High-intensity downhill running exacerbates heart rate and muscular fatigue in trail runners

Abstract

This study explores the cardiorespiratory and muscular fatigue responses to downhill (DR) vs uphill running (UR) at similar running speed or similar oxygen uptake (V̇O2). Eight well-trained, male, trail runners completed a maximal level incremental test and three 15-min treadmill running trials at ±15% slope: i) DR at ~6 km·h−1 and ~19% V̇O2max (LDR); ii) UR at ~6 km·h−1 and ~70% V̇O2max (HUR); iii) DR at 19 km·h−1 and ~70% V̇O2max (HDR). Cardiorespiratory responses and spatiotemporal gait parameters were measured continuously. Maximal isometric torque was assessed before and after each trial for hip and knee extensors and plantar flexor muscles. At similar speed (~6 km·h−1), cardiorespiratory responses were attenuated in LDR vs HUR with altered running kinematics (all p < 0.05). At similar V̇O2 (~3 l·min−1), heart rate, pulmonary ventilation and breathing frequency were exacerbated in HDR vs HUR (p < 0.01), with reduced torque in knee (−15%) and hip (−11%) extensors and altered spatiotemporal gait parameters (all p < 0.01). Despite submaximal metabolic intensity (70% V̇O2max), heart rate and respiratory frequency reached maximal values in HDR. These results further our understanding of the particular cardiorespiratory and muscular fatigue responses to DR and provide the bases for future DR training programs for trail runners.

Full text

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rjsp20
Journal of Sports Sciences
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rjsp20
High-intensity downhill running exacerbates heart
rate and muscular fatigue in trail runners
Marcel Lemire , Romain Remetter , Thomas J. Hureau , Blah Y. L. Kouassi ,
Evelyne Lonsdorfer , Bernard Geny , Marie-Eve Isner-Horobeti , Fabrice
Favret & Stéphane P. Dufour
To cite this article: Marcel Lemire , Romain Remetter , Thomas J. Hureau , Blah Y. L. Kouassi ,
Evelyne Lonsdorfer , Bernard Geny , Marie-Eve Isner-Horobeti , Fabrice Favret & Stéphane P.
Dufour (2020): High-intensity downhill running exacerbates heart rate and muscular fatigue in trail
runners, Journal of Sports Sciences
To link to this article: https://doi.org/10.1080/02640414.2020.1847502
Published online: 15 Nov 2020.
Submit your article to this journal
View related articles
View Crossmark data

SPORTS PERFORMANCE
High-intensity downhill running exacerbates heart rate and muscular fatigue in trail
runners
Marcel Lemire
a,b
, Romain Remetter
a,c
, Thomas J. Hureau
a,b
, Blah Y. L. Kouassi
a,b
, Evelyne Lonsdorfer
a,c
, Bernard Geny
a,c
,
Marie-Eve Isner-Horobeti
a,d
, Fabrice Favret
a,b
and Stéphane P. Dufour
a,b
a
Faculty of Medicine, University of Strasbourg, Translational Medicine Federation (FMTS), Strasbourg, France;
b
Faculty of Sport Sciences, University of
Strasbourg, Strasbourg, France;
c
Physiology and Functional Explorations Department, University Hospitals of Strasbourg, Civil Hospital, Strasbourg,
France;
d
Physical and Rehabilitation Medicine Department, University of Strasbourg, University Institute of Rehabilitation Clémenceau, Strasbourg,
France
ABSTRACT
This study explores the cardiorespiratory and muscular fatigue responses to downhill (DR) vs uphill
running (UR) at similar running speed or similar oxygen uptake (V
̇O
2
). Eight well-trained, male, trail
runners completed a maximal level incremental test and three 15-min treadmill running trials at ±15%
slope: i) DR at ~6 km·h
−1
and ~19% V
̇O
2max
(LDR); ii) UR at ~6 km·h
−1
and ~70% V
̇O
2max
(HUR); iii) DR at
~19 km·h
−1
and ~70% V
̇O
2max
(HDR). Cardiorespiratory responses and spatiotemporal gait parameters
were measured continuously. Maximal isometric torque was assessed before and after each trial for hip
and knee extensors and plantar exor muscles. At similar speed (~6 km·h
−1
), cardiorespiratory responses
were attenuated in LDR vs HUR with altered running kinematics (all p < 0.05). At similar V
̇O
2
(~3 l·min
−1
),
heart rate, pulmonary ventilation and breathing frequency were exacerbated in HDR vs HUR (p < 0.01),
with reduced torque in knee (−15%) and hip (−11%) extensors and altered spatiotemporal gait para-
meters (all p < 0.01). Despite submaximal metabolic intensity (70% V
̇O
2max
), heart rate and respiratory
frequency reached maximal values in HDR. These results further our understanding of the particular
cardiorespiratory and muscular fatigue responses to DR and provide the bases for future DR training
programs for trail runners.
ARTICLE HISTORY
Accepted 4 November 2020
KEYWORDS
Oxygen uptake; heart rate;
muscle torque; running
kinematics; inclined
treadmill
Introduction
Paragraph 1. Trail running has become very popular over the
last 20 years. These outdoor mountain running races extend
from short trails to extreme events (i.e., ultra-long distances
and/or large cumulative ascent-descent) but systematically
involve uphill (UR) as well as downhill running (DR) sections.
Physiological studies have highlighted the marked dierences
in DR vs UR with direct consequences in pacing as well as in
training strategies (Born et al., 2016). Most of the dierences
between DR and UR are considered to be related to the type of
muscle actions on which they rely (Vernillo et al., 2016). For
instance, DR is known to require preferentially eccentric muscle
actions (i.e., lengthening muscle actions) to generate the neces-
sary braking component within the major extensor muscles of
the lower limbs whereas UR predominantly relies on concentric
muscle actions (i.e., shortening muscle actions) to move the
runner’s body up despite gravity (Lindstedt et al., 2001).
Paragraph 2. Downhill running is characterized by the high
running speeds that can be achieved despite very low energy
requirements. Conversely, UR features much lower running
velocities but considerably greater energy cost (Minetti et al.,
2002). When both running modes are compared at similar
slopes and running speeds (thereby limiting the comparison
to very low running speed in UR) (Lemire et al., 2018) or similar
heart rate (HR) (Garnier et al., 2019), oxygen uptake (V
̇O
2
) and
ventilation (V
̇
E
) are lower in DR. However, when DR and UR are
compared at the same slope and similar V
̇O
2
, running speed is
much greater in DR, together with higher HR (Kolkhorst et al.,
1996; Lemire et al., 2020b). Previous studies in this area have
used either short stages during incremental tests (i.e., 2 min
stages) (Breiner et al., 2018; Kolkhorst et al., 1996; Lemire et al.,
2020b) and/or were limited to low exercise intensities (i.e., V
̇O
2
levels < 40 ml∙kg
−1
∙min
−1
) (Kolkhorst et al., 1996; Pokora et al.,
2014). Moreover, the ability to maintain a stable V
̇O
2
and there-
fore the tolerance to high-intensity DR at steep slopes (i.e.,
>10%) has not been specically addressed. Indeed, UR at 10%
slope has been shown to increase the amplitude of the V
̇O
2
slow component compared to level running (Pringle et al.,
2002). Although an upward drift in V
̇O
2
has been described
during low intensity (i.e., <50% V
̇O
2max
) prolonged DR (Dick &
Cavanagh, 1987; Westerlind et al., 1994), the behaviour of V
̇O
2
and other cardiorespiratory parameters during high-intensity
constant speed DR at steep slope remains unexplored, albeit
this knowledge would help prescribing specic training ses-
sions and designing training programs for trail runners.
Paragraph 3. Despite the low energy cost of DR, some
recent studies have shown that trail races can induce marked
muscular fatigue with up to 40% loss of lower limb extensors
muscle strength (hip, knee and ankle) (Baiget et al., 2016;
Easthope et al., 2010). This fatigue is considered to be mainly
related to a combination of local muscle metabolic perturba-
tions potentially altering muscle contractile processes and/or
CONTACT Marcel Lemire marcel.lemire@unistra.fr Faculty of Sport Sciences, University of Strasbourg, Strasbourg 67084, France
JOURNAL OF SPORTS SCIENCES
https://doi.org/10.1080/02640414.2020.1847502
© 2020 Informa UK Limited, trading as Taylor & Francis Group

