claim. If your racing involves repeated surges, hilly courses, tactical track and cross country racing, or hard interval sessions in training, the mechanism and several trials point toward a benefit for those specific moments. If your racing is a metronomic flat marathon, the interval data mostly doesn’t apply to you. That’s a narrower claim than either the fans or the skeptics usually make, and it’s what the evidence supports.
Heat, Hydration, and the Old Cramping Myth
For years, runners were warned that creatine causes dehydration and heat cramps, advice that spread through coaching circles faster than any study ever did. The logic sounded plausible: creatine pulls water into muscle cells, so surely less water is available elsewhere when you’re sweating through a hot half marathon.
The controlled evidence never cooperated with the story. The 2025 review of common questions in the Journal of the International Society of Sports Nutrition addresses the dehydration and cramping claims directly and finds the research does not support them, a conclusion its author group first reached in an earlier review and has only strengthened since. If anything, the reasoning has flipped. Because creatine expands total body water, some researchers have explored it as a deliberate hyperhydration strategy for hot conditions, which is exactly what the crossover study of creatine and glycerol in well trained endurance runners was probing: whether several hundred milliliters of retained fluid could blunt thermal and cardiovascular strain, and whether the extra mass would cost running economy in exchange.
I’d stop short of recommending creatine as a heat strategy, because that literature is small and the protocols are specialized. But the direction of the modern research matters. The supplement runners avoided for fear of dehydration is now being studied as a tool for staying hydrated. When the science and the folklore point in opposite directions for this long, the folklore should lose.
Who These Findings Apply To, and Who They Don’t
Nearly everything above was measured in trained men. That’s not a throwaway line; it changes how confident anyone should be in applying the findings elsewhere.
A 2025 systematic review in Nutrients on creatine in active females examined 27 studies and judged the overall findings inconclusive, with wide variation in athlete caliber, dosing, and study quality, and with methodological considerations specific to female athletes poorly addressed across the literature. That is a verdict on the research, not on creatine, but it means a female marathoner reading this article is extrapolating further than a male one, and she deserves to know that rather than be handed conclusions borrowed from studies of thirty year old men.
Vegetarians and vegans start with lower muscle creatine, since dietary creatine comes almost entirely from meat and fish, and they tend to show larger responses to supplementation in muscle saturation studies. The endurance specific data in this group is thin, though, so the same caution applies.
Older runners sit in a slightly different position. Masters athletes lose muscle mass and power with age faster than they lose aerobic capacity, and the broader creatine literature in older adults, combined with resistance training, is one of the stronger corners of the field. That research wasn’t run in endurance athletes, so I won’t stretch it further than it goes, but a sixty year old runner who also lifts is closer to the populations where creatine’s training benefits are best documented than a twenty five year old marathoner is.
And if you have kidney disease or any other condition that affects how your body handles creatine or creatinine, the research summarized here was not done in people like you, and the only sensible move is to talk to your doctor before supplementing. The safety record in healthy people, documented across the ISSN position stand’s review of studies up to 30 grams daily for as long as five years, is genuinely strong, but it is a record about healthy people.
What the Studies Used Versus What Runners Actually Do
There’s a quiet mismatch running through this whole literature that almost never gets pointed out. The impressive findings, glycogen supercompensation, reduced post race inflammation, better repeated surges, mostly used loading protocols of around 20 grams per day for five to seven days, often timed right before an event or a depletion protocol. What most people actually do, and what most modern guidance suggests, is three to five grams daily, indefinitely.
Those approaches reach the same muscle saturation eventually, which is why the maintenance dose is reasonable for anyone taking creatine for general training support. But you cannot assume that every acute finding from a loading study reproduces itself under a casual maintenance routine, and the one 28 day maintenance dose trial in trained cyclists discussed above found no performance effect at all. Equally, the negative VO2max signal came substantially from loading era studies where rapid water gain was maximal, so the maintenance approach plausibly softens the main cost too. The honest position is that dose and timing are underexplored in endurance contexts specifically, and anyone who tells you the protocol details are settled for runners is selling certainty the trials don’t contain.
One more practical note from the studies themselves: in the 30 km race trial and the half Ironman trial, athletes tolerated full loading doses through race week without reported cramping or gut trouble. Even so, nobody should first try any supplement in race week. Whatever you decide, decide it in training.
What Would Change the Picture
I’ll end with what’s missing, because that tells you how much weight today’s conclusions can bear.
The study I most want to see is simple to describe and apparently hard to fund: a properly powered randomized trial in trained distance runners, women included, using a maintenance dose, measuring actual race or time trial performance plus recovery and training quality across a full block, with body composition tracked well enough to separate water from everything else. Nothing like it currently exists. Until it does, creatine for endurance runners sits in an odd evidentiary spot: convincingly null for steady state speed, mildly negative for laboratory VO2max, promising for glycogen storage, promising but thin for post race recovery, reasonably supported for repeated surges, and nearly silent on the outcome runners care most about.
A 2026 narrative review in the Journal of the International Society of Sports Nutrition put the current consensus about as fairly as it can be put: creatine is not a classical endurance enhancer for steady state performance, but shows promise for high intensity bursts within endurance sports and for aiding recovery. That’s a much less exciting sentence than either “creatine is useless for runners” or “every runner needs creatine”, and it happens to be the one the evidence supports.
If you’re an endurance runner deciding today, the decision comes down to your own tradeoffs. The costs are a small amount of water weight and the price of a cheap, well studied powder. The plausible benefits are better carbohydrate storage, possibly gentler recovery from your longest efforts, and stronger surges and interval sessions, with the direct steady state benefit firmly not on the list. Neither the dismissal nor the hype survives contact with the actual papers, and you now know enough about those papers to make the call yourself.