Effects of acute carbohydrate supplementation on endurance performance: a meta-analysis.

UNLABELLED: Research on the performance effects of acute carbohydrate supplementation is comprehensive. Here we present the first meta-analytic review of this research.
METHODS: Eighty-eight randomized crossover studies in which carbohydrate supplements were consumed with or without protein before and/or during exercise provided 155 estimates for performance effects in time-to-exhaustion tests or in time trials with or without a preload. For the mixed-model meta-analysis, all effects were converted into percentage changes in mean power in a non-preloaded time trial and weighted using percentage standard errors derived from exact p-values (in a minority of studies) or from estimated errors of measurement (in all other studies). Publication bias was assessed with a plot of t-values for the random-effect solutions versus standard errors. Probabilistic inferences were derived with reference to thresholds for small, moderate and large effects on performance of 0.5, 1.5 and 2.7%.
RESULTS: Publication bias was reduced by excluding studies with a standard error >1.25%. In the remaining 73 studies and 122 estimates, the meta-analysed performance effects of carbohydrate supplements ranged from clear large improvements of ∼6% to clear moderate impairments of ∼2%. The best supplement inferred from the analysis consisted of a ∼3-10% carbohydrate-plus-protein drink providing ∼0.7 g/kg/h glucose polymers, ∼0.2 g/kg/h fructose and ∼0.2 g/kg/h protein. Substantial increases in the benefit of a supplement were probably small with an additional 9-hour fast and with the inclusion of ∼0.2 g/kg/h of protein, probably small to moderate with ingesting the first bolus not at the start of exercise but 1-4 hours before exercise, and possibly small with increasing the frequency of ingestion by three boluses per hour. Substantial reductions in the benefit of a supplement were possibly moderate with a supplement providing >0.25 g/kg/h fructose, and possibly small with an increase in ambient temperature of 10°C. The effect in subjects with maximal oxygen consumption higher by 10 mL/kg/min was probably trivial, and the effects of exercise duration were dependent on the concentration of carbohydrate plus protein in the supplement. The effect of including salt was unexpectedly trivial, and the effect of gender was unclear.
CONCLUSIONS: Carbohydrate supplements with an appropriate composition and administration regimen can have large benefits on endurance performance. More research and better reporting are required to investigate the moderating effects of gender and salt.