REF-819 · Hawley et al., 1997 · Sports Medicine, Vol. · Paper · PubMed

Carbohydrate-Loading and Exercise Performance: An Update

B+moderate evidenceMeasured on a mixed groupWritten before October 2026, being re-checked

The review concludes that raising pre-exercise muscle glycogen above normal resting values has little or no effect on a single exhaustive high-intensity bout lasting less than 5 minutes, and no benefit for moderate-intensity running or cycling lasting 60 to 90 minutes, because substantial glycogen is still left in the working muscles at the end.

In endurance events lasting more than 90 minutes, elevated starting glycogen postpones fatigue by about 20%; exhaustion in such exercise usually coincides with critically low muscle glycogen (25 mmol/kg wet weight), suggesting blood glucose oxidation cannot replace the energy lost from glycogen.

Supercompensation may also improve set-distance performance, where high-carbohydrate diets have been reported to improve performance by 2 to 3%.

Participants

Not applicable (review). The abstract does not give the number of studies reviewed or their participants.

Protocol

Narrative review of studies that raised pre-exercise muscle glycogen and measured performance.
Findings grouped by exercise type and duration: high-intensity bouts under 5 minutes; moderate-intensity running or cycling of 60 to 90 minutes; endurance events over 90 minutes (time to fatigue); and set-distance performance (time trials).
The abstract describes no search strategy, inclusion criteria or statistical pooling.

Key Results

ExerciseEffect of raising starting muscle glycogen (per the review)
Single exhaustive high-intensity bout, less than 5 minLittle or no effect
Moderate-intensity running or cycling, 60 to 90 minNo benefit; substantial glycogen remains at the end
Endurance exercise longer than 90 min (time to fatigue)Fatigue postponed by about 20%
Set distance covered as fast as possibleHigh-carbohydrate diets reported to improve performance by 2 to 3%
At exhaustion in prolonged exercise, muscle glycogen is usually critically low (25 mmol/kg wet weight).

Practical Application

In this narrative review of carbohydrate-loading studies in endurance athletes, loading helped only when the event lasted more than 90 minutes — it postponed fatigue by about 20% and improved set-distance performance by 2 to 3% — while it did nothing for efforts under 5 minutes or for 60-90-minute running or cycling. Load for a marathon or longer; for shorter races, including a half marathon finished inside about an hour and a half, normal eating is enough on this evidence.
The duration cut-off explains the trial in REF-774, where loading did not speed up a half-marathon-length run: at that duration glycogen is not the limit.
The 20% is a time-to-fatigue figure; the race-relevant number is the 2 to 3% for set-distance performance. Inference, not tested here: the smaller figure is the one to quote to a marathoner.
Not from this paper; the base's own practice: load without a depletion phase (the moderate regimen in REF-774 matched the classic one on stored glycogen), and fuel during the race as in SYNTH-02; the review gives no dose, so any grams-per-kilogram figure is practice.

Limitations

Narrative review: study selection and weighting are not described, and the percentages are not pooled estimates with confidence intervals.
The studies behind it predate modern in-race fuelling practice; how much loading adds on top of high in-race carbohydrate intake is not addressed in the abstract.
The abstract gives no loading protocol, dose or sex breakdown.
Written from the abstract and record metadata only (full text not read).
How this entry was checked and changed
Why this level

authoritative narrative review from a leading carbohydrate-metabolism group (Hawley, Noakes); no systematic method, and the percentages are the authors' summary of the literature, not a pooled estimate