mechanical loading on the muscle-tendon complex in the
lower limbs, possibly leading to profound skeletal muscle
damage (Giandolini et al., 2016b). In this context, signicant
reductions in knee extensor torque have been observed after
both pure UR and DR (Giandolini et al., 2016b; Lazzer et al.,
2015), but the respective contribution of the mechanical and
metabolic loads to the muscular fatigue observed after DR vs
UR remains unclear.
Paragraph 4. Therefore, the main purpose of this study
was to determine the cardiorespiratory responses of well-
trained trail runners to DR vs UR at steep slopes performed
at similar running speed and at similar oxygen uptake (i.e.,
high intensity running exercise). A secondary purpose of
this study was to explore the exercise-induced torque loss
observed after high-intensity DR vs UR performed at steep
slopes.
Materials and methods
Participants and ethical approval
Paragraph 5. Eight well-trained male trail runners partici-
pated in this study (age: 29 ± 10 [mean ± SD] years; height:
1.74 ± 0.03 m; body mass: 61.8 ± 5.0 kg; BMI:
20.4 ± 2.0 kg·m2; maximal oxygen uptake (V
̇O
2max
):
68.0 ± 6.4 ml·min
−1
·kg
−1
; maximal HR: 183 ± 8 bpm;
12 ± 4 years training history and 2503 ± 1092 km·year
−1
of training). The number of subjects included in this study
has been calculated to detect a 10% torque loss in DR
(Giandolini et al., 2016a). Therefore, a statistical power of
0.9 and an alpha risk of 0.05 required 8 subjects to be
included. All athletes were informed of the benets and
risks of this investigation prior to giving their written
informed consent to participate in the study. They were
also instructed not to consume alcohol or caeine in the
3 h before each test and to refrain from strenuous and
exhaustive exercise 24 h prior to each test. All were healthy,
without current injuries and did not take any medication.
The experiment was previously approved by our
Institutional Review Board and complied with the
Declaration of Helsinki (CPP18-039a/2108-A00700-55).
Experimental design
Paragraph 6. In eight separate sessions (Figure 1), each athlete
performed each of the following experimental test: i) a maximal
level running incremental test on a treadmill to determine level
running V
̇O
2max
; ii) four DR familiarization sessions; iii) three
constant speed inclined (+ or – 15% slope) running trials (one
UR and two DR). All constant speed trials were randomized,
except high-intensity UR which had to be performed before
low-intensity DR at the same running speed. All trials were
performed at the same time of the day and were separated
by at least 4 days.
Familiarization sessions to downhill running
Paragraph 7. To minimize muscle damage, the subjects were
familiarized with DR using 4 separate familiarization trials with
a constant negative slope of −15%: 1) 5-min DR at 60–70% level
running speed at V
̇O
2max
(vV
̇O
2max
), 2) 10 min DR at 60–90% v_
V
O
2max
(4 min at 60% vV
̇O
2max
, then 2 min stages with 10%
vV
̇O
2max
increment), 3) 16 min DR (5 min DR at 60%, 5 min at
80%, 3 times 1 min at 100% vV
̇O
2max
and 1 min DR recovery at
60% vV
̇O
2max
between each bout), 4) the same workout than
the third familiarization session except that the intensity of
3 × 1 min trials was set at 120% vV
̇O
2max
.
Incremental and constant speed running tests
Paragraph 8. All athletes performed an incremental test until
exhaustion on a level treadmill (Pulsar, H/P Cosmos, Nussdorf-
Traunstein, Germany). The participants began the rst stage at
13 km·h
−1
during 2 min and the running speed was increased
by 1 km·h
−1
every 2 min until volitional exhaustion.
Paragraph 9. Three constant speed running trials of
15 min duration were also performed by all athletes: i)
one DR bout at 70% V
̇O
2max
(−15% slope, HDR), ii) one UR
bout at 70% V
̇O
2max
(+15% slope, HUR) and iii) one DR bout
at similar treadmill speed to the UR trial (−15% slope, LDR).
Previous data from the same subjects collected during DR
(−15%) and UR (+15%) maximal incremental tests (Lemire
et al., 2020b) were used to establish the running speed
Figure 1. Protocol design.
2M. LEMIRE ET AL.

