REF-1544 · Ronsen et al., 2004 (with companion papers 2001-2002) · European Journal of Applied Physiology · Paper · PubMed

Does a 3-h gap cost more than 6 h?

Second rides after 3 h showed higher VO2, heart rate and hormone responses than single rides; 6 h plus an extra meal went with lower values for most. Nine men, one day, no performance measured.

Main caveat: Abstracts only. All four papers; no copy in PubMed Central or Europe PMC.
limited, mixed evidenceMeasured on elite athletes

All four abstracts report nine athletes and the same bout: 75 min on a cycle ergometer, described in full as 10 min at 50% and then 65 min at 75% of VO2max. They come from one group, and the shared design makes it likely they are one programme with the same nine men (inference; no abstract says so).

The trials were a day without exercise, one bout, two bouts 3 h apart and two bouts 6 h apart. Compared with a single bout, the second bout after 3 h showed higher mean VO2, heart rate, rectal temperature and excess post-exercise oxygen consumption, a lower respiratory exchange ratio, and a higher peak in IL-6 and IL-1ra.

Compared with a first or single bout, it showed higher epinephrine, norepinephrine, ACTH, cortisol and growth hormone and lower testosterone.

Compared with 3 h, the 6-h gap with an additional meal went with lower VO2, heart rate and temperature and a higher exchange ratio in the second bout, no effect on metabolism afterwards, and smaller rises in catecholamines, ACTH, cortisol and white-cell shifts, though not in growth hormone or lymphocyte activation.

IL-6 and IL-1ra did not differ significantly between 3 h and 6 h. Our reading: these papers show a real acute carry-over from a first hard ride to a second, smaller with more time and food, in physiology and hormones. They say nothing on whether a double is better or worse for the athlete.

Protocol

Who: nine in each paper, as described under Tier. Age, VO2max and sport are not given in the abstracts.
Bout: 10 min at 50% then 65 min at 75% of VO2max on a cycle ergometer (2001, 2004); "75 min at 75%" in the two 2002 abstracts.
Timing: first bout 08:00-09:15 for the 6-h gap and 11:00-12:15 for the 3-h gap; the afternoon bout falls at the same hour in every arm, 15:15-16:30 by the 2001 abstract, so time of day is matched for that bout. The 2001 abstract prints "12:15 a.m.", plainly a slip for midday.
Food: the 2004 abstract gives one meal in the 3-h rest and two in the 6-h rest. The other three abstracts name no meal, so the food confound in the hormone, white-cell and IL-6 comparisons rests on the inference that these are one programme. Composition and amount are not in any abstract.
Rest: bed rest at all hours outside exercise in the 2001 paper; "complete bed rest" names the no-exercise arm in 2001 and in Am J Physiol 2002.
Sampling: 11 blood samples to 07:30 the next morning in the 2001 paper; trials ran 24 h or 25 h. Glycogen is not reported as measured in any abstract.
Funding and interests: PubMed lists non-US government support for all four; the funders and any conflict statements are not in the abstracts.

Key Results

ComparisonMeasureResultPaper
Second bout after 3 h vs single boutVO2, heart rate, rectal temperature, post-exercise oxygen usehigher2004
Second bout after 3 h vs single boutRespiratory exchange ratiolower2004
Second bout after 3 h vs first or singleEpinephrine, norepinephrine, ACTH, cortisol, growth hormonehigher2001
Second bout after 3 h vs first or singleTestosteronelower2001
Second bout after 3 h vs first or singleInsulin, FSH, LH, TSH, free thyroxin, IGF-1no clear difference2001
6 h + extra meal vs 3 h, second boutVO2, heart rate, temperature lower; exchange ratio higherpost-exercise metabolism: no effect found2004
3 h vs 6 h, second boutEpinephrine, norepinephrine, ACTH, cortisol; neutrophil rise, lymphocyte fallmore pronounced after 3 hAm J Physiol 2002
3 h vs 6 h, second boutGrowth hormone; lymphocyte activationno clear differenceAm J Physiol 2002
Peak IL-6, pg/mL3 h 8.8 ± 1.3; single 5.2 ± 0.7; 6 h 5.9 ± 1.23 h above single (significant); 3 h vs 6 h not significantJ Appl Physiol 2002
Peak IL-1ra, pg/mL, 1 h after3 h 1,774 ± 373; single 302 ± 53; 6 h 1,276 ± 4513 h above single (significant); 3 h vs 6 h not significantJ Appl Physiol 2002
Our arithmetic: IL-6 after 3 h was about 1.7 times the single-bout peak, IL-1ra about 5.9 times. The 6-h IL-1ra mean sits far closer to the 3-h value than to the single bout, with wide spread.
No clear difference, not "the same." Nothing here is an equivalence test, and n = 9.
Numbers. The other abstracts print no values for VO2, heart rate, temperature or hormones, so the size of those differences is not known from what was read.

Mechanisms

Measured: larger metabolic, sympatho-adrenal, pituitary-adrenal and cytokine responses to a second bout at 3 h; smaller metabolic and hormonal responses at 6 h with more food.
Authors' interpretation: the J Appl Physiol 2002 abstract suggests the IL-6 and IL-1ra rise may be linked to glycogen depletion. No abstract reports glycogen, so this remains a proposal from what was read.
Inference: a lower exchange ratio in the second bout is compatible with less carbohydrate use; these abstracts do not test why.

Practical Application

Does a 3-h gap cost more than 6 h? Second rides after 3 h showed higher VO2, heart rate and hormone responses than single rides; 6 h plus an extra meal went with lower values for most. Nine men, one day, no performance measured.
Good for: expecting a higher heart rate and oxygen use at the same prescribed intensity in a second session soon after a first, in trained people on the same day.
Good for: a reason not to read 3-h and 6-h gaps as interchangeable, while saying plainly that the 6-h arm also ate more.
Not good for: a rule on the best gap. Only two gaps were tested, confounded with food, and neither is linked here to adaptation or to performance in the second bout.
Not good for: the claim that doubles are bad. A larger acute response is not harm, and it could equally be the stimulus (inference); these papers cannot tell.
Not good for: interval sessions, split sessions of matched total work, or easy doubles. These are two continuous 75-min rides at one intensity.
Not from this paper: spacing hard sessions as far apart as the day allows and eating in the gap is common practice; here it rests on one day of physiology in nine men.

Limitations

Abstracts only. All four papers; no copy in PubMed Central or Europe PMC. Order of trials, randomisation, diet, absolute workloads and exact statistics for most outcomes were not read.
Gap and food moved together. The 6-h arm had one more meal, so any 3 h against 6 h difference cannot be credited to time alone.
The double adds work. Against the single bout, the double arm rode 150 min instead of 75 (our arithmetic), so this is more exercise as well as repeated exercise.
Not independent. Four papers, probably one set of nine men and one set of trials (inference). They should count as one programme of evidence, not four replications.
Multiple outcomes, small n. Many hormones and cell types were tested in nine people across several papers.
Bed rest between bouts, stated in the 2001 paper, is not how anyone lives on a training day.
Not measured, on the abstracts: performance in the second bout, glycogen, training adaptation, sleep, more than one day, any woman by stated sex.
How this entry was checked and changed
Why this level

(from the abstracts) — a crossover under 15 earns B by design, lowered one step because the outcomes are acute physiology and blood markers rather than performance, and because all four papers were read as abstracts only, so order, diet and glycogen methods could not be checked