Background: Creatine supplementation (CS) has been reported to increase body mass and improve performance in high-intensity, short-duration exercise tasks. Research on CS, most of which has come into existence since 1994, has been the focus of several qualitative reviews, but only one meta-analysis, which was conducted with a limited number of studies.

Purpose: This study compared the effects of CS on effect size (ES) for body composition (BC) variables (mass and lean body mass), duration and intensity (< or = 30 s, [ATP-PCr = A]; 30-150 s [glycolysis = G]; >150 s, [oxidative phosphorylation = O]) of the exercise task, type of exercise task (single, repetitive, laboratory, field, upper-body, lower-body), CS duration (loading, maintenance), and subject characteristics (gender, training status).

Methods: A search of MEDLINE and SPORTDiscus using the phrase "creatine supplementation" revealed 96 English-language, peer-reviewed papers (100 studies), which included randomized group formation, a placebo control, and human subjects who were blinded to treatments. ES was calculated for each body composition and performance variable.

Results: Small, but significant (ES > 0, p < or = .05) ES were reported for BC (n=163, mean +/- SE=0.17 +/- 0.03), ATP-PCr (n=17, 0.24 +/- 0.02), G (n=135, 0.19 +/- 0.05), and O (n=69, 0.20 +/- 0.07). ES was greater for change in BC following a loading-only CS regimen (0.26 +/- 0.03, p=.0003) compared to a maintenance regimen (0.04 +/- 0.05), for repetitive-bout (0.25 +/- 0.03,p=.028) compared to single-bout (0.18 +/- 0.02) exercise, and for upper-body exercise (0.42 +/- 0.07, p<.0001) compared to lower (0.21 +/- 0.02) and total body (0.13 +/- 0.04) exercise. ES for laboratory-based tasks (e.g., isometric/isotonic/isokinetic exercise, 0.25 +/- 0.02) were greater (p=.014) than those observed for field-based tasks (e.g., running, swimming, 0.14 +/- 0.04). There were no differences in BC or performance ES between males and females or between trained and untrained subjects.

Conclusion: ES was greater for changes in lean body mass following short-term CS, repetitive-bout laboratory-based exercise tasks < or = 30 s (e.g., isometric, isokinetic, and isotonic resistance exercise), and upper-body exercise. CS does not appear to be effective in improving running and swimming performance. There is no evidence in the literature of an effect of gender or training status on ES following CS.