allowing to reach 70% V
̇O
2max
during HUR and HDR trials.
Pilot testing also demonstrated that 70% V
̇O
2max
was the
highest exercise intensity that our athletes were able to
maintain for 15 minutes during DR at −15% slope.
Gas exchange measurements
Paragraph 10. During all running trials, V
̇O
2
, carbon dioxide
output (V
̇CO
2
), V
̇
E
, respiratory frequency, tidal volume (V
T
) and
respiratory exchange ratio (RER) were collected breath-by-
breath though a facemask with an open-circuit metabolic cart
with rapid O
2
and CO
2
analysers (Metamax Cortex, Leipzig,
Germany). Before each exercise test, the pneumotachograph
was calibrated according to manufacturer’s instructions. Heart
rate was continuously measured (Polar S810, Polar, Kempele,
Finland). Maximal oxygen consumption was dened as the
highest 30 s V
̇O
2
value. The level running speed associated with
V
̇O
2max
(vV
̇O
2max
) was determined where V
̇O
2
reaches a plateau
or if the increase in V
̇O
2
was less than 2.1 ml·kg
−1
·min
−1
for
a speed increase equal to 1 km·h
−1
(Billat & Koralsztein, 1996).
During the constant speed trials, slow components in the car-
diorespiratory responses were established using the dierence
between end-exercise and 3
rd
minute values (Poole & Jones,
2012).
Blood lactate analyses
Paragraph 11. Blood lactate concentration (b[La]) was
assessed from earlobe blood samples (200 µl of blood) before
and after 1 and 3 min of recovery after each running test
(Lactate Scout+, EKF Diagnostics, Leipzig, Germany).
Isometric maximal torque assessment
Paragraph 12. Before and within the rst 2 min after each
constant speed running trial, hip extensor, knee extensor and
plantar exor strengths were assessed using a handheld
dynamometer (microFET®2; Hoggan Scientic, Salt Lake City,
UT, USA) during isometric maximal voluntary contraction. The
coecient of variation for force measurement using this device
has previously been demonstrated to be 3.2–4.2%, with limits
of agreement of 13.8–19.2% (Mentiplay et al., 2015; Mu et al.,
2016). All tests were performed with the right lower limb for
each athlete (dominant limb), in duplicates with 1-min rest
between attempts in the following order: plantar exor, knee
extensor and hip extensor. The same experimenter conducted
all the measures and was standing systematically with his back
or his elbow against the wall to secure a solid position during all
maximal voluntary contractions (Figure 2). These measure-
ments were initially collected in Newton and subsequently
converted in torque using anatomic measures of hip and
knee extensors and plantar exor lever arms. The best value
was kept for analyses.
Hip Extensors. In order to assess isometric maximal volun-
tary contraction in hip extensors the participant was asked to
place his hip exed at 90° in dorsal decubitus position on the
examination table (Figure 2(c)). The dynamometer was placed
on the posterior side of the thigh, proximal to the knee joint
and the participant was asked to perform a maximal isometric
hip extension contraction for 5 s.
Knee Extensors. The athlete sat comfortably on the edge of
the examination table with a high-density corner cushion
Figure 2. Schematic representation of isometric maximal voluntary contraction measurement for knee extension, plantar flexion, and hip extension (Panel a, b and
c respectively).
JOURNAL OF SPORTS SCIENCES 3

under the knees, with hips and knees exed at 90° (Figure 2(a)).
The dynamometer was placed on the anterior aspect of the
shank, just above the ankle joint and the participant was asked
to perform a maximal isometric knee extension contraction for
5 s.
Plantar Flexors. The participant was lying supine with ankle
at 0°and hips and knees at 180°. The dynamometer was placed
under the metatarsal heads (Figure 2(b)). The participant was
then asked to perform a maximal isometric plantar exion
contraction for 5 s.
Spatiotemporal parameters
Paragraph 13. Spatiotemporal parameters were measured
using the OptoGait system (OptoGait; Microgate, Bolzano,
Italy) (Healy et al., 2019). The device used two parallel bars,
covering the length of the treadmill belt, that were placed on
the side edges of the treadmill as near as possible to the
contact surface. Before each test, the device was systemically
calibrated for each subject and the shoes length was measured
as a number of LEDs by maintaining a foot on the treadmill
between the bars. Contact time, step length and step frequency
were measured over 30 s at the 3
rd
and the 15
th
min for the
constant running speed trials.
Statistical analyses
Paragraph 14. Statistical analyses were performed using
Statistica (13.5, Tulsa, Oklahoma, USA). All data are expressed
as mean ± standard derivation (SD). After testing for data
distribution normality with Kolmogorov–Smirnov test and
equal variances with Bartlett test, two-way analysis of variance
(ANOVAs) on repeated measures were performed to assess the
eect of exercise time (baseline, 3 min and 15 min) and exercise
modes (HDR, HUR, LDR) on the cardiorespiratory responses and
the spatiotemporal parameters. One-way ANOVAs on repeated
measures were performed to assess the eect of exercise mode
on linear regression slope and y-intercept of the relationship
between HR and V
̇O
2
. Three-way ANOVAs on repeated mea-
sures were also performed to assess the eect of exercise time,
exercise mode and joint on lower limb torque (ankle, knee and
hip extensors). When signicant interactions were observed,
Tukey’s honestly signicant dierence post-hoc tests were
used to localize the signicant dierences. For all statistical
analyses, p ≤ 0.05 was considered statistically signicant.
Results
All cardiorespiratory parameters are presented in Table 1 and
no dierence occurred at rest among conditions (p > 0.05)
(Table 1).
Comparing downhill and uphill running at similar running
speed
Paragraph 15. At similar running and vertical velocities, V
̇O
2
and V
̇CO
2
were ~3 fold higher (p < 0.001) despite similar RER in
HUR vs LDR at the 3
rd
and 15
th
min of exercise (Figure 3(a) and
Table 1). Blood lactate was not signicantly dierent between
conditions at 1- and 3-min after end-exercise (p = 0.807 and
p = 0.381, respectively).
Pulmonary ventilation was also ~1.8 and 2.4-fold higher in
HUR than in LDR after 3 and 15 min, respectively (both
p < 0.001). The greater V
̇
E
response in HUR vs LDR was asso-
ciated with a higher tidal volume (all p < 0.001) despite similar
respiratory frequency at the 3
rd
min of exercise (p = 0.487;
Figure 4(a-c)) whereas both tidal volume and respiratory fre-
quency were greater in HUR vs LDR after 15 min exercise
(p < 0.001 and p = 0.005, respectively). Despite similar running
velocities, HR was higher in HUR vs LDR at both the 3
rd
and 15
th
min time points (Figure 3(b), p < 0.001).
All cardiorespiratory parameters achieved a steady state
with no further changes between the 3
rd
and 15
th
min during
LDR and HUR (p = 0.734–1.000), except for HR which signi-
cantly increased during HUR only (p = 0.007).
No dierence was observed in hip extensor, knee extensor and
plantar exor isometric torque before HUR and LDR (Figure 5), and
these values remained unchanged post-exercise (both p = 1.000).
While stride frequency was ~8% higher in HUR than in LDR
at the 3
rd
min and ~11% higher at the 15
th
min (all p < 0.001),
stride length was ~9% longer and contact time was ~40%
greater in LDR than in HUR at both the 3
rd
and 15
th
min time
Table 1. Metabolic and cardiorespiratory responses to downhill vs uphill running at similar speed or oxygen uptake.
Time Rest 3 min 15 min
Condition HDR HUR LDR HDR HUR LDR HDR HUR LDR
Treadmill speed (km·h
−1
) - - - 18.9 ± 2.0† 6.2 ± 0.7 6.2 ± 0.7 18.9 ± 2.0† 6.2 ± 0.7 6.2 ± 0.7
Vertical speed (m·s
−1
) - - - −0.78 ± 0.08† 0.26 ± 0.03 −0.26 ± 0.03† −0.78 ± 0.08† 0.26 ± 0.03 −0.26 ± 0.03†
V

