REF-828 · Lemire et al., 2021 · Journal of Science and Medicine in Sport, Vol. · Paper · PubMed

Physiological factors determining downhill vs uphill running endurance performance

Blimited evidenceMeasured on highly trained athletesWritten before October 2026, being re-checked

Ten highly trained men from endurance sports did 5-km time trials outdoors, one downhill and one uphill (average grade ±8%), plus tests of maximal oxygen uptake, maximal strength of the lower-limb extensors, leg musculotendinous stiffness, local muscle endurance, vertical jump, sprint velocity and explosivity/agility.

They ran faster downhill than uphill (20.4 ± 1.0 vs 12.0 ± 0.5 km/h) with a similar mean heart rate (95 ± 2% vs 94 ± 2% of maximum) yet a lower mean VO2 (85 ± 8% vs 89 ± 7% of VO2max).

Uphill performance was predicted by velocity at VO2max, BMI and maximal extensor strength (r2 = 0.94); downhill performance by velocity at VO2max, leg musculotendinous stiffness and maximal extensor strength (r2 = 0.84).

The two efforts share two predictors (vVO2max and maximal strength), weighted differently, and each has one specific third predictor: BMI for uphill, leg stiffness for downhill.

Participants

Ten highly trained male endurance athletes. The abstract does not give age or training history.

Protocol

Outdoor 5-km downhill (DR) and uphill (UR) time trials, average grade ±8%.
Laboratory and field tests: maximal oxygen uptake, maximal strength of the lower-limb extensors, leg musculotendinous stiffness, local muscle endurance, vertical jump ability, sprint velocity and explosivity/agility.
Correlation and commonality regression analyses to find predictors of DR and UR performance.

Key Results

Downhill 5-kmUphill 5-km
Running velocity20.4 ± 1.0 km/h12.0 ± 0.5 km/h
Average heart rate (% of maximum)95 ± 2%94 ± 2%
Average VO2 (% of VO2max)85 ± 8%89 ± 7%
Significant predictorsvVO2max, leg musculotendinous stiffness, maximal extensor strengthvVO2max, BMI, maximal extensor strength
Variance explainedr2 = 0.84r2 = 0.94
Velocity differed between conditions (p < 0.05); heart rate did not (p > 0.05); VO2 was lower downhill (p < 0.05).

Practical Application

In ten highly trained male runners, a 5-km downhill and a 5-km uphill time trial were each explained by three predictors: both depended on velocity at VO2max and maximal leg-extensor strength, but the uphill added body mass index and the downhill added leg musculotendinous stiffness (r2 = 0.94 uphill, 0.84 downhill). Climbing and descending are separate skills: an aerobic engine and strong legs help both, a lighter body helps on the climbs, and a stiff, spring-like leg helps on the descents.
Heart rate was the same up and down (95 vs 94% of maximum) although VO2 was lower downhill (85 vs 89% of VO2max). Inference, not tested here: heart rate over-reads the metabolic cost of a fast descent, so it is a poor guide to downhill effort.
Not from this paper; the base's own practice: train downhill running as a skill with gradual exposure (REF-660, REF-782) and climbing with uphill intervals (REF-781); keep heavy strength work in the week (SYNTH-44).
The study did not test a stiffness-training programme; "leg stiffness" here is a measured trait, not a prescription.

Limitations

Ten men only; the predictors come from regression in a very small sample and risk overfitting.
Short (5-km) time trials at one average grade; long races add fatigue and muscle damage that were not studied.
Correlational: no training intervention.
The "randomized controlled trial" label refers to the order of the two time trials as far as the abstract shows; there is no control group.
Written from the abstract and record metadata only (full text not read).
How this entry was checked and changed
Why this level

crossover, n=10 men; the determinant analysis is correlational in ten runners