REF-979 · Costa et al., 2017 · Applied Physiology, Nutrition, and Metabolism, Vol. · Paper · PubMed

Gut-training: the impact of two weeks repetitive gut-challenge during exercise on gastrointestinal status, glucose availability, fuel kinetics, and running performance

B+moderate evidenceMeasured on trained to highly trained athletesWritten before October 2026, being re-checked

Endurance runners (n = 25) ran a gut-challenge trial: 2 h at 60% VO2max while consuming gel-discs containing 30 g carbohydrates (2:1 glucose/fructose, 10% w/v) every 20 min, then a 1 h distance test. They were then randomly assigned to 2 weeks of repetitive gut-challenge with a carbohydrate gel-disc (CHO-S), carbohydrate food (CHO-F) or placebo (PLA), and repeated the trial.

Gastrointestinal symptoms fell on CHO-S (60%) and CHO-F (63%), more than on placebo; the distance test improved on CHO-S (5.2%) and CHO-F (4.3%) but not on placebo (-2.1%). The gel-disc group also showed a lower breath-hydrogen peak and higher blood glucose. Oxidation rates, plasma I-FABP and cortisol did not differ.

Participants

Endurance runners (n = 25), randomly assigned to three groups (CHO-S, CHO-F, PLA). Group sizes, sex and fitness are not given in the abstract.

Protocol

Gut-challenge trial (GC1, repeated as GC2): 2 h running at 60% VO2max while consuming gel-discs containing 30 g carbohydrates (2:1 glucose/fructose, 10% w/v) every 20 min, followed by a 1 h distance test.
Intervention: 2 weeks of repetitive gut-challenge with a carbohydrate gel-disc (CHO-S), carbohydrate food (CHO-F) or placebo (PLA).
Measures: gastrointestinal symptoms, breath H2 (malabsorption), blood glucose, oxidation rates, plasma I-FABP, cortisol, distance covered.

Key Results

Gastrointestinal symptoms reduced in GC2 on CHO-S (60%; p = 0.008) and CHO-F (63%; p = 0.046); the reductions were greater than PLA (p < 0.05).
Distance test: improved on CHO-S (5.2%) and CHO-F (4.3%), not on PLA (-2.1%) (trial × time: p = 0.009).
H2 peak in GC2: CHO-S 6 (4-8) ppm vs CHO-F 9 (6-12) ppm and PLA 12 (2-21) ppm (mean (CI); trial × time: p < 0.001).
Blood glucose in GC2: CHO-S 7.2 (6.3-8.1) mmol/L vs CHO-F 6.1 (5.7-6.5) mmol/L and PLA 6.2 (4.9-7.5) mmol/L (trial × time: p = 0.015).
No difference in oxidation rates, plasma I-FABP or cortisol between groups and trials.

Practical Application

"Can I train my gut, and how long does it take?" In 25 endurance runners randomised to 2 weeks of repeated carbohydrate intake while running (gel-disc or food) or placebo, gut symptoms during a 2 h run with 30 g of carbohydrate every 20 min fell by 60% and 63% in the carbohydrate groups, more than on placebo, and the 1 h distance test improved 5.2% and 4.3% against -2.1%. Two weeks of practising the intake was enough to see it.
Both carbohydrate forms worked for symptoms and performance; only the gel-disc (the same product as in the test) lowered the malabsorption marker and raised blood glucose. Practise with the product you will race on.
The benefit did not come through intestinal injury or fuel use: I-FABP and oxidation rates did not change.
Not from this paper; the base's own practice: change one variable per long session, and build the rate by the ramp in SYNTH-02; the troubleshooting order is SYNTH-08.

Limitations

n = 25 split across three groups; group sizes are not in the abstract.
Laboratory running at 60% VO2max for 2 h; no race, no heat, no cycling.
The size and frequency of the training feedings over the 2 weeks are not given in the abstract.
Symptoms are self-reported; blinding of the food arm is not described in the abstract.
Inference, not stated in the abstract: the cohort may overlap with REF-980 (same laboratory, same protocol, same period) — do not count the two as fully independent.
Written from the abstract and record metadata only (full text not read).
How this entry was checked and changed
Why this level

parallel RCT, n=25 across three groups (group sizes not in the abstract)