O
2
(l·min
−1
) 0.311 ± 0.139 0.315 ± 0.137 0.318 ± 0.107 2.575 ± 0.282†‡ 2.876 ± 0.199‡ 0.804 ± 0.217†‡ 2.946 ± 0.379§ 2.949 ± 0.225 0.796 ± 0.208†
V

O
2
(ml·kg
−1
·min
−1
) 7.3 ± 2.1 6.7 ± 1.8 5.5 ± 1.6 41.8 ± 4.7†‡ 46.5 ± 3.5‡ 13.0 ± 3.3†‡ 47.7 ± 6.4§ 47.9 ± 3.3 12.8 ± 3.2†
b[La] (mmol·l
−1
) R1 1.2 ± 0.4 1.1 ± 0.5 1.3 ± 0.5 - - - 2.9 ± 1.1‡ 1.8 ± 0.4 1.2 ± 0.4
b[La] (mmol·l
−1
) R3 3.3 ± 1.9†‡ 1.8 ± 0.7 0.9 ± 0.4
RER 0.78 ± 0.06 0.76 ± 0.06 0.78 ± 0.7 0.83 ± 0.04‡ 0.84 ± 0.04‡ 0.90 ± 0.06‡ 0.90 ± 0.09 0.86 ± 0.06 0.80 ± 0.06§
V

E
(l·min
−1
) 9.8 ± 3.5 10.0 ± 4.9 10.1 ± 3.0 72.7 ± 9.6‡ 66.3 ± 8.1‡ 24.0 ± 5.5†‡ 96.7 ± 17.4§ 70.4 ± 7.7 21.2 ± 5.0†
RF (breaths·min
−1
) 17.5 ± 4.6 16.1 ± 4.3 16.9 ± 3.4 47.3 ± 7.2†‡ 32.5 ± 5.3‡ 28.1 ± 4.8‡ 56.1 ± 9.3§ 35.8 ± 5.9 26.9 ± 3.7†
V
T
(l) 0.59 ± 0.24 0.63 ± 0.20 0.61 ± 0.24 1.61 ± 0.32†‡ 2.06 ± 0.26‡ 0.88 ± 0.26†‡ 1.73 ± 0.39 2.03 ± 0.40 0.81 ± 0.36†
t
I
(s) 1.26 ± 0.34 1.24 ± 0.26 1.27 ± 0.32 0.67 ± 0.13†‡ 0.92 ± 0.18‡ 0.96 ± 0.26‡ 0.53 ± 0.20† 0.85 ± 0.24 0.98 ± 0.25
t
E
(s) 1.95 ± 0.49 2.10 ± 0.59 2.06 ± 0.41 0.67 ± 0.12‡ 0.98 ± 0.14‡ 1.15 ± 0.24‡ 0.56 ± 0.28 0.90 ± 0.27 1.19 ± 0.45
t
I
/t
TOT
0.39 ± 0.06 0.38 ± 0.05 0.39 ± 0.04 0.50 ± 0.04‡ 0.48 ± 0.02‡ 0.46 ± 0.05‡ 0.49 ± 0.04 0.48 ± 0.03 0.45 ± 0.05
HR (bpm) 66 ± 17 58 ± 13 62 ± 15 157 ± 10†‡ 141 ± 9‡ 92 ± 12†‡ 179 ± 13†§ 151 ± 12§ 93 ± 12†
Note. Values are means ± SD of 8 athletes in all conditions. HDR: downhill running at 70% V
̇O
2max
, HUR: uphill running at 70% V
̇O
2max
and LDR: downhill running at
similar running speed than HUR. † p < 0.05 vs HUR at the same time point, ‡ p < 0.05 vs rest in the same condition and § p < 0.05 vs 3 min in the same condition;
oxygen uptake (V
̇O
2
), blood lactate (b[La]) 1- (R1) and 3-min (R3) after end-exercise, respiratory exchange ratio (RER), minute pulmonary ventilation (V
̇
E
), respiratory
frequency (RF), tidal volume (V
T
), inspiration time (t
I
), expiration time (t
E
), duty cycle (t
I
/t
TOT
), heart rate (HR).
4M. LEMIRE ET AL.

