REF-196 · Valenzuela et al., 2022b · International Journal of Sports Physiology and Performance · Paper
Durability in Professional Cyclists: A Field Study on the Relationship Between Physiological Characteristics and Fatigue-Induced Power Decline
Protocol
n=12 male professional cyclists, age 26 (5) y, VO2max 83.0 (3.6) mL·kg⁻¹·min⁻¹
20-minute time trial (TT) on 2 sessions separated by 48 h: (1) with no previous fatigue (TT-Fresh) and (2) immediately after a long submaximal ride of approximately 4 h / 40 kJ/kg (TT-Fatigue)
Outcome: percentage decay in mean power output from TT-Fresh to TT-Fatigue
Correlates tested: laboratory endurance indicators from a previous maximal incremental test (ventilatory threshold, peak power output, VO2max) and training loads during the 4 weeks preceding the TTs
Key Results
Mean power fell from 386 (29) W to 375 (28) W, a decay of -2.9% (P = .007)
Large interindividual variability: range -8.5% to +1.1%, coefficient of variation 105% — some riders lost nothing, others lost over 8%
Average heart rate did not differ between the two TTs (177 (7) vs 176 (6) beats·min⁻¹, P = .118)
No significant associations between the power decay and any of the analysed indicators — ventilatory threshold, peak power output, VO2max or preceding 4-week training load (all P > .05)
Fat oxidation, FTP and body composition were not among the indicators analysed in the abstract
Practical Application
In 12 professional cyclists (VO2max 83.0), a 20-min time trial done after ~4 h / 40 kJ/kg of submaximal riding produced on average 2.9% less power than the same test fresh (386 vs 375 W), with a range from -8.5% to +1.1% between riders, and the size of the loss was not related to ventilatory threshold, peak power, VO2max or the previous 4 weeks of training load — durability has to be measured directly; it cannot be read off a lab profile or a threshold test.
The authors' conclusion: performance is significantly impaired after a certain amount of work (~40 kJ/kg) even in professionals, and the impairment is not explained by "traditional" lab indicators or training-load markers — which supports assessing durability as its own parameter.
Inference, not tested: the wide spread (some riders unchanged, others down more than 8%) is why a fresh 20-min test can rank two riders one way and a fatigued test the other; for long road events the fatigued number is the relevant one.
Limitations
n=12, correlational: a null association in a sample this small does not rule out real relationships (the abstract itself calls for "confirming potential determinants")
Single fatiguing protocol (~4 h, 40 kJ/kg) at submaximal intensity; other durations and intensities untested
Professional male cohort only — transfer to amateurs is not tested here
Contrast with REF-424 (strong correlations after a high-intensity fatiguing protocol in professionals; not from this paper) — the fatiguing protocol and the analysed variables differ between the two studies
How this entry was checked and changed
Why this level
regraded 2026-09-27 from B+: correlational field study, n=12
Content correction 2026-09-27
Per-rider power table, mean -8.7% ± 2.1% and "392 → 358 W" -> the abstract reports 386 (29) → 375 (28) W, -2.9%, range -8.5% to +1.1% (CV 105%); no per-rider data in the abstract.
Correlation table (VO2max r = -0.15, FTP r = -0.08, LT r = 0.22, %fat r = -0.34, fat oxidation r = 0.61 p<0.05, training age r = -0.45) -> deleted; the abstract reports no significant associations with any analysed indicator (ventilatory threshold, peak PO, VO2max, 4-week training load), and fat oxidation, FTP, body composition and training age were not analysed.
"Fat oxidation capacity is the primary determinant of durability", the fat-oxidation mechanism block and the "Molecular Mechanisms" section (PGC-1α, carnitine shuttle, angiogenesis, fasted rides, omega-3) -> deleted; not from this paper.
Level B+ -> B (correlational field study, n=12); VO2max SD 2.5 -> 3.6; FTP 6.2 W/kg deleted (not in the abstract).
2026-10-01: cross-reference repointed from a merged framework to the one that replaced it (SYNTH-01→31, 05→34, 09→70, 12/15→32, 19→26).