What a creatine "lean mass" gain counts

Creatine is the rare supplement with a deep research base, so a positive trial is not a surprise. What interests me as a formulator is the measurement. When a headline says a supplement "added muscle," the question I ask first is: which instrument said so, and what does that instrument actually count?

What the trial did

The paper is by Jisun Chun, Richard Kreider and colleagues at Texas A&M's Exercise & Sport Nutrition Lab, published open access in the Journal of the International Society of Sports Nutrition in August 2026. Seventy-three healthy, sedentary adults first chose for themselves whether to join a supervised exercise-and-diet program (resistance plus aerobic training, eating 300 to 500 kcal a day below their needs) or to change nothing. Within each of those two groups, they were then randomized, double-blind, to creatine monohydrate or a maltodextrin placebo, 5 g twice a day (10 g total) for 12 weeks. Sixty-four people aged 45 to 65 finished and were analyzed. Body composition was measured by DXA (dual-energy X-ray absorptiometry) at weeks 0, 6 and 12.

The headline numbers are DXA "lean tissue mass." In the exercise-and-diet group, creatine users were up 1.69 kg at week 6 and 1.27 kg at week 12, versus essentially no change on placebo. In the no-exercise group, creatine users were up 0.99 kg at week 6 and 1.1 kg at week 12. Leg press and bench press strength also rose more with creatine. That last point matters, and I come back to it below.

What "lean" means on a DXA scan

A DXA scanner sorts the body into three bins: fat, bone mineral, and everything else. "Lean" is the everything-else bin. It holds muscle, but also organs, skin, blood, and the water inside and between cells, plus whatever is in your gut on scan day. Muscle is mostly water to begin with, so this is not a flaw in the machine. It just means a DXA lean gain is a weight gain in the non-fat, non-bone compartment, and the scan alone cannot say how much of it is new contractile protein and how much is fluid.

Creatine is exactly the supplement where that distinction bites. It is pulled into muscle by a sodium-dependent transporter, and that rapid uptake may draw water in with it to balance the change inside the cell. A small 2023 study by Emily Buck and colleagues gave young adults a high dose (0.3 g per kg of body weight) for one week. Per the abstract, fat-free mass on DXA rose about 1.1 kg more than on placebo, and total body water measured by bioimpedance rose about 2%. The authors concluded the body-composition changes were driven by that water increase. One week is far too short to build a kilogram of muscle, which is the point.

The trial that looked before training began

The best test of this I have found is a 2025 trial in Nutrients by Imtiaz Desai, Mandy Hagstrom and colleagues at UNSW Sydney. They gave 63 adults (average age 31) creatine at 5 g a day, but scanned them after a seven-day "wash-in" with no training, before a 12-week program started. After that one sedentary week, the creatine group had gained 0.51 kg more DXA lean mass than controls. Over the next 12 weeks of training, both groups gained about 2 kg, with no difference between them.

Where that first-week gain sat is the telling part. Arms and legs showed no difference; the trunk did (0.32 kg versus −0.10 kg). The authors wrote: "Given the increase in trunk LBM, coupled with the influence of body water and stomach content on DXA measures of LBM, it is plausible that the LBM increases are owing to the changes in fluid in the trunk region."

Two honest caveats on that trial. Its control group received no placebo, and the authors didn't control or measure hydration, so it is not proof of water either. And they note the water evidence across the creatine literature is itself mixed: several 5 to 6 week studies found no change in body water, while others found increases. So the fair statement is not "creatine lean gains are just water." It is that an early DXA lean gain on creatine is ambiguous, and a single scan can't resolve it.

Reading the Texas A&M numbers with that in mind

The shape of the data fits that ambiguity. In both creatine groups, most of the lean gain had already appeared by the first follow-up scan at week 6, and in the exercise group it was smaller at week 12 than at week 6. A fluid shift that arrives early and then plateaus would look like that. So would early muscle gain. I searched the paper and it does not report any measure of body water, intracellular water, or muscle size by ultrasound or MRI, which are the measurements that could tell those apart.

The strength results are the stronger evidence, because water in a muscle doesn't lift a weight. The paper reports that creatine without exercise also increased one-rep-max strength on leg press and bench press, which points to a functional effect beyond fluid. That is consistent with creatine's known role in rapid energy supply during short, hard efforts. It still doesn't tell you how many of the reported kilograms are muscle tissue.

Other things worth knowing about this trial

Who exercised was not randomized. Participants picked their own arm (exercise or not); only the supplement was randomized within each arm. The comparisons that matter most here (creatine versus placebo inside the same arm) are protected by randomization. Comparisons across arms are not, and the authors flag it.

Registration came after enrollment began. The paper states the first participant enrolled on 10 July 2024 and the trial was submitted to the ISRCTN registry on 11 April 2025. Registering before the first participant is the standard that protects against outcomes being chosen after the fact. Late registration is not evidence of anything wrong, but it is weaker protection.

The cognition results are labeled exploratory by the authors themselves. The paper says individual cognitive and biomarker findings "should be interpreted as exploratory," because many outcomes were tested without adjusting for multiple comparisons. If you see this trial cited as proof creatine sharpens memory in midlife, that is further than the authors went. (I covered creatine and cognition separately in Report 028.)

A kidney number moved, as expected. Creatine raises blood creatinine, which makes creatinine-based eGFR look worse without any change in kidney function. The authors note they didn't measure cystatin C or measured GFR, so the eGFR shift "should not be interpreted as evidence of reduced renal filtration." Worth knowing if you take creatine and get routine bloodwork.

Funding and conflicts are disclosed. The study was funded by an unrestricted gift from the WoodNext Foundation. Alzchem, a creatine manufacturer, supplied the creatine and funded the homocysteine assays, and the paper states "The sponsors were not involved in data collection, analysis, or interpretation." The senior author discloses that he chairs Alzchem's scientific advisory board and has conducted industry-sponsored creatine research. That is common in this field and openly stated. It is also a reason to want the next trial run by someone else.

How to read the next creatine "muscle" headline

Which instrument? DXA and bioimpedance both count water as lean. Muscle thickness by ultrasound, cross-sectional area by MRI, or a body-water measurement alongside the scan gets closer to tissue.

When was the baseline? If the first scan was taken before any creatine and the next one weeks later, the early fluid shift is baked into the "gain." The UNSW design, scanning again after a week of creatine alone, is the cleaner way to separate them.

Did strength move too? A lean gain with a strength gain is more convincing than either alone.

What does it mean for one person? In the UNSW lab, the DXA's own test-retest error for an individual scan was 0.59 kg. Group averages wash that out; your own before-and-after scan doesn't. A one-kilogram change on your personal report is within shouting distance of the machine's noise.

None of this makes creatine a bad bet. The overall evidence for it is far stronger than for most of the shelf, and this trial adds a careful data set in an age group that matters. It just means "added lean mass" and "added muscle" are two different claims, and the trial measured the first.

Not medical advice. This is educational analysis, not a recommendation — a study is not a prescription. Talk to a qualified clinician before acting on anything you read here. Full disclaimer →

Disclosure, plainly: I founded and run Shroombiosis (a company I run), which formulates and sells supplements. That is a commercial stake in how supplement research gets read, and the same measurement questions apply to anything I make. I have no relationship with any author, lab, sponsor or brand discussed here. Nothing here is sponsored and no link earns a commission; here's the full policy. A recommendation with no stake at all: for performance nutrition, Die Tryin Co. is a fellow combat-veteran-owned brand I'm glad to point people to. I don't own it and earn nothing from the link.