points (Figure 6; all p < 0.001). No change was noted in any of
these parameters between the 3
rd
and 15
th
min of exercise.
Comparing downhill and uphill running at similar oxygen
uptake
Paragraph 16. Despite ~2-fold greater treadmill and vertical
running velocities in HDR vs HUR (Table 1), V
̇O
2
and V
̇CO
2
were
~10% lower in HDR vs HUR after 3 min of exercise (p < 0.004).
This dierence did not persist further (Figure 3(a)), such that
similar V
̇O
2
, V
̇CO
2
and RER were observed between trials at the
15
th
min (p = 1.000, 0.766 and 0.668, respectively), due the
progressive emergence of signicant V
̇O
2
and V
̇CO
2
slow com-
ponents during HDR (p < 0.001). Similar V
̇O
2
and V
̇CO
2
were also
observed when averaged between the 3
rd
and 15
th
min in each
trial (V
̇O
2
:2.828 ± 0.285 l·min
−1
in HDR vs 2.934 ± 0.218 l·min
−1
in
HUR, p = 0.163; V
̇CO
2
: 2.495 ± 0.246 l·min
−1
in HDR vs
2.518 ± 0.192 l·min
−1
in HUR, p = 0.800). At end exercise, b[La]
was similar after 1 min recovery (p = 0.131), but higher after
3 min recovery in HDR vs HUR (p = 0.017).
Heart rate was greater in HDR vs HUR at both the 3
rd
and
15
th
min time points (Table 1). Pulmonary ventilation was simi-
lar between trials after 3 min of exercise but became increas-
ingly greater thereafter in HDR vs HUR, describing a V
̇
E
slow
component (p < 0.001, Figure 4(a)). The larger V
̇
E
response
observed in HDR was accompanied by lower tidal volume but
higher respiratory frequency (p = 0.003 and p < 0.001, respec-
tively, Table 1).
The HR/V
̇O
2
and V
̇
E
/V
̇O
2
relationships were dierent
between HDR and HUR either because of greater y-intercept
(HR/V
̇O
2
, p = 0.002, Figure 7(a)) or greater slope (V
̇
E
/V
̇O
2
,
p = 0.030, Figure 7(b)). Conversely, the V
̇CO
2
/V
̇O
2
relationships
were similar in HDR and HUR (Figure 7(c)).
Of note, the respiratory frequency and HR achieved at the
15
th
min during HDR were similar to the maximal values mea-
sured during the level running incremental test (58
breaths
−1
·min
−1
and 183 bpm, respectively, p = 0.579 and
p = 0.240, respectively).
No dierence was observed in hip extensor, knee extensor
and plantar exor isometric torque before HUR and HDR (Figure
5). However, HDR induced substantial torque losses in hip
extensor (−11%) and knee extensor (−15%) whereas plantar
exor torque loss did not reach signicance (p = 0.353).
Similar percentage change and signicance level were
obtained regarding torque normalized to body mass
(2.30 ± 0.64 to 2.03 ± 0.57 Nm·kg
−1
for hip extension,
3.07 ± 1.04 to 2.66 ± 0.98 Nm·kg
−1
for knee extension and
1.67 ± 0.59 to 1.52 ± 0.65 Nm·kg
−1
for plantar exion).
Greater stride frequency (+24%) and longer stride length
(+141%) together with much shorter contact time (−60%) were
Figure 3. Time course of the oxygen consumption (panel a) and heart rate (panel b) responses during the 15 min trials. Black symbols: DR at 70% V
̇O
2max
(HDR); white
symbols: UR at 70% V
̇O
2max
(HUR); grey symbols: DR at the same running speed than UR (LDR). Horizontal dashed lines indicate maximal values measured during the
level running incremental test. Values are means ± SD of 8 athletes.
JOURNAL OF SPORTS SCIENCES 5

observed in HDR vs HUR at both the 3
rd
and 15
th
min time points
(Figure 6; all p < 0.001). None of these parameters displayed
signicant changes between the 3
rd
and 15
th
min of exercise.
Discussion
Paragraph 17. This study is the rst comparison of sustained
(15 min) UR vs DR performed at steep slope (± 15%) either at
similar running speed or at similar oxygen uptake in well-
trained athletes. The main ndings of this study are: i) cardior-
espiratory responses are dramatically reduced in LDR vs HUR,
but maximal HR and respiratory frequency can be achieved in
HDR despite very submaximal V
̇O
2
(70% V
̇O
2max
); ii) greater
slow components of V
̇O
2
, HR, V
̇
E
and respiratory frequency
occurred in HDR vs HUR or LDR; iii) muscular fatigue developed
in lower limbs extensor muscles specically after HDR; iv) spa-
tiotemporal parameters of the running gait are mechanical and
metabolic intensity-dependent but remained constant
throughout each trial.
Specicity of cardiorespiratory responses to downhill vs
uphill running
Paragraph 18. When comparing HUR vs LDR at steep slopes (±
15%) (similar running speed, 6.2 km·h
−1
), 25% to 70% lower
cardiorespiratory responses, together with reduced b[La] were
observed in LDR. This observation is in line with the reduced
metabolic cost of DR of trained runners at this slope (Lemire
et al., 2018) and extend previous ndings collected during
shorter duration running trials with smaller slope (5 min at ±
5%) (Kolkhorst et al., 1996). However, when comparing HUR
and HDR at steep slope (± 15%) (similar oxygen uptake), much
higher running speed was achieved in HDR (18.9 vs 6.2 km·h
−1
)
together with greater HR and V
̇
E
(Figures 3(b) and 4(a)). The
Figure 4. Time course of the ventilatory responses during the 15 min trials. Panel a, pulmonary ventilation; panel b, respiratory frequency; panel c, tidal volume. Black
symbols: DR at 70% V
̇O
2max
(HDR); white symbols: UR at 70% V
̇O
2max
(HUR); grey symbols: DR at the same running speed than UR (LDR). Horizontal dashed lines indicate
maximal values measured during the level running incremental test. Values are means ± SD of 8 athletes.
6M. LEMIRE ET AL.

greater V
̇
E
response observed in HDR was associated with
a higher respiratory frequency, suggesting a more supercial
ventilation pattern (Figure 4(a,b)) but Ti/Ttot ratio was not
altered, indicating that the relative contribution of inspiration
and expiration phases into the ventilation cycle was unchanged
in HDR vs HUR. The exacerbated HR and ventilatory responses
progressively developed throughout the 15 min HDR trial led to
maximal HR and maximal respiratory frequency despite sub-
maximal V
̇O
2
(70% V
̇O
2max
, Figure 3). Our athletes approached
the upper limit of their running capacity at the end of the HDR
trial and pilot works demonstrated that most of them would
not have been able to complete a 15 min DR trial performed at
80% V
̇O
2max
, despite 4 familiarization sessions in DR. Given the
lower V
̇O
2max
recently reported in highly trained athletes
during DR at −15% slope (Lemire et al., 2020b), the HDR trial
in the present study was performed at higher relative metabolic
intensity (84% of condition-specic V
̇O
2max
), possibly contribut-
ing to the higher HR and V
̇
E
responses observed in HDR vs HUR,
despite similar absolute metabolic intensity (V
̇O
2
in l·min
−1
).
Paragraph 19. Collectively, the present results suggest that
the HR/V
̇O
2
response to DR is dierent than the one observed
for UR, supporting a dierent coupling of the cardiovascular
and metabolic responses during DR vs UR (Figure 7(a)). This
possibility is in line with previous reports comparing UR vs DR
(Kolkhorst et al., 1996) or eccentric vs concentric cycling
(Dufour et al., 2007, 2004; Lipski et al., 2018). Of note, blood
lactate was signicantly higher at the 1
st
and 3
rd
min of recov-
ery after HDR vs HUR despite similar exercise V
̇O
2
. This
Figure 5. Maximal isometric strength measured pre and post trials. Hip extensor (Panel a), knee extensor (Panel b) and plantar flexor (Panel c). DR at 70% V
̇O
2max
(HDR);
UR at 70% V
̇O
2max
(HUR); DR at the same running speed than UR (LDR). * p < 0.05 pre vs post in HDR. Values are means ± SD of 8 athletes.
JOURNAL OF SPORTS SCIENCES 7

observation suggests that HDR might require a greater contri-
bution of anaerobic glycolysis to total energy turnover, albeit
remaining somewhat modest (b[La] = ~3.3 mmol·l
−1
) as it did
not lead to greater CO
2
production (i.e., via increased buer
activity) in HDR vs HUR conditions (Figure 7(c)). Although the
physiological tenets of the specic HR and ventilation pattern
observed in DR vs UR remain unclear, they might be related to
a combination of several factors pertaining to DR such as
increased eccentric muscle actions, higher running velocities,
exacerbated muscle work for trunk stabilization and/or altered
locomotor/ventilation coupling with cardiovascular and venti-
latory consequences.
Rapid and slow components of the cardiorespiratory
responses during downhill vs uphill running
Paragraph 20. After rapid initial responses from exercise onset,
V
̇O
2
, V
̇
E
, respiratory frequency and tidal volume remained
stable between 3
rd
and 15
th
min in HUR and LDR (Figures 3
and 4). Conversely, despite its 3 times faster-running speed,
V
̇O
2
was lower (−300 ml·min
−1
) after 3 min of exercise during
HDR vs HUR, and progressively increased between 3
rd
and 15
th
min during HDR, describing a slow component (+371 ml·min
−1
),
which also occurred for HR (+22 bpm), V
̇
E
(+24 l·min
−1
) and
respiratory frequency (+9 breaths·min
−1
). These observations
suggest our athletes incurred a larger O
2
decit in the rst
3 min of the HDR vs HUR trial, despite similar end-exercise
V
̇O
2
and similar V
̇O
2
averaged over the 3
rd
to 15
th
min time
window. These results extend previous ndings showing simi-
lar V
̇O
2
kinetics in eccentric vs concentric cycling performed at
much lower metabolic intensity (~26% V
̇O
2max
) (Perrey et al.,
2001) and also add on recent observations that V
̇O
2
kinetics
during LR and UR are correlated in elite mountain runners
(Willis et al., 2019). The physiological mechanisms underlying
the progressively increasing cardiorespiratory responses in HDR
could be related to the higher relative metabolic exercise
intensity in HDR (84% of DR V
̇O
2max
) vs HUR (68% UR V
̇O
2max
)
(Lemire et al., 2020b) and/or to the greater mechanical con-
straints associated with DR presumably impairing O
2
delivery
(i.e., greater vascular collapse during muscle actions) and/or
increasing O
2
demand as lower leg muscles fatigue (i.e., altered
muscle recruitment) (Poole, 2019).
Lower limbs muscular fatigue in downhill vs uphill
running
Paragraph 21. Signicant torque losses were observed in the
knee and hip extensors (range −11 to −15%) only after HDR
(Figure 5). Of note, none of these strength measures signi-
cantly correlated to the cardiorespiratory responses, including
the magnitude of the slow components. The torque losses
observed after 15-min DR at 70% V
̇O
2max
in the present
study are in line with previous reports showing depressed
knee extensor isometric torque (−15%) after 30-min DR
(10 km·h
−1
at −20% slope) (Martin et al., 2004) or reduction
in knee extensors isometric torque (−19%) after an 8.5 km
mostly downhill trail run (~34 min duration, about −17%
slope with running speed ranging from 11 to 19 km·h
−1
)
(Giandolini et al., 2016a).
This attenuated torque after HDR could be mainly ascribed
to the deleterious eect of lower limb mechanical loading, as
oxygen uptake (i.e. metabolic load) was globally similar to HUR.
Muscle damage have been repeatedly reported to contribute
to muscular dysfunction after DR (Giandolini et al., 2016b),
therefore their involvement in the present results cannot be
ruled out but their role was probably minimal as our subjects
were used to perform trail races/training and all realized 4
specic familiarization sessions to treadmill DR. Whatever the
Figure 6. Stride length (Panel a), stride frequency (Panel b) and contact time (Panel c) measured at the 3
rd
and 15
th
min in each trial. DR at 70% V
̇O
2max
(HDR); UR at
70% V
̇O
2max
(HUR); DR at the same running speed than UR (LDR). * p < 0.05 vs HDR and $ p < 0.05 vs HUR at same time points. Values are means ± SD of 8 athletes.
8M. LEMIRE ET AL.

Figure 7. Relationships between heart rate (Panel a), ventilation (panel b), carbon dioxide production (Panel c) and oxygen uptake respectively. Black symbols: DR at
70% V
̇O
2max
(HDR); white symbols: UR at 70% V
̇O
2max
(HUR); grey symbols: DR at the same running speed than UR (LDR). * p < 0.05 vs linear regression slope in LDR, ¤
p < 0.05 vs linear regression slope in HDR and # p < 0.05 vs linear regression intercept in HUR. Values are means ± SD of 8 athletes.
JOURNAL OF SPORTS SCIENCES 9

exact underlying mechanisms, the reduction in extensor mus-
cle torque observed after HDR supports the recent proposition
that lower limb muscle strength contributes to incline running
performance (Lemire et al., 2020a) as encountered during trail
or road races.
Running kinematics in downhill vs uphill running
Paragraph 22. Despite similar running speed (6.2 km·h
−1
), HUR
elicited higher stride frequency, shorter stride length and
shorter contact time compared to LDR (Figure 6) in line with
previous observations collected from ± 5 to 15% slope
(Gottschall & Kram, 2005; Minetti et al., 1994). When compared
at similar V
̇O
2
(~2.900 l·min
−1
), HDR (18.9 km·h
−1
) induced 2.5
times longer stride length, higher stride frequency and more
than 2 times shorter contact time compared to HUR
(6.2 km·h
−1
). To our knowledge, this is the rst kinematic
description of DR vs UR gait at severe slope (±15%) combined
with similar but high metabolic intensity (i.e., especially in DR)
in well-trained athletes.
Within each trial, we observed stable running gaits between
the 3
rd
and 15
th
min, even during HDR where signicant max-
imal isometric muscle torque losses were identied. Therefore,
running gait during DR at 70% V
̇O
2max
with −15% slope can be
preserved for 15-min by well-trained athletes despite signi-
cant lower limb extensor muscle torque losses and signicant
cardiorespiratory slow components.
Limitations
Paragraph 23. The running speed used during HUR and HDR
to achieve 70% of level running V
̇O
2max
have been deter-
mined using previously performed maximal incremental
tests using 2-min stages at the same slopes (+15/-15%). This
method does not allow to anticipate the development of slow
components in the cardiorespiratory responses but the
observation of similar end-exercise and average V
̇O
2
in HUR
vs HDR suggests our objective of similar V
̇O
2
among condi-
tions was reached. The assessments of lower limb muscle
strength were not performed with a gold-standard isokinetic
ergometer but instead with a handheld dynamometer.
Although such device may have accuracy limitations, it has
been validated in the literature against isokinetic ergometry
and we implemented specic positions of both the investi-
gator and the subject during each evaluation to secure accu-
rate measurements. In the same line, although limited, our
number of subjects (N = 8) was determined after sample size
calculation to ensure a statistical power > 0.9 and an alpha
risk < 0.05. Lastly, our calculations of cardiorespiratory slow
components over the 15-min running trials were not based
on parameters of V
̇O
2
on-kinetics modelizations but rather on
the dierence between end-exercise and the 3
rd
minute value
(Poole & Jones, 2012). We are condent this approach is
appropriate to highlight the absence of steady-state
observed in the HDR condition and opens the way for future
characterization of proper V
̇O
2
on-kinetics in inclined vs level
running conditions.
Paragraph 24. In conclusion, cardiorespiratory
responses to DR vs UR heavily depend on both metabolic
(i.e., oxygen uptake) and mechanical exercise intensity (i.e.,
running speed). At similar running speed, cardiorespiratory
responses are attenuated in DR vs UR, with no signicant
impairment in muscular function but altered running gait
kinematic. During high intensity running exercise and
despite similar oxygen uptake, cardiorespiratory responses
are exacerbated in DR vs UR, with signicant reductions in
lower limb extensor muscle torque and markedly dierent
running kinematics. Of note, DR running at 70% V
̇O
2max
sustained for 15 min at −15% slope led HR and respiratory
frequency to maximal values in well-trained athletes. These
results demonstrate that high intensity DR sessions at
−15% slope can hardly be performed above 70% of
V
̇O
2max
in competitive trailers.
Acknowledgements
The authors thank the athletes for their willingness to participate in the
present work. They also warmly thank the whole Respiratory and
Functional Exploration department of Strasbourg’s New Civil Hospital.
Compliance with ethical standards
The experiment was previously approved by our Institutional Review Board
and complied with the Declaration of Helsinki (CPP18-039a/2108-A00700-55).
Disclosure statement
No conicts of interest, nancial or otherwise, to declare by the authors.
Funding
No funding was received for this study.
ORCID
Marcel Lemire http://orcid.org/0000-0002-8023-0329
References
Baiget, E., Peña, J., Borràs, X., Caparrós, T., López, J. L., Marin, F., Coma,
J., Eduard Comerma, E. (2016). Eects of a trail mountain race on neu-
romuscular performance and hydration status in trained runners. The
Journal of Sports Medicine and Physical Fitness, 58(1-2), 43–49. https://doi.
org/10.23736/S0022-4707.16.06792-X
Billat, V., & Koralsztein, J. P. (1996). Signicance of the velocity at VO2max
and time to exhaustion at this velocity. Sports Medicine, 22(2), 90–108.
https://doi.org/10.2165/00007256-199622020-00004
Born, D. P., Stoggl, T., Swaren, M., & Bjorklund, G. (2016). Near-Infrared
Spectroscopy: More Accurate Than Heart Rate for Monitoring Intensity
in Running in Hilly Terrain. International Journal of Sports Physiology and
Performance, 12(4), 440–447. https://doi.org/10.1123/ijspp.2016-0101
Breiner, T. J., Ortiz, A. L. R., & Kram, R. (2018). Level, uphill and downhill
running economy values are strongly inter-correlated. European Journal
of Applied Physiology, 119(1), 257–264. https://doi.org/10.1007/s00421-
018-4021-x
Dick, R. W., & Cavanagh, P. R. (1987). An explanation of the upward drift in
oxygen uptake during prolonged sub-maximal downhill running.
Medicine and Science in Sports and Exercise, 19(3), 310–317. https://doi.
org/10.1249/00005768-198706000-00019
Dufour, S. P., Doutreleau, S., Lonsdorfer-Wolf, E., Lampert, E., Hirth, C.,
Piquard, F., Lonsdorfer, J., Geny, B., Mettauer, B., & Richard, R. (2007).
Deciphering the metabolic and mechanical contributions to the
10 M. LEMIRE ET AL.

exercise-induced circulatory response: Insights from eccentric cycling.
American Journal of Physiology-Regulatory, Integrative and Comparative
Physiology, 292(4), R1641–1648. https://doi.org/10.1152/ajpregu.00567.
2006
Dufour, S. P., Lampert, E., Doutreleau, S., Lonsdorfer-Wolf, E., Billat, V. L.,
Piquard, F., & Richard, R. (2004). Eccentric cycle exercise: Training appli-
cation of specic circulatory adjustments. Medicine & Science in Sports &
Exercise, 36(11), 1900–1906. https://doi.org/10.1249/01.MSS.
0000145441.80209.66
Easthope, C. S., Hausswirth, C., Louis, J., Lepers, R., Vercruyssen, F., &
Brisswalter, J. (2010). Eects of a trail running competition on muscular
performance and eciency in well-trained young and master athletes.
European Journal of Applied Physiology, 110(6), 1107–1116. https://doi.
org/10.1007/s00421-010-1597-1
Garnier, Y., Lepers, R., Assadi, H., & Paizis, C. (2019). Cardiorespiratory
Changes During Prolonged Downhill Versus Uphill Treadmill Exercise.
International Journal of Sports Medicine, 41(2), 69–74. https://doi.org/10.
1055/a-1015-0333
Giandolini, M., Horvais, N., Rossi, J., Millet, G. Y., Morin, J. B., & Samozino, P.
(2016a). Acute and delayed peripheral and central neuromuscular altera-
tions induced by a short and intense downhill trail run. Scandinavian
Journal of Medicine & Science in Sports, 26(11), 1321–1333. https://doi.
org/10.1111/sms.12583
Giandolini, M., Vernillo, G., Samozino, P., Horvais, N., Edwards, W. B.,
Morin, J. B., & Millet, G. Y. (2016b). Fatigue associated with prolonged
graded running. European Journal of Applied Physiology, 116(10),
1859–1873. https://doi.org/10.1007/s00421-016-3437-4
Gottschall, J. S., & Kram, R. (2005). Ground reaction forces during downhill
and uphill running. Journal of Biomechanics, 38(3), 445–452. https://doi.
org/10.1016/j.jbiomech.2004.04.023
Healy, A., Linyard-Tough, K., & Chockalingam, N. (2019). Agreement
Between the Spatiotemporal Gait Parameters of Healthy Adults From
the OptoGait System and a Traditional Three-Dimensional Motion
Capture System. Journal of Biomechanical Engineering, 141(1), 014501.
https://doi.org/10.1115/1.4041619
Kolkhorst, F. W., Mittelstadt, S. W., & Dolgener, F. A. (1996). Perceived
exertion and blood lactate concentration during graded treadmill
running. European Journal of Applied Physiology and Occupational
Physiology, 72(3), 272–277. https://doi.org/10.1007/BF00838651
Lazzer, S., Salvadego, D., Taboga, P., Rejc, E., Giovanelli, N., & Di Prampero, P. E.
(2015). Eects of the Etna uphill ultramarathon on energy cost and
mechanics of running. International Journal of Sports Physiology and
Performance, 10(2), 238–247. https://doi.org/10.1123/ijspp.2014-0057
Lemire, M., Hureau, T. J., Favret, F., Geny, B., Kouassi, B. Y. L., Boukhari, M.,
Lonsdorfer, E., Remetter, R., & Dufour, S. P. (2020a). Physiological factors
determining downhill vs uphill running endurance performance. Journal
of Science and Medicine in Sport, S1440-2440(20), 30664-2. https://doi.org/
10.1016/j.jsams.2020.06.004
Lemire, M., HUREAU, T. J., REMETTER, R., GENY, B., KOUASSI, B. Y. L.,
LONSDORFER, E., ISNER-HOROBETI, M.-E., FAVRET, F., & DUFOUR, S. P.
(2020b). Trail Runners Cannot Reach VO2max during a Maximal
Incremental Downhill Test. Medicine and Science in Sports and
Exercise, 52(5), 1135–1143. https://doi.org/10.1249/MSS.
0000000000002240
Lemire, M., Lonsdorfer-Wolf, E., Isner-Horobeti, M. E., Kouassi, B. Y. L.,
Geny, B., Favret, F., & Dufour, S. P. (2018). Cardiorespiratory Responses
to Downhill Versus Uphill Running in Endurance Athletes. Research
Quarterly for Exercise and Sport, 89(4), 511–517. https://doi.org/10.1080/
02701367.2018.1510172
Lindstedt, S. L., LaStayo, P. C., & Reich, T. E. (2001). When active muscles
lengthen: Properties and consequences of eccentric contractions. News
in Physiological Sciences: An International Journal of Physiology Produced
Jointly by the International Union of Physiological Sciences and the
American Physiological Society, 16(1), 256–261. https://doi.org/10.1152/
physiologyonline.2001.16.6.256
Lipski, M., Abbiss, C. R., & Nosaka, K. (2018). Cardio-pulmonary responses to
incremental eccentric and concentric cycling tests to task failure.
European Journal of Applied Physiology, 118(5), 947–957. https://doi.
org/10.1007/s00421-018-3826-y
Martin, V., Millet, G. Y., Martin, A., Deley, G., & Lattier, G. (2004). Assessment
of low-frequency fatigue with two methods of electrical stimulation.
Journal of Applied Physiology, 97(5), 1923–1929. https://doi.org/10.
1152/japplphysiol.00376.2004
Mentiplay, B. F., Perraton, L. G., Bower, K. J., Adair, B., Pua, Y.-H.,
Williams, G. P., McGaw, R., & Clark, R. A. (2015). Assessment of Lower
Limb Muscle Strength and Power Using Hand-Held and Fixed
Dynamometry: A Reliability and Validity Study. PLoS One, 10(10),
e0140822. https://doi.org/10.1371/journal.pone.0140822
Minetti, A. E., Ardigo, L. P., & Saibene, F. (1994). Mechanical determinants of
the minimum energy cost of gradient running in humans. The Journal of
Experimental Biology, 195(1), 211–225.
Minetti, A. E., Moia, C., Roi, G. S., Susta, D., & Ferretti, G. (2002). Energy cost of
walking and running at extreme uphill and downhill slopes. Journal of
Applied Physiology, 93(3), 1039–1046. https://doi.org/10.1152/japplphy
siol.01177.2001
Mu, G., Dufour, S., Meyer, A., Severac, F., Favret, F., Geny, B., Lecocq, J., &
Isner-Horobeti, M.-E. (2016). Comparative assessment of knee extensor
and exor muscle strength measured using a hand-held vs. isokinetic
dynamometer. Journal of Physical Therapy Scienc, 28(9), 2445–2451.
https://doi.org/10.1589/jpts.28.2445
Perrey, S., Betik, A., Candau, R., Rouillon, J. D., & Hughson, R. L. (2001).
Comparison of oxygen uptake kinetics during concentric and eccentric
cycle exercise. Journal of Applied Physiology, 91(5), 2135–2142. https://
doi.org/10.1152/jappl.2001.91.5.2135
Pokora, I., Kempa, K., Chrapusta, S. J., & Langfort, J. (2014). Eects of
downhill and uphill exercises of equivalent submaximal intensities
on selected blood cytokine levels and blood creatine kinase activity.
Biology of Sport, 31(3), 173–178. https://doi.org/10.5604/20831862.
1111434
Poole, D. C. (2019). Edward F. Adolph Distinguished Lecture. Contemporary
Model of Muscle Microcirculation: Gateway to Function and Dysfunction.
Journal of Applied Physiology, 127(4), 1012–1033. (1985). https://doi.org/
10.1152/japplphysiol.00013.2019
Poole, D. C., & Jones, A. M. (2012). Oxygen uptake kinetics. Comprehensive
Physiology, 2(2), 933–996. https://doi.org/10.1002/cphy.c100072
Pringle, J. S., Carter, H., Doust, J. H., & Jones, A. M. (2002). Oxygen uptake
kinetics during horizontal and uphill treadmill running in humans.
European Journal of Applied Physiology, 88(1–2), 163–169. https://doi.
org/10.1007/s00421-002-0687-0
Vernillo, G., Giandolini, M., Edwards, W. B., Morin, J. B., Samozino, P.,
Horvais, N., & Millet, G. Y. (2016). Biomechanics and Physiology of
Uphill and Downhill Running. Sports Medicine, 47(4), 615–629. https://
doi.org/10.1007/s40279-016-0605-y
Westerlind, K. C., Byrnes, W. C., Harris, C., & Wilcox, A. R. (1994). Alterations in
oxygen consumption during and between bouts of level and downhill
running. Medicine & Science in Sports & Exercise, 26(9), 1144–1152.
https://doi.org/10.1249/00005768-199205001-00343
Willis, S. J., Gellaerts, J., Mariani, B., Basset, P., Borrani, F., & Millet, G. P.
(2019). Level Versus Uphill Economy and Mechanical Responses in
Elite Ultra-Trail Runners. International Journal of Sports Physiology
and Performance, 14(7), 1001–1005. https://doi.org/10.1123/ijspp.
2018-0365
JOURNAL OF SPORTS SCIENCES 11